March 2000

Ref : 2000-220-0004

......................................................................................................

Metering, Load Profiles and Settlement

in Deregulated Markets
......................................................................................................

System Tariff Issues Working Group

......................................................................................................
 TABLE OF CONTENTS

1. Introduction.................................................................................................................................................. 2
2. General overview of existing metering, load profiles and settlement systems.................................................. 4
2.1 Overview of threshold level established for metering................................................................................... 4
2.2 Overview of load profile models - the area and the consumer category model .............................................. 6
2.3 Overview of the settlement methods used in power markets....................................................................... 10
2.4 Time restraints and other conditions for changing supplier......................................................................... 14
2.5 Costs resulting from changing supplier (who pays) .................................................................................... 16
2.6 Costs of metering – included in the tariff?.................................................................................................. 17
3. Conclusions and recommendations.............................................................................................................. 23
3.1 Threshold levels for metering based on time interval ................................................................................. 23
3.2 Which load profile model is adequate where? ............................................................................................ 24
3.3 Settlement of the deviation between estimated and actual consumption...................................................... 27
Appendix 1: Description of the already existing metering load profiles and settlement systems ........................ 28
Finland........................................................................................................................................................... 28
Norway .......................................................................................................................................................... 35
United Kingdom............................................................................................................................................. 43
Sweden........................................................................................................................................................... 50
Proposed German System ............................................................................................................................... 53
New Zealand .................................................................................................................................................. 55
Appendix 2: Future plans for metering load profiles and settlement systems in other countries......................... 61
Ireland............................................................................................................................................................ 61
Slovenia ......................................................................................................................................................... 61
Spain.............................................................................................................................................................. 62
The Netherlands ............................................................................................................................................. 64
Turkey............................................................................................................................................................ 65
1. Introduction
With the opening of the European electricity markets, the first step in most countries has been
to look at the wholesale market, and the relatively large consumers. For those consumers the
metering of the consumption per hour is relatively straightforward, and it is easy to establish
at which price the electricity consumption should be billed, since large consumers have
interval meters installed already, for hourly or half-hourly metering.
But as thresholds become lower and some countries go towards a full opening of their
markets, this is no longer the case. For the retail market, the question is therefore how to
establish a practical but economical alternative to hourly metering for each individual end-
consumer.
Load profile-based metering and settlement is a feasible and cost effective way to give the
opportunity for small consumers to participate in open markets. The users are mainly
residential consumers, but also low energy users, that are now able to change their electricity
supplier in several of the deregulated markets in Europe. The aim is to allow real access to the
retail market without introducing costly and complex metering systems as a necessary
requirement. This can be seen either as a permanent or a transitional solution.
Load profiling involves:
• determining an estimate of the average load profile of a class of customers over a given
period, and
• allocating or ‘deeming’that load profile to all customers in that customer category.
That class of customers can be either all those consumers that are not metered on time interval
basis within the geographic region covered by a network. With this method, the non-metered
customers constitute the residual profile, which is an adjusted load-profile for the node or for
the area under consideration. The other option is grouping customers with a similar load-
pattern into categories, where each individual customer is then associated with a
predetermined representative load-profile. There are different criteria used to create these
profiles, but the precondition is always that load measurements (metering measurements)
have been made at an earlier stage.
Profiling enables an electricity supplier to calculate the electricity consumption for every
pricing period on the market (usually half-hourly or hourly time intervals, in some cases
quarter-hourly) of its customers who do not have a time interval meter installed. This will
form the basis for the supplier to pay for the electricity purchased from the wholesale market.
Fitting total consumption, which can be measured using the existing time interval meters, to a
load profile would allow a customer’s consumption to be dis-aggregated into time interval
segments (quarter-hourly, half-hourly or hourly). Therefore, this approach would avoid the
expense of replacing existing meters with time interval meters while still creating the data
necessary to perform wholesale settlement for each time interval.
Ideally, all consumers in an open electricity market should have time-interval meters with on-
line communication. This would ensure accurate billing of consumers and eliminate cross-
subsidies between customers with differing load profiles. Further, it would eliminate the risks
to suppliers of estimating load profiles and assigning network loss residuals, it would enable
accurate and more final reconciliation of wholesale settlements between suppliers, and
eliminate the purchase of the meter as a barrier to competition. On-line communication would
allow the data from the metering to be downloaded frequently, to allow weekly or monthly
settlement.

2
The problem with meters and communications from meters is that this solution is relatively
costly for small consumers. This includes both the capital costs of meters and installing
meters and the ongoing costs of operating and maintaining meters. However, the cost of
meters will probably continue to fall as a mass market is established. The cost of
communication, i.e. the collection of the meter-reading, will depend on the frequency of
meter-reading as well as the method, i.e. whether it is done manually or made possible
through telecommunication lines. By installing meters and communication facilities only for
customers that do switch supplier, the costs can be kept relatively low. At a later stage, global
meter reading can be considered. It must also be considered whether the costs of metering
should be allocated only to the users who do change, or spread on all customers. Spreading
the cost on all customers could avoid the cost of metering making customers abstain from
changing supplier.
Although the metering solution is attractive as an ideal solution, the costs of installing meters
and communication with the meters makes a load-profiling solution an attractive alternative,
at least in a transitional phase. The report therefore mainly focuses on the load profiling
solutions.

This report intends to cover the following main topics:

• Description of the already existing metering, load profiles and settlement systems in some
of the countries that have new legislation adopted for a load-profiling solution.
• Future plans for metering, load profiles and settlement systems in the other countries that
have not yet adopted new legislation.
• Review of threshold levels established for interval metering, and analysis of load profile
models and settlement methods used in practice.
• Conditions for changing supplier and costs resulting from changing supplier.
• Settlement of the deviation between estimated and actual consumption.

*As supply companies are unbundled from distribution companies, it becomes important to distinguish between
the unbundled distribution network companies and the traditional distribution companies that are still integrated
supply and distribution companies.

3
2. General overview of existing metering, load profiles and settlement systems

2.1 Overview of threshold level established for metering

The advantage of knowing the accurate hourly consumption is obvious for a supply company
in terms of reduced reading costs, more accurate billing, better energy data to assist in the
management of energy trading, opportunities for further service depending on the type of
meter installed etc. The benefit to the consumer is generally seen as the possibility of offering
a better price for energy consumption and the prospect of additional services.
It can be seen in table 2.1 that in the countries that already have new legislation on metering
and load profiles there is a large diversity of solutions for metering systems used and for the
threshold level imposed to consumers for having those metering systems.
The metering time interval varies from a quarter of an hour to an hour. Countries where
interval metering requirements for direct access to the electricity market were implemented
most recently chose shorter intervals than the countries the system was implemented earlier.
The option was chosen probably in connection with the costs associated with the
implementation of the metering system. The data communications infrastructure is the most
complex and potentially highest cost technology issue, not always the meter itself. Further,
the long-term costs associated with the use of the telecommunication infrastructure (meter
reading costs and charges for data transmission) are added and those depend essentially on the
volume of data foreseen to exchange between the players on the market. The more data
needed to transfer and the more frequent the data is needed the higher the costs of data
transmission. This is the reason that greater attention needs to be paid to the collection of
metering data, in the time-scale required for settlement, at the minimum cost. Even while the
price of meters does decrease continuously, resulting from the technical characteristics of the
metering systems available at the moment of implementation, the cost of meters is not
negligible.
As regards the threshold level for compulsory use of interval metering systems, a broad range
of solutions have been adopted in different countries. In general customers above a certain
consumption or technical characteristics of their installations and/or who choose new
suppliers are required to install interval meters. This cut-off point (which at the same time
represents the threshold for load profiling eligibility) level varies from 20 to 200 kW in
different systems.
It is possible that a metering point with an expected power consumption below the imposed
threshold will be settled based on time interval metered values as a result of a consumer or
network owner demand. In such cases, the consumers themselves or the network owner
himself must defray the additional costs associated with the metering.

4
Table 2.1: Threshold level established for interval metering

Country Metering time Metering systems used Threshold level for metering on Communication systems used Threshold level for on-line
interval a time interval basis communication/storage
metering systems
• Hourly metering for large
Finland 1 hour 3 × 63 A main fuses for all kind Usually modem + telephone No regulatory restrictions
consumers who buy with bargain
of consumers line for storage metering
price (from new or old supplier)
• Load profile model for small
consumers who buy from new
supplier
• Hourly metering for large
Norway 1 hour Annual power consumption of
consumers (global requirement)
over 400 MWh per metering
• Load profile model for small point
consumers
• Half-hourly metering for large
United ½ hour Power demand of 100 kW and
consumers
Kingdom above.
• Load profile model for small
consumers
• Hourly metering for large
Sweden 1 hour 3 x 65 A,
consumers
• Load profile model for small 135 KW
consumers
• Quarter-hourly metering for
Germany ¼ hour Not established yet Large customer : data-lines > 5 MW
large and medium consumers
• Load profile model for small Small customer: meter reading
consumers
• Half-hourly metering for those
New Zealand ½ hour Annual power consumption of
consumers who have meters that
over 500 MWh.
met certain standards
• Load profiling methods for the
other consumers

5
2.2 Overview of load profile models - the area and the consumer category model
There are two general methods for the construction of load curves: the area method and the
category method. All analysed countries are using either one of them or a combined model
and both models present advantages as well as disadvantages. In each of these countries exists
more or less implementation particularities but the general principle is the same. A periodic
reconciliation (calculation for deviations between the suppliers) is connected with both those
methods.
In the area model (or regional method), especially large or medium size consumers are
equipped with the facilities for time interval metering. Power used by as well as losses related
to those consumers is subtracted from the total energy consumption by the considered time
interval for metering. The result, that is the difference calculated for each time interval, is
considered to illustrate the consumption of all those consumers who are entitled to use the
load curve model in the area (network) in question. Therefore this average profile can be
deemed to be the load profile for all end users that were not metered on time interval basis
within the geographic region covered by the network.

Hourly metering: one supplier Hourly metering: new supplier

Hourly metering: several suppliers

Area profile metering without customers having chosen a new supplier, with one customer having changed
supplier, and finally several customers having chosen different suppliers. (Norwegian Ministry of Energy).

6
The category model (or consumer-group-related method) defines the load of a consumer
taking into account the average load of a group. There are different criteria used to form these
profiles but the precondition is always that sufficient load measurements have been made
earlier. From the point of view of type of data used to construct the load curves there are two
main profiling methods:

A typical winter day

(Germany) with load
profiles for different
kinds of customers.
The large area in the
bottom is households,
the top area is
measured customers.

Example (from
Germany) of high
share of individually
metered customers
within an area. (top
area). Remaining
areas shown include
rest-curve, heating,
special contracts,
companies and
households.

• Static profiles
are derived from existing, historic data. If consumption data for each time interval
considered has been collected from a sufficiently large sample of customers, preferably
for more than a year, then a profile can be developed from this data for that class of
customer. The profile can also be modified to take into account factors that affect
consumption and which may vary from day to day as well as from year to year (variations
in the weather, holiday periods compared to the long term average etc.).
• Dynamic profiles are derived from current data collected from a ‘live’ sample. The
consumption data for each time interval considered of a sample is continually metered and
the resulting data is analysed to derive average daily load profiles. There is no need to

7
 correct the profile to take into account of holidays and weather phenomena, as the ‘live’
sample will respond directly to those variations.

Adjusted system load profile

30000

25000
Net losses
20000 Hourly metered
ASLP
GWh

15000

10000

5000

0
One year
Example of the components of electricity use, including Adjusted System Load Profile (ASLP), hourly metered
customers and net losses (Norwegian Ministry of Energy).

• A third profiling method can be considered in cases where the consumption pattern and
load is predictable and can be determined using a simple engineering calculation (e.g. for
street-lightning).
Table 2.2 (below) represents a synopsis of load profile models used for determining the
consumption profile for those consumers that are not metered on a time interval basis in some
of the countries considered in our analysis together with their implementation particularities.

8
Table 2.2: Load profile models used
Country Load profile Implementation particularities
model used
Category model Four national load curves for consumer groups:
Finland
(consumer-group- 1. Households without electric heating
related method) 2. Households with electric heating (temperature correction
included: +4%/°C below +15°C)
3. Other consumers below 3x35 A
4. Other consumers 3x35 A or bigger (must be equipped
with 2-time-zone meter)
Area model ASLP is the average consumption profile for those end users
Norway
that are not metered hourly in a power network. ASLP is
derived by deducting the network loss together with the
actual consumption of consumers with hourly settlement
from the hourly net input into the network.
Category model Eight half-hourly profiles areestimated by a contractor to
United
(consumer-group- the Electricity Pool, using data acquired from national load
Kingdom
related method) research customer samples; two categories for domestic
customers, and six for non-domestic customers. The load
profiles are updated annually. Corrections with temperature
and the time of sunset are made using linear regression
techniques.
Area model As Norway from 1 November 1999
Sweden
(regional method)
Category model Preferred but not implemented yet.
Germany
(consumer-group-
related method or
Synthetic Model)
Combined area Two general types of profiles:
New
(nodal) &
Zealand 1. Network Supply Point (NSP)-Derived profiles – these
category model are profiles derived through the manipulation of the
residual NSP profile (similar to ASLP):
• Interval time of use meters.
• Separately metered controlled load.
• Non-separately metered controlled load.
• Uncontrolled load 24-hour meters using Grid Exit
Point (GXP) profile.
2. Load Research/Engineering Estimate profiles – these are
profiles either derived from the statistical sampling of
consumers, or from engineering estimates of
consumption:
• Non-MARIA-compliant half-hour metered loads
(essentially a statistically sampled profile with 100%
sampling).
• Non-metered loads (e.g. street lighting).

9
2.3 Overview of the settlement methods used in power markets
There are two broad models of wholesale settlement in a fully competitive market, the
differentiation approach and the global approach.
The differentiation approach involves first assessing the incumbent supplier’s consumption by
assessing the consumption of their associated distribution network. From this network
consumption is subtracted the consumption of all customers metered by interval time meters
and the consumption of all profile customers who are supplied by competitive suppliers,
leaving a difference which is the incumbent supplier’s consumption (residual curve).
In the global approach all customers are given a load profile regardless of whether or not they
remain with their incumbent supplier. The global settlement process then sums all the metered
and profile based consumption values for all customers of each supplier. If the sum differs
from the total metered consumption of the system then the difference is proportionally
allocated across all suppliers. This procedure can be used for both initial and final settlements.

ASLP and market share

30000
Supplier E 2%
25000
Supplier D 5%
20000 Supplier C 8%
Supplier B 15%
GWh

15000 Supplier A 70%

10000

5000

0
One year
Five different Suppliers and their market shares, and illustration of the Adjusted System Load Profile over a full
year (Norwegian Ministry of Energy).

Table 2.3 is a synopsis of the settlement methods used in the power markets of some of the
countries considered in our analysis.
Although the power markets are organised in different ways everywhere the players in the
settlement process are mainly the same. Apart from the Transmission System Operator, other
players are entities with different balance settlement responsibilities, usually suppliers,
distribution network operators (Finland), large power producers and large industrial
consumers that must provide balance settlement information to each other and to the
Transmission System Operator for the ultimate reconciliation. For the countries where spot
electricity and bilateral contract markets exist all this is valid for the trading on the whole of
these markets.

10
At the same time, each network owner carries out periodic financial settlement between the
individual suppliers to settle the difference between the calculated distribution (through load
profiling) and the measured distribution (on the basis of meter reading). These meter readings
are done yearly or more often. This is only a settlement between suppliers within an
individual network and does not have any impact with regard to the power system balance but
prior to final settlement the suppliers must endure the risk of their settled consumption
differing from final consumption.
The TSO has the overall responsibility for short term system operation and hence the
balancing responsibility on the system level. The overall balance-settlement and reconciliation
is normally handled by the TSO, but the role of other balance responsible parties may vary
from country to country.
There are specific methods for establishing the difference between estimated and actual
consumption reconciliation for consumers that are not metered on time interval basis. The
price at which this difference is settled is different from system to system. Some use the tariff
of the local supplier, others a weighted spot electricity price for that period etc.
As it can be seen profiling increases ‘upstream costs’ such as settlement systems costs and the
financial management complexity across suppliers. In the same time errors stemming from
profile inaccuracies contribute to settlement residuals (alongside of losses or differences
contributed to by theft, meter inaccuracies, errors in calculating transmission and distribution
losses) and make the allocation of settlement residuals during the settlement process more
difficult.

11
Table 2.3: Settlement methods used in power markets
Country Entities with balance settlement Balance settlement system Data transmission methods for Difference between estimated and
responsibilities balance settlement information actual consumption reconciliation
for consumers that are not metered
on time interval basis
Finland 1. Fingrid System Oy (part of the Flow of balance settlement information: - EDIEL (Electronic Data Yearly (or more often) at the tariff
Transmission Network Operator) - Consumers → suppliers Interchange in the of the local supplier
2. Balance Responsible Supplier - Suppliers → BRS Electricity industry) used
(BRS) – e.g. large power - Distribution companies → System as standard of
producers Operator communication from 1
3. Distribution network operators/ - BRS → System Operator September 1999
suppliers Cross checking of calculations among BRS
companies done by the System Operator
Norway 1. Statnett (Transmission System Flow of information in connection with: - EDIEL used as standard Periodically at weighted spot
Operator) – trading concession a) trading on the spot electricity market and of communication electricity price for the period
holder who is responsible for the bilateral trading: - EDIEL messages between two successive meter
settlement of regulating power - NordPool (Spot Electricity Market) MSCONS used for the readings
(Power exchange) → Power Exchange transmission of settlement
2. Entity with balance responsibility - EBR → Power Exchange data since 1 January 1998
(EBR) – trading concession holder b) Regulating power settlement (metering
for whom balancing power is data):
settled in the network owner’s - network owner → Transmission
power network: System Operator
- Suppliers - network owner → EBR
- end users (e.g. large industrial - Transmission System Operator →
consumers) EBR
1. Electricity Pool Settlement balance is carried out by the The source for consumption data is
United
2. Suppliers Electricity Pool. Initial settlement occurs recent tariff meter readings adjusted
Kingdom
based on meter data then available, followed to an annual basis, or provisional
by four additional reconciliation runs at estimates based on historic data in
intervals over a period concluding at 14 the absence of recent readings.
months after settlement day.

12
 Country Entities with balance settlement Balance settlement system Data transmission methods for Difference between estimated and
responsibilities balance settlement information actual consumption reconciliation
for consumers that are not metered
on time interval basis
1. Swedish Power Grid Company
Sweden As Norway (from 1 November 1999) As Norway (from 1 November As Norway (from 1 November
(Transmission Network Operator)
1999) 1999)
2. Distribution network operators
New 1. NZEM (wholesale pool) trading is
Steps of settlement process (ex post system): - Data for all Half Hourly Between periodically meter
Zealand reconciled by the Reconciliation
Meters are downloaded readings (at least once every 12
Manager 1. Collection of Metering Data
electronically. months on a rolling basis) is used
2. non-NZEM trading by the
2. Receive a List of ICPs from the Registry - Data for all Non-Half Historical Estimate Accrual
National Reconciliation Manager
Hourly Meters are Methodology.
(subsidiary of Transpower – 3. Application of Accruals collected in a manner as
Transmission Network Operator).
4. Application of Local Losses to Half- agreed between the Data
Hourly Metered Data Administrator and each
Supplier that it represents.
5. Aggregation Approach

13
2.4 Time restraints and other conditions for changing supplier
Table 2.4 represents a synopsis of time restraints for changing supplier in some of the
countries considered in our analysis.
Table 2.4: Time restraints for changing supplier

Country Consumer’s possibility for changing supplier

According to regulations In practice
Finland Any time - The network owner must be informed 2
weeks in advance.
- Usually fixed-term contracts (1-2 years).
Norway Any time (Monday every - Exchange of information procedure
week for profiles) through EDIEL messages (PRODATA)
for change of supplier must begin no
later than 3 weeks before the change is
to take effect.
United Any time - The change to the new supplier must
Kingdom become effective within 28 days of the
contract date.
Sweden Slow move from “six months - As Norway from 1 November 1999
in advance” to “one month in
advance” deadline
Germany Every month - Not implemented yet.
New Zealand Any time - Approximately 1-2 weeks.
As it can be seen in the above table there is a difference between the theoretical possibility for
changing supplier (according to existing regulations in each of the studied countries) and the
practical possibility related to the needed technical procedures for this supplier changing.
Where an advanced electronic system used for exchange of information for change of supplier
is implemented the necessary time for performing this action is shorter.
The procedure for change of supplier involves the customer, the new supplier, the network
company and the old supplier. Generally, the new supplier notifies thenetwork company
about the change and the network company then informs the old supplier. Afterwards both
suppliers will agree a process for a final reading or deemed final reading of the meter of the
customer who is changing supplier.
During this switching process there are many issues that will need to be addressed. The range
of these issues and their position in the switching process are illustrated in table 2.5.

14
Table 2.5: Potential issues for the procedure for changing supplier
Process Issues
1. Customer decision to - Are the customers aware of the market?
switch supplier - Are the customers aware of the procedure to change suppliers?
- Are suppliers providing accurate information?
- Is there a need for regulation or a voluntary code of conduct?
2. Transfer - What steps does the customer have to take?
administration - What “switching” costs will the customer have to pay?
- How are the distributor and the supplier notified?
- How will a record of customer/supplier relationships be
maintained and modified?
- Who retains the customer information?
- Who has access to customer information?
3. Install new meter - Will new meters be required for customers who switch?
(and communication - Who bears the cost of a new meter, if required?
equipment) ?
- Who will decide the new meter specifications required?
- Who owns or will own the meter?
- What metering options are available for customers with
unique load profiles?
4. Obtain final meter - Is remote meter reading economically viable?
reading - Is integration with the distributor’s normal meter reading
process possible?
- Can estimates of readings be used?
5. Change data storage - Where is the metering data stored? (It is centralised, or
and aggregation decentralised?)
- Who has access to what data?
- How will the “right” to access be transferred?
- How will aggregation per supplier be altered?
- How will distributors and suppliers access the data?
- How to ensure data access is not a barrier to competition?
6. Determine (or - How many different “deemed” load profiles are required?
”deem”) load profile - Will time interval meters reduce the need to offer different
“deemed” load profiles?
- How will customers be assigned a profile class?
7. Customer transferred
to new supplier

15
2.5 Costs resulting from changing supplier (who pays)

The opening-up of retail competition is not possible without offering the possibility to all
consumers to switch retail suppliers if they wish. This is the reason that most countries have
adopted new legislation to let all consumers have the possibility of changing supplier.
In table 2.6 shows a synopsis of how the costs resulting from changing supplier are supported
in some of the countries considered.

Table 2.6: Costs resulting from changing supplier

Country Fees for changing Compulsory change of Costs associated with

supplier meter for those change of supplier
consumers who have to procedure (costs of
be metered on time balance settlement etc.)
interval basis
No, for the first changing Yes, even if the local Covered by the
Finland
of supplier. Some network supplier wins the distribution network
operators collect a charge competitive bidding. owner.
in spite of regulations. However this order is
not always followed,
because it is in the
interest of both the
consumer and the local
supplier to use the old
meter.
No. No. All customers above Remuneration covered
Norway
400 kWh already have through the network user
metering and fee by the network
communication owner.
equipment)
No. No.
United
Kingdom
No (Government Covered by the network
Sweden
proposal). owner.
Yes, after the second or
third change (some
associations and Swedish
Power Grid proposal).
No (proposal).
Germany
No. However it is possible No. Covered by the
New
that some suppliers charge Customer Switching
Zealand
the consumers for Development Fees paid
changing supplier. by MARIA and NZEM
participants, on the basis
of non-half-hour
metered load.

16
The fees for changing supplier were removed (but in some countries only for the first
changing of supplier) because it was considered that even small fees reduce the willingness of
the end users to change suppliers, and the fee would have the effect of locking the customer
into the supplier. However there are cases when some suppliers charge the consumers for the
changing of supplier and in some jurisdictions this is permitted forthe second or third change.
Generally the costs associated with change of supplier procedure are covered by the network
operator through the normal transmission fees paid by the all network users. There are not
permitted any fees or tariffs in excess. One can regard the costs a network owner incurs from
handling supplier changes as part of their duty as a monopolist.

2.6 Costs of metering – included in the tariff?

Metering and data collection are important aspects of market design. The communication with
the meter is costly and has many implications, since it
- is an important component of the interface between the customer and both the supplier
and the distributor (network company),
- is a critical component in both the retail and the wholesale market settlement process,
- impacts on feasibility of data collection methods, and
- is technically important for the control of the network.
At the same time for low costs metering data collection it is necessary that:
- Data collection should be done by one collection agency in each area if the collection is to
be done efficiently.
- Metering service and data collection technologies will need to be compatible. Standards
would be needed for the meters themselves, the meter data formats, the communication
protocols, metering data management systems, the interfaces between meters and data
collection technologies, and between metering data management systems. These standards
must have an open architecture to support data collection, communication, and
aggregation by a variety of independent metering service providers.
- Competition in metering provision and data collection could still be supported by
providers competing for the contracts to provide these services for each area.
The metering service business is high volume, low margin business in which the service cost
and efficiency of operation are key. As the industry changes with suppliers no longer being
restricted by geographical location, a comprehensive metering service throughout the whole
market provided by a third party becomes an attractive option. The change to the industry
structure is likely to result in a small number of new players emerging who can offer efficient
metering services for both gas and electricity and possibly water.
Table 2.7 responds at some of the questions related to meter ownership, meter reading meter
and data collection technologies for the analysed countries.

17
Most meters are owned by the distribution company responsible for the meter installation,
maintenance, opening reads e.g. following change of supplier. Thedistribution company is
also the asset owner and is responsible for the periodic replacement of meters. Most
distribution companies have a diverse set of meters within their operating area. This means
that for organisations that have sites that are spread across many distribution areas will not be
able to have the same information gathered from each site. With the introduction of
competition, the possibility of letting the consumer own the meter exists, so that common data
can be obtained from all sites. The consumer could own the meter outright or could lease the
meter from a third party agent (who is responsible for maintenance etc.). However, if other
parties than the distribution/network company own the meter (i.e. independent meter readers
or the consumer), it is vital to establish who is responsible for controlling the meter.
Independent meter-readers may well emerge, but a clear division of responsibilities must then
be established.

18
Table 2.7: Meter ownership, meter reading and data technologies (N = normal kWh-meter, I = interval meter)

Country Who owns the Who reads the If the consumer Data collection Standards used
meter? meter? reads the meter, is for the meter data
there a verification Who collects How often? Who pays for
formats
system? the data? it?
Finland N: Network owner N&I: Network N: Network owner Network owner or Various rules for Network owner EDIEL-messages
owner is responsible to Collection & different cases & compulsory on
I: Usually network Supplier for extra
read once per 2 Settlement Service steps 01.09.1999
owner, legally services (i.e. on-
years Provider
allowed for line data
consumer as well I: Reading by tele- transmission)
communications
Norway N&I: N&I: N: Network owner Network owner N: Household- Network owner EDIEL-messages
is responsible to normally, but the customers with according to
Network owner Network owner (MSCONS) used
read every third task may also be annual requirements.
from 1998 between
year for domestic handled by consumption
If an interval the parties in the
customers. independent above 8000 kWh/y
meter is installed market.
service-provider 4 or 6 times per
For customers which is not in
year. The previous
with measurement compliance with
standard format GS2,
transformers, Various practice requirements, the
is still used as an
meter-reading shall for other customer party who requires
internal format i.e. to
be performed every categories the special meter
collect measurement
year must pay
data from customers
I: By tele- to the network
communications. owners’database.

19
 Country Who owns the Who reads the If the consumer Data collection Standards used
meter? meter? reads the meter, is for the meter data
there a verification Who collects How often? Who pays for
formats
system? the data? it?
United N: Network owner N: Supplier’s N: Customer may N: Supplier’s N: Monthly for N and I: N: Mostly manual
Kingdom by default (as from Agent (Data ‘self read’, verified Agent (Data commercial Ultimately, the reads, some IEC 1107
01/04/00). Collector). by Suppliers Collector). Customers, Customer, bundled to hand held units
Supplier can ‘take Network owner has Agent. Obligation Network owner Quarterly (3- in with his supply
over’ but dealt with obligation to to read at least has obligation to monthly) for contract or tariff
by Suppliers’ procure meter every 2 years. procure data domestic. terms. There are I: Remote collection
Agent (Meter reading as a collection as a some contracts uses format specified
Operator). Legally default. default. directly between by the Pool interfaced
allowed for I: Not allowed for I: Daily Customer and through MV90.
Customer to own Customer to read – Data Collector (as
as well. I: Supplier’s remote collection I: Supplier’s there are between
Agent (Data as per Pool Agents (Data Customer and Meter data transfers
I: Supplier is
Collector) has Procedures Collector ensures Meter Operator) as use a ‘Data Transfer
responsible for
responsibility but data passed to a legacy of Network’ which may
provision,
read remotely each Data Aggregator) arrangements prior be by electronic
discharged through
day. via remote to 1998. means.
Agent (Meter
collection
Operator). MO
owns and leases to
customer or
customer buys and
owns.
Sweden
Germany network owner network owner sample network owner monthly supplier ERIFACT

New Zealand

20
The reading of domestic meters as part of a data retrieval activity is a very manual process.
The key aim is to reduce the number of estimated readings and to increase the accuracy of
those readings. The technology used requires that walk-orders of customer details are
downloaded and consumption readings uploaded. The data associated with customer details is
often limited to be mainly address details and possibly meter location. Additional information
such as expected meter reading (within a tolerance either absolute or as a percentage value),
preferred time of visit etc. would greatly improve the service and data quality (by reducing
reading mistakes).
In table 2.8 the regulatory status of metering and the modalities for covering the costs of
metering in some of the countries considered in our analysis are briefly described.
Table 2.8: Regulatory status of metering and the modalities for covering the costs of
metering
Covering the costs of metering Regulatory status of metering
Country
Finland Currently included in the network The acceptation of the Regulator
proposal would improve the
tariff.
possibilities of the companies
Proposal from the Regulator on a providing separate metering services
separate metering tariff. for entering the market.
Currently included in the network Network owners are responsible for
Norway
tariff. ensuring the measurements.
Network owner is free to farm the
meter job but not the ownership of the
meter.
United Metering services are included as a The supplier is responsible for
Kingdom ‘package’within the supply contract. appointing agents for meter operation,
data collection and data aggregation,
and he will pay for metering services
through contracts with each agent,
passing such costs on to the customer.
Sweden Currently included in the network Network owners are responsible for
tariff. ensuring the measurements.
Germany Metering is a separate service.
New Metering is a service provided by Data Administrators have
Zealand different companies. responsibility for gathering the meter
readings from data collectors,
preparing estimates in instances when
no meter readings are taken, and
aggregating the metering data for
provision to the NRM.
The metering costs include the costs of the interval time meters (equipment, installation and
calibration) as well as the long-term costs (meter reading costs and telecommunication
charges).
In most cases the network owners are responsible for ensuring the measurements and
gathering the metering data, but the possibility to farm the meter job out to other metering

21
companies or to let others install and possibly even own the meters exists. In this case the
costs of metering are currently included in the network tariff (in the contract).
In other cases, the responsibility for gathering the meter readings from data collectors belongs
to other entities and in such cases obviously the costs of metering are separated from the
network tariff.
Therefore, depending on the market structure, customers may receive a single bill (usually
from a supply company) or two bills (one from the supply company and one from the
distribution company which often includes the charges for metering services). A single bill
requires that the supply company pays the distribution and transmission companies for using
their systems to deliver the energy to the consumer; the charges for these services are not
identified as part of the total bill.

22
3. Conclusions and recommendations
The new small consumers market (residential and low energy users, now able to change their
electricity suppliers) will probably change and evolve over time. In the beginning suppliers
will be inexperienced in the operation of this competitive electricity market and will require
time to develop a range of service offerings. Customers will be inexperienced too and will
need to learn about the market and the benefits it can bring to them. With time the share of
customers selecting different suppliers will increase.

Change of supplier (Norway)

100000
90000 Customers not supplied
80000 by local supplier
70000
Customers that have
60000
changed supplier this
50000 quarter
40000
30000
20000
10000
0
97

98
97

98
6

9
7

8
y9

y9

y9

y9
r9

r9
ril

ril
ly

ly
ar

ar

ar

be

ar
be
Ju

Ju
Ap

Ap
nu

nu

nu

nu
cto
cto
Ja

Ja

Ja

Ja
O

In a dynamic market the flexibility in designing the market framework will be critical. The
choice of time interval metering versus load profiling is a market design issue, which is
fundamental to the development of residential and low energy users market. Whichever
solution is chosen to solve a market design issue now will not necessarily be the correct
solution in the next few years time, or for all customers in the market at the one time. Also,
the technologies and costs of the market support systems will change over time.
Therefore minimum cost solutions are recommended which can be upgraded in time.Some of
the markets described in this report provide examples of how it is possible to give residential
and low energy users true retail access without introducing costly and complex metering
systems as a necessary requirement, and that profile-based metering and settlement, if
efficiently implemented, is a feasible and cost effective way to give the third party access for
small customers.

3.1 Threshold levels for metering based on time interval

The primary concerns related to imposing time interval metering requirements relate to the
effect of such requirements on the development and operation of competitive markets. An
argument against imposition of interval metering requirements for smaller customers is that
such requirements are a deterrent to customers choosing new retail suppliers. As a result,

23
these requirements slow the development of competitive markets, particularly for smaller
customers. This argument can be addressed to some extent in terms of an analysis of meter
affordability for different groups of customers.
In favour of lower thresholds for interval metering is that interval metering means that
customers face their true cost of service. Giving customers and suppliers better information on
the true cost of supply means greater overall economic efficiency. There is general agreement
on this economic principle.
A further argument in favour of more widespread interval metering is that including larger
customers in load profiling adds to the volatility of settlements based on profiling. That is,
requiring interval metering at a lower cut-off is expected to reduce the variability of the load
profiling errors for those below the cut-off.

3.2 Which load profile model is adequate where?

In a world with no transaction costs, the most efficient method of determining electricity
consumption would be the installation of time interval meters (usually quarter-hourly, half-
hourly or hourly) at each point of consumption. This would enable the true costs of electricity
for each consumer to be identified without distortion. However, in practice there are costs
associated with installing and operating those meters, as well as the communication with
those meters, i.e. through telecommunication lines. It is generally believed that for most small
electricity consumers metering on basis of a metering time interval is not an economic
solution because the metering installation and operational costs exceed the potential cost
savings.
The use of load profiling to determine a consumer’s electricity consumption also has costs
associated with it. As with metering on basis of a metering time interval, these costs must be
weighed against any potential cost savings in order to determine whether the introduction of
profiling is economic. Today two different approaches for load profiling are used, the area
and the category model. They can be used either separately or combined. Both of them have
advantages and disadvantages in some specific conditions as it can be shown in the table
below.

24
Table 3.1: Area model versus category model: advantages and disadvantages

Area model Category model

Advantages - It can be implemented immediately in - Supports the use of
the systems where representative data multiple customer
for load profile construction are not categories or profiles;
available. - Increases the accuracy of
- It is cheaper and less complex (only the average load profile as
one consumer profile per network only members of the
area) than representative sampling for customers class are
the systems with a small number of included in the sample
networks. groups;
- Simpler to administer and requires - The total number of curves
less IT-support than the category is lower than in the area
model for the systems with a small model for the systems with
number of networks. a large number of
networks.
- The ability to easily obtain dynamic
profiles. - All players are able to
know in advance the load
- Minimise the risk to the network
profile assigned to a
owner especially in the networks
specific customer.
where there is a large variation of the
temperature.
Disadvantages - The profile must be updated - The model cannot be
regularly due to the local variations applied without systematic
and customer charges. measurements done earlier.
- Variation in the customers in the - Creating too many profiles
network area may create inaccuracies (more than 10) will
in the profiles (possible cross- increase profiling costs;
subsidising between the consumers). - The difficulty in choosing
- The need of consumption estimates a representative customer
from the suppliers every day. sample and costs of the
sample required.

Obtaining a representative deemed load profile is important because this will affect the
average price paid for the electricity by the supplier for their customers’ consumption, and the
supplier will eventually pass some variation of this price on to their customers. If there is a
large variation in the consumption patterns of customers assigned the same profile, there is the
risk that some customers are cross-subsidising other customers using the profile. This could
lead to a perception of the system being unfair, to arguments over which profile a customer is
assigned, and to some customers feeling obliged to purchase meters to reduce their electricity
costs. However, the errors or variations introduced by the area model may not be any greater
than those introduced by any other profiling system. In addition, the errors from profiling may
be less than those in the existing tariff systems.

25
 Risk according to ASLP

ASLP
Winner
Loser
Price

T im e (o n e y e a r)

Example of two customers with consumption spread differently over the year, and a price level which changes
over the year. The client with the relatively high consumption during the time with high electricity prices “wins”,
but the client with relatively low consumption “loses” and pays a higher average price than if metered.
(Norwegian Ministry of Energy).

If there is greater variation in the load profiles of customers, more customer categories and
hence average profiles, will be needed. The determinants of the number of customer
categories required will be:
- the size of the variation in the load profiles of individual customers within a customer
category;
- the extent that the wholesale market spot price varies;
- the degree of variation within the class that is acceptable to suppliers, consumers and
regulators;
- whether the additional costs and complexity of using multiple customer categories is
acceptable and worthwhile.
Increasing the number of classes or profiles should lead to increases in the accuracy of the
profiles. It is not only the accuracy of the profile which is important but the variation in
average energy price that matter. Customers may have widely varying load profiles but their
average energy price may be nearly identical. In the Nordic countries, where electrical heating
is quite common, but not the only source for heating, the electricity consumption will increase
with 4%/degree below 15 degrees Celsius. In Germany, the temperature coefficient is 0.8
%/degree. Hence, the profile of a Nordic customer will be very different, depending on
whether there is 100 % electrical heating, 50% electrical heating, or no electrical heating. The
load variation between customers would have a little impact on their average energy price if
the variation of the price is low. This is the case in Norway where one profile is used for all
customers on a distribution network and the daily variation in the spot price (used for
reconciliation) is very low. Therefore the area model is appropriate to use on such markets
and the gains in accuracy using the category model may not result in significantly different
average electricity prices being paid by the individual customers.

26
At the same time, using multiple customer category profiles does decrease the volatility and
bias in the settlement process. However introducing a profiling system without customer
category level profiles will still result in a measurement system that can support reasonable
wholesale settlements. Therefore a balance between accuracy and the number of profiles
needs to be done according to the specific conditions of the systems and markets considered.
Moreover the need and cost of regularly reviewing of the accuracy of the load profile sample
to ensure it is still representative is added.
As a general rule the area model is adequate for the first implementation of load profiling
method where representative data for load profile construction are not available. If
consumption data for each time interval considered has been collected from a sufficiently
large sample of customers, then a profile can be developed from this data for that class of
customer. The minimum period for data collection is a year and probably at least another year
will be needed to form the appropriate class of profiles for the specific conditions of the
relevant market. Therefore only after the load curves are available (i.e. after at least two years
of practice using the area model where the needed data for load profile construction are not
available from the beginning) the most suitable load profile model for the specific conditions
of each market between the area and the category model can be chosen.

3.3 Settlement of the deviation between estimated and actual consumption

The use of load profiling to determine a consumer’s electricity consumption inherently
introduces discrepancies between estimated and actual consumption. There are different
methods to settle this difference and it is recognised that the settlement based on load
profiling involves certain risks for some of the market players in certain conditions.
Obviously the risks associated with settlement based on load profiling depends on the load
profile system and the settlement method used.
To minimise these risks, firstly, it is required that end users with a high consumption of power
could have a great effect on the load profile to be metered on a time interval basis (e.g. hourly,
half-hourly etc.). If the large consumers with deviant consumption patterns are included in the
load profile used, then the profile will not be representative for a large segment of the
electricity consumers including, among others, households.
Generally, the risks associated with settlement based on load profiling supported by the
Transmission Network Operator, Power Pool and network owners are not significant. The
major risk for them is that the suppliers are insolvent at the time of settlement in situations
where their customers have consumed more power than expected.
Suppliers and individual end users are exposed to certain risks in some particular situations
related to the periodicity of the meter readings. (For instance in the settlement based on the
adjusted system load profile the meter readings shall be taken at the same time for all the end
users in the network for an equitable settlement between them). The price used for balancing
the estimated and actual consumption, loss allocation between the suppliers in the network or
a possible cross-subsidisation between large and small consumers in a network is also
important. (For instance when the small consumers with largely off-peak consumption
patterns must pay more for their electricity because of the influence of the load shape of large
electricity users consuming energy during higher-price peak consumption periods).

27
Appendix 1: Description of the already existing metering load profiles and settlement
systems

Finland
1. Background information
1.1. Electricity sector in Finland
Power producers in Finland have divided themselves into three principal categories: about 40% of all electricity
is produced by basically government-owned companies and another 40% by companies owned by private
industry. The remainder mainly consists of municipal CHP. The share of wind power and similar small-scale
power plants is only minor.
The HV power transmission network was formed by having the different power producers build lines from their
respective power plants to locations where they wished to sell electricity. Gradually, these formed an integrated
national grid, parts of which had several owners. Ultimately, it was not until the year 1997 that the single
company model was implemented in the grid.
The number of distribution organisations is about 110. About 15% of them are municipal utilities, 30% are
municipal companies, and the rest are privatecompanies.

1.2. Electricity Legislation

In the 1980s, the Finnish Government took a number of legislative measures in order to boost competition. These
measures also include the reform in the Electricity Act to create deregulated electricity markets. After lengthy
preparations, Parliament confirmed the Electricity Market Act, and the electricity markets started, step by step, to
open up between the years 1995-1996. Small consumers were practically left outside the markets until the
autumn of the year 1998, since, before that time, all consumers on the market had been required to have a meter
registering electricity consumption by the hour. Since the autumn of 1998, (private households from September 1
1998, others from November 1 1998) a consumer's electricity consumption can be estimated by means of load
curves, if the size of the consumer's main fuse is maximum 3 x 63 A.
In terms of electricity distribution, the system based on the Electricity Market Act is the following:
- The principle of the law is to avoid regulatory action and to leave all regulation to the markets, as much as
possible.
- Electricity transmission in the grid is a monopoly.
- The former electric utilities must divide their operations into accounting units (or separate companies).
Regulatory action is directed to distribution network operations (network unit), which comprise a monopoly.
Electricity supply to end customers is submitted to competition. Should a customer not wish to have vendors
compete against each other, electricity is sold to him by the local supplier (the supply unit of the former
electric utility), who must have public tariffs for this purpose.
- The distribution network operations must yield reasonable profits (instead of high profits).
- The activities are being monitored by the Electricity Market Authority (EMA, The Regulator).
The respective organisations (for example, the Finnish Electricity Association, FEA, in Finnish SENER) issue
guidelines and recommendations on the activities.
Between the grid and the distribution networks, there may exist individual regional networks that mainly own
110 kV lines. These companies are treated in the same way as the distribution networks.

1.3. Degree of market opening and future thresholds

Fall 1999, that the market has been open to all for about a year, only about 1% of all consumers have decided to
change their electricity supplier. About one third of them are small consumers. This is the prevailing situation, in
spite of the fact that some vendors have made considerable investments in their marketing. Approximately
10...20% of the customers buy their electricity from their previous suppliers, but at bargain prices. In one
company, as much as 40% of electricity sales take place at bargain prices, both to small and large-scale
customers.

28
According to the statistics collected by the Regulator, the tariffs of small customers fell clearly in the autumn of
1998, just as these customers were brought within competition. It is true that the tariff of other customers went
down as well, but the drop was sharpest on the part of small customers.
There are no major thresholds for the customer to change supplier. Because the prices of individual suppliers are
very near each other, the possible benefit for a customer who changes supplier is low. That’s why the number of
consumers who do it, is expected to increase very slowly.
Perhaps there is a threshold left for those consumers, whose main fuses are just bigger than those allowed for
load curve consumers. Sometimes the price of the hourly meter is thus too high and makes the business with a
new supplier unreasonable. No developing measures are planned to make the situation better.
The expected development of supply & distribution companies is that supply business will be incorporated into
larger units or at least into some co-operation chains. So far, only a couple of supply & distribution companies in
Finland have entirely given up the supply business. Distribution companies already have jointly founded a few
supply companies. The number of distribution companies is likely to diminish very slowly.

1.4. Political and regulatory background for the current Metering, Load Profiles and Settlement System
According to the Finnish Electricity Market Act, the network operator is responsible for the installation, function
and reading of meters. The Finnish Electricity Association (FEA) has drawn up a recommendation concerning
the properties that are required of metering equipment. The recommendation concerns, among other things, the
accuracy of metering equipment, the lengths of emergency operating times, etc.
According to law, the metering and the balance clearing are part of the duties of the distributor operator, and the
cost will be covered in the distribution tariffs. The acquisition cost of the hourly meter can be charged separately
from the customer, but the customer must also have a right to acquire the meter individually, if he so wishes.
This right has raised a lot of criticism, because it is not clear what to do, if the meter fails.
It is the liability of the network operator to do the reading of other than hourly meters at least once in every two
years, however, in leisure time residences at least once in four years. It is the consumer’s responsibility to report
the readout of the meter when the seller makes an inquiry about it. During this period of time, billing can take
place on the basis of the consumer’s estimated consumption.

2. Tariffs
2.1. Tariff System
Regardless the distances, all distribution networks must have independent tariffs, non-discriminatory to the
different customers and specified for each voltage level (in practice low voltage, medium voltage and high
voltage). There are no specific transformation tariffs. The law does not set any requirements to the tariff
structure. The Regulator only monitors the reasonableness of the overall output and the equal treatment of both
customers and electricity vendors using the network. The Regulator has issued a number of guidelines that are
connected with equal treatment and other trade policies. The Finnish Electricity Association has investigated a
way to define reasonable output, but has not taken a stand on the amount of output.
At the moment, the tariffs of the different distribution networks are individual, as to both price and structure. As
the former utilities had to divide their operations into distribution and supply, they, in most cases, divided their
old tariffs in two parts without making any fundamental changes in the structure of the tariffs.

2.2. Description of the Distribution Tariff Structure

The distribution tariffs usually include the following charges:
- Fixed charge, which often depends on the capacity of the consumer's main fuse.
- Energy charge: For the smallest consumers, the same price is applied around the year, but larger customers
are required to use a dual-works kWh meter and a timer, or network command control. Energy has a high
price on workdays in wintertime, for example, but a low price correspondingly in the summer, on weekends
and at night.
- The ratio between fixed charges and the energy charges may vary greatly between the different distribution
companies.

29
- The largest consumers may include a charge in their tariffs, connected to peak load.
- Reactive energy is free of charge up to a certain limit.
The customer may select the desired tariff. The distribution tariff may be selected regardless of the selected
supply tariff.

2.3. Connection fees

Buildings intending to be connected to the network must pay a connection fee. Traditionally, the system has been
such that the fee covers at least the construction cost of the connecting cable or the connecting overhead line.
Many utilities have charged a slightly higher connection fee that has also covered some of the other construction
costs of the distribution network. It is expected that The Regulator will later start issuing guidelines also on the
harmonization of connection fees.

3. Metering
3.1. Choice of the Threshold Level established for metering
When we talk about the costs of hourly metering, we often mean the mere purchase price of an hourly meter. It
must be remembered that hourly metering requires a modem and a telephone line for data transmission, the
acquisition and connection costs of which amount to some 100 Euro, even when there is another telephone on
the site in question. The annual telephone costs play even a more significant role, amounting to some 100 Euro
as well. As far as small consumers are concerned, such costs will easily create an insurmountable obstacle to the
use of an hourly meter, even though the meter itself can be very inexpensive. The hourly meter has also proved
to be sensitive to climate-related overvoltages in rural circumstances.
Thus it’s clear, that small consumers can take part in the free market only by load profile method. In Finland the
planning of the system was started with definition of the method and even the details. After that begun the
discussion of the possible users of the method. The threshold was finally set to 3 x 63 A for all kind of
consumers.

3.2. Description of Metering System used

The meters of the small consumers are normal kWh-meters. Other consumers than households shall however
have a 2-time zone meter, if their main fuses are equal to or bigger than 3x35 A, and if the consumer will take
part in the free market (to buy from an outside supplier or from the local supplier with bargain price).
Above the threshold level, a consumer who is not taking part into the free market, may use the meter he had
previously. So it was not necessary to make comprehensive meter change operations connected with the
beginning of the free market for small consumers.

4. Load Profiles
4.1. Choice of the Load Profile Model used
It is known that there are two methods for the formation of load curves: the regional method and the consumer-
group-related method.
In the regional method, all large consumers are equipped with the facilities for hourly metering. Energy used by
large consumers is deducted from the total energy consumption by the hour. The result, in other words, the
difference calculated by the hour, is considered to illustrate the consumption of all those consumers who are
entitled to use the load curve method in the area in question.
The consumer-group-related method defines the load of a consumer taking into account the average load of a
group. Even three groups may be enough to reach a sufficient accuracy. The precondition for this method is that
load investigations (measurements) have earlier been made enough. An annual reconciliation (calculation for
deviations between the vendors) is connected with this method.
The advantages of this method from the Finnish point of view are
- The number of curves is lower than that in the regional method

30
- Special factors (i.e. correction according to the temperature) can be taken into account
- The accuracy can be increased later by making more load investigations

4.2. Description of the Load Profile Model used

In Finland, the consumer-group-related method has been chosen. Only large consumers, who have changed
supplier, must be equipped with the facilities for hourly metering. Four national load curves for various
consumer groups have been defined:
1. Households without electric heating
2. Households with electric heating
3. Other consumers below 3x35 A
4. Other consumers 3x35 A or bigger (must be equipped with 2-time-zone meter)
Load curves can be applied to consumers whose main fuses are up to 3 x 63 A. For the group 2 the temperature
correction is used.
The method has worked well so far. Some local load curves, confirmed by the Regulator, have been adopted.

5. Settlement
5.1. Description of the settlement method used
The development of the balance settlement system in Finland has taken place gradually in line with the
electricity markets that opened in progressive steps. The details of the system have been implemented on the
basis of the recommendations drawn up by FEA. Legislation only provides guidelines, but in spite of this, it has
been necessary to revise the regulations twice in this respect.
The balance settlement system is divided into three steps. The lowest step is formed mainly by distribution
network operators/suppliers. They deliver balance information to companies who have registered as the Balance
Responsible Supplier (BRS). A typical company registering as the Balance Responsible Supplier is a large
power producer. In the early stages, there were only two BRS on the market, but by now as many as 16. The
highest step, which takes care of balance settlements nation-wide, is comprised of the companyFingrid System
Oy, which functions as part of the Grid Operator.
The advantage for a supplier to register as a BRS is the fact, that other counterparts then do not know, with
whom the supplier is making business. From the other side, a BRS has to make some extra reporting needs some
extra skills etc. This is why it’s not reasonable for a small supplier to register as a BRS.
The system underwent a significant test, as free electricity trade - taking place by means of load curves - was
introduced on the part of small customers on September 1 1998. The system has worked in a satisfactory way. It
is true that there were some difficulties at the turn of the year 1999, as many new BRS operators joined the
balance settlement system. Another reason for the difficulties was that the biggest consumers, mainly the wood
processing industry, use electricity in a different way during the Christmas season than during ordinary weeks.
It has often been difficult for network operators to comply with the deadlines that have been set for balance
settlements. This is why problems arose in situations in which a certain amount of electricity should have been
recorded in the balance sheets of two BRS. If one of them received the information concerning a certain amount
of electricity, but the other one did not, a misunderstanding could not be avoided. Deadline stipulations should
therefore be developed further. Sanctions should be available as well.
Mistakes in balance settlements accumulate on the Grid Operator level and become visible as “extra” or
“missing” electricity in Finland. Since the expiration of the fixed term reserved for balance settlements, there
have been so many unexplainable mistakes that have affected the finances of the different parties to the
electricity trade. For this reason, the distribution network operators have been requested to send their summed
vendor-related delivery information directly to the Grid Operator. With the help of this information, checking
calculations can be made among BRS companies. The distribution network operators have taken a positive
attitude to the request, and the procedure has been observed to be very beneficial.
Electricity vendors and distribution network operators are provided with an opportunity to use the www-balance
settlement pages of the Grid Operator. Here, the vendor or distribution network operator can view the
information relating to his company, connected with fixed deliveries and the summed deliveries of distribution

31
networks. In general, it is in the interest of vendors and distribution network operators to keep in touch with the
BRS-operator and ask different kinds of checking questions about the most important pieces of balance
settlement information.
Large numbers of messages must be transferred between the parties to balance settlements. The most primitive
way to transfer messages is to use the telefax, which method was used particularly by small network operators in
the beginning. A much better way is to use the Excel table that the parties have mutually agreed to use. Several
years ago, preparations were started for the implementation of automatic data transmission method for balance
settlement information in Finland. This method is based on the EDI system, which is used, for example, in the
daily connections of the grocery trade. The EDI application used in the electricity trade is called EDIEL (EDI =
Electronic Data Interchange). A corresponding system is used in Sweden and Norway. EDIEL messages account
for some 70 % of all messages, and their use will become compulsory on September 1 1999.
The advantage of the Three Step Settlement System is that the amount of data by any supplier or operator
remains relatively low. The disadvantages of the System are connected with the fact, that it’s difficult to control
whether the data is correct or not. Because of this the System will probably be developed so that distribution
network operators will report directly to the Grid Operator. The Grid Operator then collects the data, classifies it
by Balance Responsible Supplier (BRS) and sends it to them. Thus will the amount of messages decrease and the
Grid Operator can control the data if necessary. The need for distribution network operators to update where to
send the data (in accordance to the new suppliers), will decrease. The data processing system of the Grid
Operator will be comprehensive, but the development of those systems makes that it is not any more an obstacle.

5.2. Deviation between estimated and actual consumption for small eligible users
According to the Finnish legislation the Distribution network operator shall yearly make clear the differences
between the sum energy calculated for the load profile method and the sum energy really metered during the
same time. The sum energy will be calculated for the customers of any supplier. The reconciliation will be
credited and debited between the suppliers acting in that network. The basis for the crediting and debiting shall
be the tariff of the local supplier, which is suited best to each group. The reconciliation shall be made once a year
or more often. If there is a fixed charge in the supply tariff, it must not be taken into account.

6. Other questions resulting from choice of system

6.1. Time restraints and other conditions for changing supplier
The possibilities of the consumer to change supplier
According to regulations, the consumer can change supplier any time. The network operator must be informed
about the matter two weeks prior to the change. In practice, this matter is usually significant only in connection
with the first changeover, because in the free markets, the tenders submitted by suppliers usually concern fixed-
term contracts, for the periods of 1-2 years in general.
Problems often arise at the moment of expiration of fixed-term contracts. If the customer has forgotten that his
contract has expired, the network operator should in fact disconnect him from the network. In reality, the
network operator has usually informed the customer about the fact that the contract will soon expire, or has
already expired. In case the contract has already expired, the customer must find an electricity supplier, who is
willing to conclude a retroactive electricity contract. An established solution must be found to correct this detail.
On billing
The largest number of customers still buys its electricity at public tariffs and based on a contract with the local
supplier. The sales contract covers both electricity sales and distribution, and network services are managed
through the supplier.
Also an outside supplier can make sure that the consumer only receives one electricity bill. Still, in the bills of
both the local and outside supplier, a distribution charge, electricity sales charge, value-added tax, electricity tax
and possibly a fee charged by the supplier for collecting the distribution charge, must be specified.
When an outside supplier acts as the vendor, the consumer concludes a separate network service contract with
the distribution network operator. If desired, it is possible for the consumer to receive a separate bill from the
distribution network operator. If the consumer only receives one bill, the distribution network operator will send
his bill to the vendor. The consumer is responsiblefor paying for this portion as well, even at the threat of being
disconnected from the deliveries of electricity.

32
6.2. Costs resulting from changing supplier (who pays)
The charge for an extra reading of the meter
When the consumer changes electricity supplier, there must be an extra reading of his meter, unless the
changeover is scheduled to the precise moment of the annual reading. A debate is going on in Finland about the
network operator’s right to debit the consumer for this measure (about 20 - 30 Euro). All are unanimous in that a
charge can be collected from the consumer who often changes supplier. However, the stand taken by the
Regulator is that no charge must be collected for the first extra reading of the meter, because it would create an
obstacle to bringing the consumer within the sphere of free competition. Some network operators collect a
charge in spite of this, and the Regulator has taken the matter to a court of law. The court case has not been
settled yet.
Compulsory change of the meter, even if the local supplier wins the competitive bidding
When a consumer in Finland using the 3 x 63 A fuse moves to the free electricity market, he must take into
consideration the acquisition and other costs of an hourly meter in the changeover situation. If he continued to
buy electricity from the local supplier at public tariffs, it would not be necessary to change the meter. Of course,
it is the consumer who makes the choice. If the consumer invites tenders from several suppliers and ends up
buying electricity at a reduced price from the local supplier, there are no technical reasons for the use of an
hourly meter. This means that the consumer in question would still be able to use his old meter, thus saving the
costs of the hourly meter. This procedure provides a competitive edge to the local supplier. The Regulator
considers this competitive edge to be unfounded and therefore requires the introduction of an hourly meter also
in a situation like this. It has turned out that the order is not followed, because it is in the interest of both the
consumer and the local supplier to use the old meter. This problem does not exist, if hourly meters are installed
to all large–scale consumers in connection with the adoption of load curves, as has been done in Norway and
will be done in Sweden.
The costs of balance settlement
The costs caused by a customer who changes supplier
The management of balance settlement results in substantial expenses to the network operator. The costs must be
viewed as part of the price that society must pay for freely operating electricity markets. In Finland, only about
30 % of the network operators compile their balance settlements individually, while the rest are in the habit of
ordering the needed balance settlement services from other network operators or system suppliers.
When a customer changes electricity supplier, the change always results in costs, because the electricity meter
must be read. At least until now, also the other related costs have remained rather high. It has been observed that
one employee is able to deal with 10-15 changes of supplier in a working day, in other words, about 2000-3000
changes in a year. It is expected that this figure will rise as we get used to the system.

6.3. Costs of metering – included in the tariff?

The costs caused by the meter and the reading of the meter are currently included in the network tariff. The
Regulator has made a proposal on a separate metering tariff. In this case, each consumer would pay for the every
meter that is used by him at a given time (for example, an ordinary meter / a dual rate meter / an hourly meter).
This method would increase the efficiency of free electricity trade and simplify the rest of the network tariffs. At
the same time, it would improve the possibilities of the companies providing separate metering services for
entering the market. – From the angle of the consumer, the drawbacks would include a more complicated
electricity bill, because a third component would be included in it, in addition to the electricity price and the
price of network services. – It is likely that the metering tariff will be introduced in Finland at some stage.

6.4. Some detailed experiences

General Experiences
Many minor details can be found for discussion in the recently opened markets, of course. In Finland, not much
attention was paid to finding solutions to problems of this level in advance. This might explain why it has been
possible to quickly open the markets, with the help of only a restricted number of employees who were engaged
in the planning of the opening-up of the markets. It is effective to look for solutions afterwards, because the
problems are tangible and measures are normally taken fast to remedy obvious grievances. The fact that

33
consumers and also other parties sometimes end up in unclear situations can be regarded as a drawback.
However, from the point of view of consumers, it seems that the markets have been opened in a relatively
successful manner.
Discussing time zones and the dual rates of network tariffs
Dual rates of end customer tariffs are commonly used in Finland. One of the reasons for this is that about 20 %
of all consumers in Finland have electric heating. With the help of dual rates, the heating of household water, for
example, is steered to take place mainly at night. The dual rates of network tariffs are also used to implement a
lower electricity price in summer, at which time electricity is produced by means of cheaper production methods.
In Finland, each local supplier has his individual dual-rate time zones. There are, however, two common
solutions, of which either one is used by some 80 % of the local suppliers. The dual-rate (time-of-day)
distribution tariffs used by the distribution network operators normally follow the same zones as the tariffs of the
local supplier. Consequently, when tenders for dual-rate electricity are submitted by outside suppliers, they must
closely follow the time zones applied in that particular area. This can be considered an obstacle or at least a
threshold to the free electricity trade, and this is why the Regulator has decided to look into the matter. As a
solution, the Regulator could prescribe that single rates are to be applied in all distribution tariffs, in which case
the electricity supplier would be free to choose the time zones on the part of each consumer, within the
framework of the technological properties of the meter, of course. Another alternative would be to allow dual
rates, but the division into inexpensive and expensive hours (or days or periods) would be standardised in the
whole country. A third alternative would be to introduce meters equipped with four calculation devices.
The matter is still being looked into, and each distribution network operator follows his own practice.
Joint purchases and back meters
Since the liberalisation of the Finnish electricity markets, it has been possible for consumers living in the same
building to begin to buy their electricity in the form of joint purchases. A typical joint-purchase customer is a
block of flats, which after having changed over to the new method, only needs one meter for the metering of
electricity (usually an hourly meter). Each flat still has a meter, which in a joint-purchase situation is managed
by the building. The meters of the flats are left in the “back” of the metering performed by theelectricity
company, and this is why they can be called “back meters”. The house manager of the building is in charge of
dividing the electricity bill among the residents.
Joint metering is considered to be in the common interest, because the residence-related costs of the hourly meter
remain very low. In a situation like this, the residents of the building have an opportunity to make electricity
suppliers compete. Additionally, the residents of many buildings have noticed that it is possible to save in the
charges included in distribution tariffs by using joint metering, because the basic charge of joint metering is
considerably smaller than the sum of the flat-related basic charges. On the other hand, distribution companies
have started to develop their tariffs in order to remove this benefit, or at least to make it as small as possible.
However, there are residents in many buildings who for some reason do not wish to go along with joint metering.
In Finland, legislation does not usually provide sanctions that a building can use to implement joint metering.
Some interest groups, which supervise the interests of buildings, have come up with a solution, according to
which the resident who is unwilling to use joint metering could remain a direct customer of the electricity
company so that his back meter would still belong to the electricity company, and his readouts would be
deducted from the readouts of the meter that is used jointly. Considering that a joint meter is an hourly meter and
that a load curve should be applied in the back meter, what we end up with is a confusing situation, and – from
the point of view of data processing – also a nearly impossible one. Such a possibility should have been clearly
prevented by legislation.

34
Norway
1. Background information
1.1. Electricity sector in Norway
Electricity generation in Norway is almost exclusively based on hydroelectricity. The maximum generation
capacity available is 27,100 MW in 845 power stations. There is currently more than 60 production companies
;
33 producers own 96% of the production capacity. Of the 108 TWh of average annual production capacity,
54.6% is owned by municipalities and 27.4% byStatkraft SF, with the remaining 18% privately owned.
The national transmission network (above 130 kV) is operated by the state-owned company Statnett SF, the
regional network (130 kV and 60 kV) is owned by regional utilities, and the local grid (20 kV and below) is
owned by distribution utilities.
There are 54 owners of medium-voltage networks – above 110 kV – and some 200 local distribution network
owners. Approximately half of the distribution companies own power production facilities, and roughly 60% of
the total energy is supplied through vertically integrated companies, including power intensive industrial
customers that supply their own power.
All utilities and other large companies (traders and brokers) participate in the wholesale market. Also large
consumers can participate instead of purchasing through utilities brokers or traders.
The power market consists of three distinct markets:
- The regulation market (physical) – is primarily a balancing market where bids are used to price imbalances
between participants and contracted generators.
- Elspot (physical) – is a day-ahead market where power is priced for five daily intervals during the week and
three intervals on the weekend. Short-term delivery, i.e., for delivery next day, is linked to physical delivery.
- Elfutures (financial) – is a week-ahead market extending two years into the future.

1.2. Electricity Legislation

Key features behind the Energy Act who entered in force on 1st January 1991 include:
- Remove responsibility of operating the national transmission grid from the main generating company
(Statkraft) to a separate new company (Statnett), to ensure transmission access on a fair and non-
discriminatory basis;
- Open access to the transmission and distribution grid to any licensed generator or supplier, so that both are
able to offer tailored bilateral contracts to their customers. All qualified participants have access to the
national grid and a spot-market;
- The Act does not require utilities to separate distribution and supply functions into separate corporate
entities, but requires separation by accounting only.
Due to the fact that the development of the power market has changed so much since the Energy Act and
regulations pursuant to the Act were written in 1990, the Ministry of Petroleum and Energy and Norwegian
Water Resources and Energy Administration (NVE) found in the spring of 1998 that it was necessary to update
the rules and regulations and give the NVE a more precisely defined legal authority. The Ministry of Petroleum
and Energy had also received a legal report from the Norwegian Electricity Federation which included a
discussion of the NVE's legal authority.
The work to revise the regulations pursuant to the Energy Act started in the spring of 1998 and was led by the
Ministry of Petroleum and Energy. This work resulted in the distribution of revised Energy Act regulations by
the Ministry of Petroleum and Energy for comments in September 1998. In these regulations the Norwegian
Water Resources and Energy Administration (NVE) was given the authority to issue its own regulations in
specific areas that included metering and settlement.

35
1.3. Degree of market opening and future thresholds
The table below lists the annual changes in the regulations governing market access for end users in Norway.
Table 1: Annual changes in the market access regulations for end users
Introduction of the Energy Act.
1991
Hourly metering requirement for change of supplier.
Local supplier has a significant market advantage.
NOK 5000 per year per customer stipulated as maximum fee for using a supplier other than the
local supplier.
1994 Maximum fee reduced to NOK 4000 per year per customer.
Hourly metering requirement for change of supplier eliminated.
1995
Settlement based on the adjusted system load profile.
Non-hourly metered end users can change suppliers on a quarterly basis.
NOK 246 stipulated as maximum fee for changing suppliers.
Network owners can collect up to NOK 4000 from each supplier for which regulating power is
settled in his network.
Hourly metering for electricity consumption in excess of 500MWh per year.
1996
Standard GS2 file format requirement.
1997 Fees are eliminated.
All end users can change suppliers on a weekly basis.
1998
Network owners must send settlement data by means of EDIEL.
Hourly metering for all installations larger than 400MWh.
1999
Messages regarding change of supplier must be sent by means of EDIEL
Mandatory continuous balance settlement.
2000
Reading of all end users at the end of the year.
Before the elimination of the hourly metering requirement hardly any household customers changed their
suppliers. In 1995 and 1996, which were the first two years after the elimination of the hourly metering
requirement, the movements on the market were very small. This is probably due in part to the fee for changing
suppliers of NOK 246 and the fact that the suppliers had to pay NOK 4000 to each and every network they sold
power in.
Another element is the fact that the power suppliers only made an active effort on the market to a limited extent,
and in many cases it was difficult for the households to find another supplier. Furthermore there was probably
little knowledge about the open power market. In the second half of 1996 the wholesale prices rose sharply due
to low inflow and low magazine levels. Primarily for reasons of local politics, most of the local suppliers did not
increase their prices to the households correspondingly. This resulted in low and in many cases negative margins
for the suppliers, and thus the household market was not of any interest to those who wanted to find new
customers.
As of 1997 the network owners’ opportunity to charge fees in connection with changing suppliers was
eliminated. This fact combined with more normal power prices started to move the household market. At the end
of 1997 almost 35 thousand households had a supplier other than the local supplier. This development has been
reinforced in 1998, as it is possible to change suppliers at the start of every week as opposed to the start of every
quarter earlier. According to the Norwegian Water Resources and Energy Administration’s NVE’ ( s) survey over
90 thousand or 4.5% of the households had a supplier other than the local supplier as of 4 October 1998.

1.4. Political and regulatory background for the current Metering, Load Profiles and Settlement System
The network owner's obligation to make provisions so that the end users could change their supplier was
established in the Energy Act of June 1990 and the Regulations of December 1990 pursuant to this Act. Section
4-4b of the regulations states: "The concessionaire is obligated to make unused transmission capacity in the
network system available to others engaged in the supply of electricity, as well as the producers and users of
electrical energy. This obligation also applies when the users of electrical energy buy power on the power market
or from other suppliers." Since it is not permitted to operate a network or sell energy without a concession, the
entire Norwegian electricity supply market is thus subject to this rule.
Section 4-4b of the regulations also states: "The concessionaire must not discriminate against any users of the
network, and he must offer them the same tariffs, adjusted according to differences in the period of use, quality

36
of delivery, etc." This rule must be interpreted to mean that a network owner must treat all power suppliers
equally. This entails that the local supplier,which was in any case integrated with the same company as the local
network owner in 1991, cannot be given any special rights in relation to the other suppliers. In 1991 no model or
regulations had been established that would allow compliance with the rule stipulating equal treatment of
suppliers.
In 1991 each individual supplier was responsible for submitting hourly settlement data to the power pool on a
weekly basis, i.e. all power inputted into the network from production or consumed through sales to end users. In
a situation where the trading and network were both part of the same company and this company was responsible
for power sales within the network, it was appropriate that the supplier itself submitted the settlement data to
Statnett. It was also easy to obtain the exchange data, as all that was necessary was to meter the exchange with
adjacent networks.
In the very beginning of 1991 the first change of supplier took place, whereby an end user changed over from his
local integrated power company to another supplier. It was the suppliers who had to cover the cost of providing
the metering data, normally through a bill he received from the network owner, as the network owner was
permitted to do in accordance with the applicable regulations. The dominant local supplier, i.e. the integrated
company, avoided these costs, because only new or so-called "foreign" suppliers had to meter the end user's
consumption hourly. In order to determine his exchange volume, the local supplier used the network owner's
exchange with other networks, and then he deducted the hourly values from the other suppliers in the network.
The local supplier demanded by virtue of his integration with the network owner, access to the hourly values
from the other suppliers.
This was clearly unfair as regards competition. As a result, only the largest end users found it profitable to
change suppliers because of the significant costs associated with having a supplier other than the local supplier.
The Norwegian Water Resources and Energy Administration (NVE) regulated the amount the network owner
could demand from the suppliers to cover the cost of hourly metering. An important goal of the NVE efforts to
improve real market access was to discontinue the practice of forcing anyone who had a supplier other than the
local power company to install meters that measured their consumption hourly.
As a result the new guidelines for the metering and settlement of trading in electricity issued by NVE entered
into force as of 1 January 1995. An important change was that the actual system load profile was to be used and
not the historical profile for the settlement of end users that were not metered hourly.

2. Tariffs
Pricing of power transport is based on the following elements:
• General terms – Each user of the grid pays for power from the grid, and each generator pays a transmission
fee for feeding power into the system. Each user of the grid pays only to the entity this user is linked directly
to. The fee includes the costs of operating the grid at higher levels (paid by the distribution owner to the
regional owner).
• The fee for feeding power into the grid has a fixed (for all grid levels) capacity component and a variable
energy component. The energy component reflects marginal grid losses, which can be considerable when
close to grid capacity limits. The value of these losses is equal to the power price in the spot-market, i.e., at
the time of occurrence of the loss. The loss varies depending on voltage level, the point at which the
generator feeds into the grid, and the time of the day and year. For example, a generator can have a negative
energy component if its location reduces losses instead of increasing them. Further, the energy component is
large in areas of surplus production. The energy component is lowest atnight-time and during summer when
load usually is low.
• The fee for taking power out of the grid varies between distributors (due to the geographical and load
variations) by voltage level. The fee has three components: A fixed component which is the same for all
customers, a capacity component depending on the customer’s maximum demand (kW), and an energy
component depending on the energy consumption (kWh). Residential customers’ fee consists of an energy
component only.

3. Metering
An upper limit of 500 MWh was set in Norway for the annual energy consumption per metering point for
settlement based on adjusted system load profiles. The Norwegian Water Resources and Energy Administration

37
(NVE) has estimated that there are approximately 10000 metering points in Norway over the limit of 500MWh
which do not have any equipment for hourly metering at the start of 1995. The network owners were given until
1 January 1996 to phase in the installation of metering equipment. As of 1 January 1999 this limit has been
reduced to 400 MWh. The original intention was to reduce the limit of 400MWh all the way to 100 MWh, but
this did not offer any benefits on the basis of a cost/benefit analysis.

4. Load Profiles
4.1. Choice of the Load Profile Model used
In 1994 a proposal was put forth which entailed that the small end users would not have physical access to the
power market, but that they could hedge the price of the electricity delivered to them through financial contracts.
The network owner would on its part be obligated to deliver power at the spot market price to its network
customers. Thus all physical trading of power would take place on the spot market.
In the same year the Social and Commercial Research Foundation (SNF) carried out an analysis of this model for
NVE. Based on this analysis the NVE concluded that the spot market model would not give small end users real
market access, and that it could result in unfortunate market-driven adaptations with regard to the integrated
power companies, which would have been forced to play an unfortunate double role. Thus the NVE found that it
should attempt to develop another model in which all the end users would have direct access to the physical
power market.
The first stage in the development of a method that would eliminate the need for hourly metering involved the
use of standard curves for the consumption profile for the various customer categories. The standard curves
could be used to distribute an assumed consumption at the hourly level so that the suppliers could calculate the
hourly settlement volume in relation toStatnett for those end users that were not metered hourly. The assortment
of standard curves that could be used for customer categories could easily grow to become very large. The
standard curves would differ in accordance with the various types of dwellings and commercial buildings, and
the different types of commercial activities, and there would also be divisions for geographical areas. The NVE
drew the conclusion that the work required to establish a standard curve that was as accurate as possible for each
individual end user would easily become complicated and difficult to manage. The NVE also envisioned that
conflicts between the end user, supplier and network owner could arise with regard to what curve should be
used. In addition, every discrepancy between the standard curve and the actual consumption profile would
represent a loss risk to the network owner. The sum total of the discrepancies between the standard curve and the
actual consumption profile would represent a significant loss risk to the network owner.
The second stage involved the use of a curve that was unique for each individual network area instead of specific
curves for customer categories. This curve would be calculated on the basis of each individual network owner's
system load profile, i.e. hourly input into the network owner's power network. The system load profile for the
last known year would be used, meaning that the profile for 1994 would be used for 1995. To adapt the profile as
accurately as possible to the end users that were not metered hourly, the network owners would deduct the
hourly values for those end users that were metered hourly. The profile would be used for all end users that were
not metered hourly.
Since the system load profile was from the previous year and random temperature fluctuations could have made
an impact on that profile, it would be adjusted in relation to the standard temperature. The profile would
subsequently be adjusted in accordance with the actual temperature. The "actual" temperature would be based on
forecasts of the weekly mean temperature issued by the Norwegian Meteorological Institute.
For the network owner this method would also represent a certain risk if the system load profile based on the
previous year that was used differed from the current year, and the temperature adjustment did not completely
counterbalance the differences.

38
4.2. Description of the Load Profile Model used
The network owner's system load profile represents the hourly net input into the network owner's power
network. The adjusted system load profile is derived by taking the system load profile as a point of departure,
deducting the network loss, and then deducting the actual end users and producers with hourly settlement. The
adjusted system load profile thus represents the average consumption profile for those end users that are not
metered hourly. In this connection we would like to point out that the profile is not a so-called predefined profile
determined prior to consumption of the power. This is a profile that is derived on the basis of the actual hourly
power input.
The adjustment for network losses should be made based on empirical data. This means that the hourly loss
values shall be deducted from the hourly power input. The empirical data that many network owners have with
regard to losses is quite inadequate. Firstly it is uncertain how large the loss will be on an annual basis, due,
among other things, to the fact that the meter readings are scattered throughout the year. Secondly, there has
been little knowledge of how the loss is distributed throughout the year.
Both of these conditions are of significance if the total cost of losses to be charged to the transmission tariff is to
be calculated. The purpose of the method of settlement based on the adjusted system load profile is, however, not
to calculate losses with a high level of accuracy. An inaccurate estimate will only result in an incorrect
distribution of the loss throughout the year. The loss volume will be correct, because all the suppliers are settled
in the end according to the actual volume of energy supplied.
The estimates the network owner has for the annual loss combined with the hourly network load and network
configuration is an adequate basis for calculating the hourly loss.

5. Settlement
5.1. Description of the settlement method used
The discrepancy between consumption/input and purchase/sales commitments must be identified and a price
must be determined. In Norway, Statnett is responsible for settlement of this difference or discrepancy. Statnett
must be able to identify the hourly discrepancy for each individual supplier who trades power in the Norwegian
power system. The value of the discrepancy is determined by the regulation power market, which is managed by
Statnett, the organisation responsible for the Norwegian power system.
Power trading is carried out by participants who are called entities with balance responsibility in the regulations.
An entity with balance responsibility is responsible for the difference between its contractual purchase and sale
obligations and the metered consumption or input in the network areas he is operating in. The entity with balance
responsibility is therefore charged for regulating power based on the physically metered power.
An entity with balance responsibility that sells power toend users is called a supplier in the Norwegian rules and
regulations. An entity with balance responsibility that only buys and sells on the wholesale market is therefore
not a supplier pursuant to the rules and regulations. Both end users and network owners can themselves be
entities with balance responsibility if they desire to be charged directly for regulating power. These end users
will normally be large industrial concerns with a high consumption of energy that buy power on the wholesale
market themselves. Network owners with balance responsibility will be the entities that buy network losses
directly on the wholesale market.
The network owner meters all electricity consumption or power input associated with his power network. He
also prepares and submits the settlement data to the power pool (Statnett) on behalf of the individual entities with
balance responsibility who sell power in their concession area. Pursuant to the regulations the network owner
shall send settlement data to Statnett three days after the delivery week. This means that the network owner shall
send Statnett the 7 times 24 hourly values for the previous week on Wednesday for each individual entity with
balance responsibility that settles power in his network. The network owner shall not distinguish between the
hourly metered and calculated values in relation to Statnett, he shall send the accumulated settlement data for
each entity with balance responsibility. The network owner shall send the same data to the entities with balance
responsibility. The entities with balance responsibility can use this data as a basis for controlling the settlement
from Statnett. The network owner shall also send hourly values for the installations that the individual entities
with balance responsibility deliver power to. In addition, the network owner shall send the adjusted system load
profile share for the entities with balance responsibility. This is data that the entities with balance responsibility
use as a basis for invoicing their customers.
The entities with balance responsibility themselves report their power commitments to the power pool, but they
do not provide settlement data. The power commitments of an entity with balance responsibility may be

39
established through buying or selling on the spot electricity market or through bilateral contracts with another
entity with balance responsibility.
The power pool settles the accounts for the other entities with balance responsibility in a corresponding manner.
Statnett has nine days after the delivery week to carry out a settlement in relation to the entities with balance
responsibility by means of a debit or credit.
This method of settlement entails that any inaccuracies in the settlement of the entities with balance
responsibility are incorporated into the network owners' loss calculations. If the consumption of an entity with
balance responsibility is listed in the settlement at a value lower than the actual consumption, then the calculated
network loss will increase by the difference between the actual consumption and the calculated consumption.
This represents a cost to the network owner. To avoid that the network owner is charged excessive costs due to
the inaccurate settlement of the entities with balance responsibility, metered hourly values, or estimated hourly
values based on the network owner's adjusted system load profile shall be used in the settlement against the
power pool.
Continuous balance settlement builds on the method for settlement based on the adjusted system load profile.
The method entails a requirement to read all of the end users at the same time so that the balance settlement adds
up to zero from the perspective of the network owner. This method also entails that the suppliers bear a price risk
for any discrepancy between the expected consumption and actual consumption for new customers gained by the
supplier after the last balance settlement until the next balance settlement.
By distributing the adjusted system load profile according to the end user/meter level instead of the supplier level
the weaknesses mentioned above can be avoided. The volume for an hour is distributed among each individual
metering point based on the expected annual consumption. In order to obtain settlement data for an individual
supplier, the network owner must sort the distributed volume at the meter level by supplier. This means that the
network owner must have a link in his meter value database between the meter number and the supplier number.
The network owner can add up the settled volume at all times for the individual meter points. The network
owner needs this total when the meter point is read. Then he can compare the difference between the settled
volume and read volume for every meter point.
This volume is priced according to the spot electricity price weighted by the adjusted system load profile from
the first hour, which was the last reading, to the last hour of considered period. The value of the discrepancy is
added to the supplier’s debit and credit balance settlement account. This account is periodically settled with the
supplier. This method entails that the network owner updates the accounts for suppliers each time a meter
reading is taken. Pricing the discrepancy according to the spot electricity price weighted by the adjusted system
load profile between reading results in a correct and relevant pricing of the discrepancy. Thus the suppliers’ price
risk is eliminated for new customers between the periodic balance statements. The continuous balance settlement
method entails that settlement based on the adjusted system load profile does not place any limitations on when
the meters are to be read. Thus the meter points do not have to be read at the same time.
Under certain circumstances continuous balance settlement entails an increased risk for the network owner. This
is the case when the network owner reads some end users more often than others. Since there are price
differences throughout the year, the sum total of the priced discrepancies will not be zero, as is the case for the
volume discrepancies. If the meter points that are read more often than others have an actual consumption profile
that is unbiased and equal to the adjusted system load profile, then the sum total of the positive and negative
discrepancies will compensate for each other over time. If this is not the case, then the sum total of the suppliers’
balance settlement accounts will not equal zero over time.

5.2. Deviation between estimated and actual consumption for small eligible users
The percentage distribution of the consumption of the individual suppliers calculated by the network owner will
not be in agreement with the actual distribution. Settlement by means of the adjusted system load profile requires
therefore that the meters belonging to the supplier's customers be read periodically, for example once a year.
It is important to point out that this settlement is not analogous to Statnett’s regulating power settlement. This is
only a settlement between suppliers within an individual network. The discrepancies between the suppliers
within a network do not have any impact with regard to the power system because the suppliers will on the
whole purchase power for the network in question in accordance with the expected consumption for non-hourly
metered end users. Therefore it is not relevant to price the discrepancies at the regulating power price, because
the regulating power price shall be an expression of the system related costs of upward and downward
adjustments due to underconsumption or overconsumption after a market has been cleared.

40
The difference between the calculated distribution and the measured distribution shall be settled between the
network owner and the individual suppliers at the weighted spot electricity price. The spot electricity price shall
be weighted in relation to the adjusted system load profile during the calendar period. The reason why the spot
electricity price is used is because it best represents the suppliers’ alternative price if the end users had been
metered hourly.
The meter readings shall be taken at the same time for all the end users in the network. This will primarily be
performed by self-reading. Self-reading systems will often be the most cost-effective systems.
Another alternative would be to read each individual supplier separately.One supplier in January, one in
February, etc. If this is based on manual reading by the network owner himself, then it may not be possible to
take the readings in an efficient manner, as the supplier's customers may be geographically scattered.
A third alternative can be an efficient method in cases where one supplier is dominant in a network. In this
alternative the customers of the small suppliers are read together, for example in the course of January, while the
customers of the dominant supplier are read scattered throughout the rest of the year. This is possible under the
assumption that the dominant supplier accepts such a solution. The settlement for the small suppliers will be
accurate, but the settlement in relation to the dominant supplier will be less accurate.

6. Other questions resulting from choice of system

6.1. Time restraints and other conditions for changing supplier
Every time when an end user changes his supplier the network owner had to perform the calculations of
percentage distribution of the consumption between the suppliers inside the network. To avoid the network
owners having to perform these calculations on a continuos basis, restrictions can be placed on when a change of
supplier can take place. Up to 31 December 1997 end users that were not metered hourly could change their
suppliers at the beginning of every quarter.
Such a limitation is a clear weakness because an end user can risk being locked into his supplier for up to 3
months. This means that the end user does not have any real opportunity to change suppliers during a quarter
even if the power prices fall. The rules were changed as of 1998, and end users now have the right to change
suppliers every Monday. Theoretically an end user can now change his supplier every week.
Exchange of information for change of supplier between the network owner, new supplier and old supplier is
made through electronic message types (EDIEL - Electronic Data Interchange in the ELectricity industry) that
are defined in the regulations. The new supplier sends a message to the network owner containing the name,
address, meter number etc. no later than three weeks before the change is to take effect. One week before the
change the network owner sends a supplier change confirmation to both the new and old supplier. The new
supplier receives in this connection more detailed information on the customer such as the expected annual
consumption. The network owner has three weeks after the change date to send meter data at the time of the
change to the new and old suppliers.

6.2. Costs resulting from changing supplier (who pays)

In 1995 and 1996 there were fees for changing suppliers and the network owner's handling of suppliers. These
fees were to be set so that there was not any discrimination against any of the network's users. The purpose of
these fees was to make the end users and suppliers aware of what costs they would incur in connection with
changing suppliers. In this manner the end users and suppliers could adapt their behaviour on the market in
accordance with the costs associated with their behaviour.
The network owner was permitted to charge a fee each time an end user changed his supplier. The idea behind
the size of the fee was that the costs incurred by the network owners were to be defrayed, and the fee was at the
same time not to be so high that it prevented small end users from changing their supplier.
In addition, the network owner could charge a fee for each individual supplier that sold power in his concession
area. The fee was supposed to defray the costs incurred by the network owners in connection with handling the
suppliers and to make it more favourable for a supplier to have many customers in a few networks, rather than
having its customers scattered across many networks.
As of 1997 the Norwegian Water Resources and Energy Administration (NVE) eliminated the network owners'
opportunity to charge fees or to receive any other payment for services in connection with the metering and
settlement of trading in electricity from suppliers delivering power in their own network. The reason was the fact
that the network owners had had two years of experience from changing suppliers, and it should be easier for

41
them to handle a larger number of suppliers and supplier changes now. Another element is that even small fees
will reduce the willingness of the end users to change suppliers, and the fee would have the effect of locking the
customer into the supplier. As a result, the suppliers would be able to have higher margins without any risk of
losing customers.
In general one can regard the costs a network owner incurs from handling supplier changes as part of their duty
as a monopolist. As regards the other costs incurred by the network owner, these are covered through the
network user fee. It is therefore not unnatural that the costs associated with handling suppliers are covered in a
corresponding manner.

6.3. Costs of metering – included in the tariff?

End users with a metering point that have an annual power consumption of over 400MWh must be metered
hourly. Network owners are responsible for ensuring that measurements are taken and that they are correct.
This does not entail that it is the network owner himself who must perform all the measurements or own the
meters. The network owner is free to farm the meter job out to others or to let others install and own the meters.
The costs associated with such equipment shall be defrayed by the network owner and included in the calculation
basis for the transmission tariff for the relevant network level.
The end user can demand that a metering point with an expected annual energy consumption of less than 400
MWh shall be settled based on hourly metered values. The end user must thenhimself defray the additional costs
associated with the hourly metering. Network owners are also responsible for the hourly values for hourly
metering below the limit of 400 MWh per year.
Some network owners may themselves desire to hourly meter certain end users or to reduce the general limit for
hourly metered consumption. The reason for this may for example be the fact that the network owner desires to
meter transmission for certain customers hourly. The network owner must in such cases himself defray the costs
associated with the hourly metering.

42
United Kingdom
1. Background information

1.1. Electricity sector in United Kingdom

The Structure of the Industry
The UK electricity system in fact comprises three separate and differently-organised electricity markets. All
meet the fundamental principles of the Electricity Directive.
1. England and Wales comprises a regulated TPA system based on a Pool system of establishing prices
using an economic merit order of offers for generation;
2. Scotland again comprises a regulated TPA system, with two separate but fully vertically-integrated
electricity companies which also supply to the England and Wales Pool via a singleinterconnector;
3. Northern Ireland consists of a small electricity system generating only 7-8TWh, with a monopoly
Power Purchaser at the wholesale level. Any licensed supplier can sell electricity to any final customer.
England and Wales
The England and Wales system, based on the operation of an electricity Pool through which electricity is traded,
qualifies as a “regulated system of access procedure” within the meaning of Article 17.4. Generation capacity is
subject to an authorisation procedure, with the Department of Trade and Industry as the authorisation body for
plant above 50 MW, and the local planning authorities for plant of 50 MW and below.
Any generator exporting more than 50 MW onto the system is required to hold a generation licence and therefore
will be obliged to trade the output via the Pool. Offers to generate are received one day in advance from
competing generating companies at each individual power station unit, and prices are set on a half-hourly basis.
Bids from generators in Scotland and France for electricity to be transmitted across the interconnectors with
these systems are dispatched in the same way as bids from generators in England and Wales.
The National Grid Company is the Transmission System Operator responsible for dispatching, following a
transparent economic merit order. Dispatching criteria are published in the Grid Code and transmission licence.
The National Grid Company also owns and manages the high voltage transmission system in England and
Wales. Transmission tariffs are published and are regulated by the Office of Electricity & Gas Regulation
(Ofgem) according to a price control formula based on RPI - X. Services such as frequency response, reserve
and reactive power are purchased through contracts with generators so as to ensure that system stability is
maintained while ensuring transparency, encouraging competition and reducing costs.
All generators and distributors seeking connection to the transmission system must meet the appropriate
standards to ensure that technical difficulties are not caused for others connected to the system. Technical rules
for access to the transmission system are published in the Grid Code, which covers items such as planning
requirements, connection conditions, demand forecasting, scheduling and despatch, operational liaison, safety
co-ordination, and testing requirements. The terms for access are non-discriminatory, and access may only be
refused in certain clearly-defined circumstances, e.g. for safety reasons. The settlement of disputes and
application of the competition rules are the responsibility of Ofgem and, in specific cases, the Monopolies and
Mergers Commission, Office of Fair Trading and the DTI.
Generally speaking, any company which supplies electricity to premises (other than small suppliers up to 500
kW) must hold a supply licence, and only holders of a supply licence may purchase electricity from the Pool.
There are two types of supply licence. Firstly, each of the Regional Electricity Companies holds a licence giving
it rights and obligations relating to supplies to franchise customers within its authorised area. These licences are
called Public Electricity Supply (PES) Licences, and set out a number of public service and other obligations.
These can be classified as follows:
- Obligation to supply and to ensure continuity and quality of supply;
- Regulated prices for customers in their area; non-discrimination and prohibition of cross-
subsidies;

43
 - Standards of performance on service; provision of services for the elderly and disabled;
establishment of a customer complaints procedure and of a code of practice on the payment of
bills;
- Standards of performance on the efficient use of electricity.
There are no restrictions on electricity trade on the basis of these Public Service Obligations.
There is a requirement on PESs to purchase a specified amount of renewable energy. A Non-Fossil-Fuel Levy on
electricity consumption is designed to fund the cost burden arising from this renewable take-off obligation. The
Levy system itself incorporates a market mechanism in which renewable generators are required to bid for
subsidy payments, thus encouraging generators to reduce their costs.
The other type of licence is called a Second Tier Licence. Regional Electricity Companies holding such a licence
can supply eligible customers outside their region, and the licence also allows generators or other undertakings to
supply customers direct. Second tier licences are held by all the Regional Electricity Companies, the main
generators in England and Wales, the Scottish companies and some new suppliers. All of these licence holders
are competing with each other to supply eligible customers, with the result that prices have been subjected to
significant downward pressure.
Competition in electricity supply was introduced gradually, with initial opening of the market to distributors and
final customer sites with a demand of over 1 MW as of April 1990. The threshold was lowered further to 100
kW in April 1994, and by June 1999, all 26million electricity customers in the UK will be able to shop around
for their electricity supply.
Public Electricity Suppliers also act as distributors of electricity on a non-discriminatory and transparentbasis
which includes the publication of separate accounts for distribution and supply. No cross-subsidy is permitted,
with the distribution business required to offer the same services and prices to all suppliers, including thePES’s
own supply business. As a monopoly activity, the charges made for the use of the distribution network are
regulated by an RPI -X formula.
The framework which exists in England and Wales is designed to facilitate competition in the generation and
supply of electricity. The framework as a whole is subject to regulation.
Scotland
The basic features of the electricity market in Scotland are as follows:
- Both the existing generators and new entrants may construct and operate generating stations in
Scotland and throughout the UK;
- Suppliers based anywhere in the UK may supply customers in Scotland;
- Trade between companies within Scotland;
- Participation in the England and Wales electricity Pool;
- Open access to transmission and distribution systems;
- Regulation by OFGEM
The licence granted to Scottish Power and Hydro-Electric allows each company to construct and operate
generating stations anywhere in Great Britain and to transmit, distribute and supply electricity in its authorised
area to tariff customers. Although Scottish Power and Hydro-Electric operate as vertically integrated electricity
companies, under the regulatory regime their generation, transmission, distribution and supply activities are each
treated as separate businesses and each is regulated separately. The licence obliges each company to publish
separate accounts for each of its businesses, to ensure there is no cross-subsidy and that excessive profit is not
made from use-of-system charges.
Generation prices are regulated by reference to prices on the England and Wales market.
Competition in supply is made possible as a result of third party access to the transmission and distribution
systems on a non-discriminatory basis, both in relation to granting use of the system and in relation to the
charges applied. Use-of-system charges are published annually, and the companies are obliged to charge their
own separate supply business and customers on the same basis. While third party access is available to anyone
who wishes to use the system, they must abide by the grid code and/or distribution code. These codes cover all
material technical conditions for connection to, operation and use of the respective systems. If the third party
does not agree to be bound by the codes, a company can refuse access to its systems.

44
Northern Ireland
In Northern Ireland, there are four privately-owned power stations, and Northern Ireland Electricity acts as
Independent System Operator. It is also has responsibility for both the transmission and distribution systems and
for the power procurement function - the purchase of electricity from the generating companies. A separate
supply business has been established within the company, and a framework for competition in supply has been
introduced. As in the rest of the UK, strict provisions are in operation to ensure the unbundling ofNIE's core
businesses.
Responsibility for the regulation of the industry is in the hands of OFREG - the Office for the Regulation of
Electricity & Gas - which oversees the development of competition and protects the interests of customers in
Northern Ireland.

1.2. Electricity Legislation

The 1989 Electricity Act laid the legislative foundations for the restructuring and privatisation of the industry.
The Act resulted in:
- a change of ownership from the state to private investors;
- the transfer of employees to the successor companies;
- the introduction of competitive markets;
- a system of independent regulation.

1.3. Degree of market opening and future thresholds

The introduction of competition in the supply market had to be phased in over a number of years because of the
sheer size of the task in terms of the number of customers and the technical complexities involved.
At the time of privatisation 5,000 of the largest industrial sites, those with a maximum demand of 1 MW, were
given the ability to choose their supplier. In 1994 this market was extended to cover medium-sized customers,
such as hotels, supermarkets and cinemas, when the threshold was lowered to 100 kW.
Before privatisation, it had sometimes been argued that only the very largest electricity consumers would be
interested in choosing their supplier. However, the reality has been very different, and many large and medium-
sized companies have been keen to participate in the competitive market. Offer had estimated that in 1997/98 the
competitive electricity supply market in Great Britain covered more than 70,000 sites and accounted for almost
146 TWh (about 50%) of electricity sold in Great Britain. Two fifths (40%) of these sites were receiving
electricity from suppliers other than their local REC, that is from second tier suppliers, accounting for 62% of
electricity supplied to this market.
Large customers have shown limited loyalty to their local suppliers, and profit margins have been kept low in an
increasingly competitive market. Because one kWh of electricity is essentially the same as another, many large
customers buy it as a commodity, with price the major differentiating factor. It has so far proved difficult to sell
value-added services, though energy efficiency advice has been a factor which has influenced the decisions of
some customers.
Britain completed the world’s first fully competitive energy market in May 1999 when the staged introduction of
competitive supply into the domestic sector came to an end. From then on, all 26 million customers could shop
around for the electricity and gas suppliers offering them the best deal.
With the number of potential 'switchers' increasing from tens of thousands to tens of millions, it was considered
unwise to introduce competition for small electricity customers in a single “big bang” because of the strain this
might impose on the new technological infrastructure. Competition was therefore phased in over nine months, in
stages based on regional post codes.
At the time of the final roll-out more than 1.5 million electricity users had already switched supplier, along with
4.5 million gas consumers for whom the open market started sooner. At the peak of early activity, one leading
UK supplier was handling 35,000 changes - losing existing customers and gaining new ones - every week. Each
one of these changes involved 15 or 20 separate actions by company staff. It has been suggested that in the
longer term, annual turnover among customers switching and then switching again could run at more than ten
per cent.

45
1.4. Political and regulatory background for the current Metering, Load Profiles and Settlement System
Currently, any generator exporting more than 50 MW on to the system is required to hold a generation licence
and to trade its output via an open commodity market, the Electricity Pool. The Pool is a simple name for what
is, in effect, a very complex trading mechanism. Set up in about a year, it has seen continuous evolution and
development. Each day prices are set for active energy supplies for each half hour of the following day
dependent upon bids made into the Pool by potential producers (generators). These prices with some subsequent
adjustments then form the basis of payments for bulk supplies to generators and by the Public Electricity
Suppliers (PESs) supplying end customers.
There have been specific legal metrology provisions for electricity meters in the UK since the 1890s. Current
provision are set out in the Electricity Act 1989. This requires that meters usedare Approved “..as to pattern and
construction..” and, in the case of domestic customers, are Certified (i.e. calibrated to specified degrees of
accuracy). To administer and police conformance to the requirements the Act established the Electricity Meter
Examining Service (EMES), the major functions of which are to oversee the approval, testing and calibration of
meters and to arbitrate in disputes as to meter accuracy between customers. It has powers to ensure that adequate
secondary legislation is in place to support these functions.

2. Tariffs
2.1. Transmission Charges
Transmission charges are binomial (i.e. two-part). The first part is an annual connection charge, payable over
twelve months, which is designed to recover the costs incurred by NGC in providing connection assets and
maintenance of those assets. These are calculated on an individual basis and are not published, although the
methodology used in the calculations is consistent and published.
The second part is the use of system charge. This will vary depending on whether inputting to or taking from
National Grid Company’s transmission system and also on the geographical location of connection.

2.2. Distribution Charges

Distribution charges are published by each distribution business. These often mirror the tariff on which the
customer is being supplied although it is becoming popular that a standard two-part Distribution Use of System
tariff may be applied, irrespective of the final supply tariff’s structure. This is particularly the case for domestic
and small commercial distribution tariffs.
The principles behind the design of tariffs remain largely unchanged in competitive markets. Tariffs continue to
be based on the provision of assets, a return on those assets, maintenance and renewal, corporate overheads in
providing services, and depreciation.

2.3. Connection fees

The standard tariff for domestic customers consists of a fixed standing charge and typically a single kWh rate
applied to all units consumed. The fixed charge is designed to recover the customer-related costs incurred in
providing the supply. These include meter reading and billing, metering equipment and the cost of the service
connection to the main.
As an alternative to the standard tariff, customers may elect to be supplied on a day/night tariff. These tariffs
have different day and night kWh rates (the night period normally lasts 7 hours in England, Wales and Northern
Ireland but 8 or 8 ½ hours in Scotland). There are also several tariffs with weekend, evening and seasonal
features; also, new tariffs are becoming available which may also help support either the environment or other
causes.
3. Metering
3.1. Choice of Threshold Level established for metering
Prior to 1998 the threshold level there was a Pool requirement in the competitive supply market to install half-
hourly registering metering for any customer having a demand of 100 kW and above. Data from such meters was
(and still is) collected by remote means on a daily basis.

46
Since 1998 below this level it is for the supplier to decide whether to install half-hourly metering and, further,
whether the associated data should be collected remotely on a daily basis. If so, it is unlikely that such provisions
will be economically justifiable for customers below a 70 kW demand threshold. If not, trading will be on
profiled (estimated) consumption with the option to use the existing meter or to install a new meter which
records half-hourly data for a period of 3-12 months (as specified by the purchaser) but which does not have
remote communications features.

3.2. Description of Metering System used

The metering systems used are described in Codes of Practice, issued by the Electricity Pool of England and
Wales, relating to metering for different levels of customer demand. Code 1 covers above 100 MVA; Code 2
covers 10-100 MVA; Code 3 covers 1-10 MVA; Code 5 covers up to 1 MW (above 100 kW); Code 6 covers LV
supplies of 100 amps or less per phase.

4. Load Profiles
4.1. Choice of Load Profile Model∗
The decision on load profiling strategy for the below 100 kW sector was made by the Electricity Pool, following
consultations with Pool members, and taking into account the results of simulation studies on the proposed
energy settlements system. Historic data have consistently shown significant differences in the load
characteristics of specific customer categories. This factor, together with the expectation that most suppliers will
have a wide range of customers in varying proportions, and that half-hourly Pool (i.e. wholesale) prices will
continue to show substantial daily and seasonal differences, led to the preference for load profiles based on
clearly defined customer categories.
Eight half-hourly profiles are estimated by a contractor to the Electricity Pool, using data acquired from national
load research customer samples; two for domestic customers, and six for non-domestic customers. The load
profiles are updated annually. One of the domestic profiles covers customers on the two rate Economy 7 tariff.
This profile is further broken down into “switched” and “base” load components, where “switched” refers to
electricity (primarily for storage space and water heating) that is controlled by a time switch orteleswitch. An
algorithm is then applied to the switched component so that a modified profile can be created for each of the
large number of time regimes typical of Economy 7 tariffs. A similar approach is followed with the non-
domestic Economy 7 profile.
In order to allow for the influence of regional variations in temperature and daylight hours, the national demands
are converted into statistical relationships with temperature and time of sunset using linear regression techniques.
These regression relationships are calculated for separate seasons and day types, for all half-hours.

4.2. Description of How the Load Profile Model Is Used*

In January of each year a set of load profiles - expressed in the form of regression equations - are provided to the
Electricity Pool, together with estimates of annual average consumption for each profile. On every settlement
day, commencing 1 April, regional temperatures and sunset times are input into these equations. The resultant
profiles estimate for each region (12 in total) the proportion of annual consumption taken during every half-hour
of each settlement day, by an average customer. This stage in the process thus gives a regional interpretation of
nationally based profiles.
The half-hourly proportions are then multipliedby an estimate of the aggregated annual consumption of all of a
supplier’s customers on a particular profile. The annual consumption estimates are derived from tariff meter
readings for customers who are not on half-hourly meters. The result is an estimate of a supplier’s half-hourly
consumption by profile, which can then be summed across the eight profiles to give the supplier’s half-hourly
consumption for the total of all “profiled” customers. This total is known as the supplier’s “deemed take”.

∗
This description refers to the arrangements followed by the Electricity Pool, responsible for the rules and
procedures of the wholesale electricity market in England & Wales. A similar approach applies in Scotland,
where the equivalent body is Scottish Electricity Settlements Ltd.

47
5. Settlement
5.1. Choice of Settlement Method Used*
Companies wishing to generate and supply electricity are required to acquire a license, or to satisfy criteria for
exemption. Most electricity generated and sold has to be traded through the Pool, the companies concerned being
required to sign the Pooling and Settlement Agreement. This agreement governs the operation of the Pool, its
trading arrangements, and the obligations of participants.
The settlement system incorporates both half-hourly metering and load profiling, and is intended to ensure that
there is a reasonable balance between risk and cost, with the risks associated with profiling shared equitably.
Metering information from all customers, suitably aggregated, is fed into the settlements process.

5.2. Settlement Method Used*

Payments to generators are required to match exactly payments from suppliers, so ensuring that the Pool clears
on a daily basis. The payments themselves are made approximately 28 days after the settlement day, i.e. the day
when the electricity was generated and consumed. “Initial settlement” occurs at this stage based on meter data
then available, followed by four additional “reconciliation runs” at intervals over a period concluding at 14
months after settlement day. The reconciliation procedure recalculates the amount each individual supplier owes.
This timetable is designed to allow the efficient capture of data from customers on different meter reading
patterns - ranging from customers on maximum demand meters that are read monthly, to domestic customers
whose meters are read annually.
The settlements process utilises data from both half-hourly and non half-hourly meters. The former are
aggregated for each supplier in each region, and adjusted for line losses. Multiplication by the half-hourly Pool
Selling Price gives each supplier’s purchase costs in respect of its half-hourly customers.
In any one region, half-hourly metered data are also summed across all suppliers. This half-hourly amount is
then deducted from the region’s total metered input from the national network, and the resultant figure
compared with the total half-hourly “deemed take” (derived as described above), summed across all suppliers
and adjusted for line losses. The ratio of these two half-hourly quantities indicates the error arising primarily
from the profiling process.
The error, or “correction factor”, is applied to each supplier’s half-hourly “deemed take”. This mechanism thus
ensures, in the interests of equity, that all suppliers bear some proportion of the correction factor, and that the
total of all suppliers’ half-hourly consumption is equivalent to theknown metered input for the region. Finally,
multiplication of a supplier’s adjusted deemed take by the Pool Selling Price gives the supplier’s half-hourly
purchase costs for customers without half-hourly meters.

5.3. Deviation Between Estimated and Actual Consumption for Small Eligible Users*
As the outline in 4.2 and 5.2 indicates, aggregated consumption covering all customers without half-hourly
metering is a fundamental part of the settlement process. The source for consumption data is recent tariff meter
readings adjusted to an annual basis, or provisional estimates based on historic data in the absence of recent
readings. Consequently the consumption data increase in accuracy, as more meter readings become available.
Any errors associated with provisional estimates of annual consumption are incorporated in the regional
correction factor. This error tends to decrease with each reconciliation run.

5.4. Supplier Responsibility for Metering Consumption, Change of Supplier, Settlement Balance*
Settlement balance is carried out by the Electricity Pool, as already described. Supplier responsibilities regarding
the registration of metering systems and customer change of supplier (see 6.1 below) are governed by a 'Master
Registration Agreement', which suppliers are required to sign. The cost of most of the services provided under
the Agreement form part of the charges that distribution businesses are able to recover from suppliers.
Each customer's metering system is formally registered, ensuring that the meter is assigned to the correct
supplier. For customers without half-hourly metering, the registration process involves allocation of the
appropriate profile. Agents acting for suppliers carry out the tasks of meter operation, data collection, and data
aggregation.

48
6. Other questions resulting from choice of system
6.1. Time constraint and other conditions for changing supplier
Once a customer has signed a contract with a new supplier, the change to the new supplier must become
effective within 28 days of the contract date. The new supplier is then required to register that a customer has
switched suppliers within 28 days of taking over liability for supplying the customer. The supplier change
procedures also set down the requirements for conducting, agreeing and validating a meter reading, and for
registering a new data collector, if appointed.

6.2. Costs resulting from changing supplier (who pays)

There is no cost to the customer in changing supplier. Each Public Electricity Supplier (PES) has a license
obligation to administer a Master Registration Agreement whereby it registers details relating to each customer
in its area including who is the supplier. When a ‘new’ supplier signs a contract with a customer, he registers this
with the PES MRA which then contacts the ‘old’ supplier, who has the right to block the change in certain cases
(e.g. excessive unpaid debt). In the absence of any objection the transfer is confirmed. The old and new suppliers
then exchange specified data relating to the customer (e.g. final meter reading).
In theory there is no limit to the number of changes of supplier a customer may make or the timing. However, in
practice suppliers are unlikely to offer contracts allowing exit of less than a month and most will seek to tie the
customer into an annual deal – obviously the prices offered will reflect the ‘security’ of the contact accepted.

6.3. Cost of metering - included in the tariff?

Prior to 1998 the customer appointed a Meter Operator (remembering these are customers of 100 kW and above)
and paid or metering through the contract with the Meter Operator on 3 levels:
- purchase or hire of the meter
- charges for meter operation services
- Pool charges for data collection
These contracts will continue for their lifetime (being eventually replaced as below).
Since 1998 the supplier is responsible for appointing agents for meter operation, data collection and data
aggregation, and he will pay for metering services through contracts with each agent, passing such costs on to
the customer as a ‘package’ within the supply contract. A breakdown of the bill may be available to the customer
so that he can identify the amount of such costs (but this is a contractual matter).

49
Sweden
1. Background information
1.1. Electricity sector in Sweden
There is a broad ownership structure on the Swedish electricity market, with state owned, municipally owned
and privately owned electricity production and electricity distribution utilities. There are about 300 producers of
electricity, many of them small. The eight largest power companies account for about 90% of the total electricity
production. The state owned company Vattenfall AB owns around 50% of the electricity producing capacity.
The transmission and distribution network is usually classified in three levels. The national grid covers the whole
country and transmits electricity at voltage levels from 220 kV to 400 kV. The next level (sub- transmission) is
composed of the lines which connect the national grid with the larger power users. They are owned and operated
by companies which in most cases are subsidiaries to thepower companies normally responsible for wholesale
supplies of power within the area. Finally, thelocal networks are owned by the distribution companies.

1.2. Electricity Legislation

In October 1995 the Swedish Parliament decided to revise the Electricity Act of 1902 and introduce a completely
open electricity market. The new Swedish regulations came into force on 1 January 1996. The Act of 1902 was
in its entirety replaced by a new electricity act on 1 January 1998.
A main principle behind the reform of the electricity market is to achieve a clear separation between the
production and sale of electricity on the one hand and the transmission and distribution on the other. The
production and sale of electricity should be subject to competition, while network operations should continue to
be specifically regulated and monitored to ensure that they are efficient.

1.3. Political and regulatory background for the current Metering, Load Profiles and Settlement System
Generally, the goal of the Swedish proposal is to “exempt customers with low electricity consumption from the
fundamental principle of hourly metering of electricity consumption”. The former demand of having hourly
metering to change supplier has been a considerable hindrance when consumers have considered changing
supplier, and therefore has slowed down the liberalisation process and lessened competition for small consumers.
An example of the more detailed considerations is the Swedish proposal that the responsibility of having a
balance settlement at each consumption point becomes a supplier responsibility rather than a consumer
responsibility.
Further, the obligation to supply consumers has been made valid for all suppliers, except when another supplier
takes over or the consumer has not paid the bills.
Rules for the costs of installing an hourly meter for suppliers below the threshold between load profiles and
hourly metering have also been settled. The Swedish Government proposes that the maximum price level applied
so far be abolished.

2. Tariffs
The electricity Act states that holders of network concessions should be obliged to connect the lines and plants of
others to their own network on reasonable terms and to transmit electricity on reasonable terms regardless of
who the owner of the power is. This also applies to the price charged for transmission services, i.e. the network
tariff. Network tariffs should be reasonable and based on objective calculations. They should be unbiased and
non-discriminating. The terms of connection may be object of examination by the Network Authority in each
individual case. The assessment of the authority should be based on the interests of the customers. The tariffs
should also lead to revenues which provide reasonable returns.
From the electricity act also follows:
- The tariff for a consumer must be independent of from which supplier the consumer buys electricity.
- The tariff for a producer must be independent of to which consumers the electricity is sold.
- Network tariffs for individual consumers must be independent of distance and theconsumers location in the
network.

50
- There may be different tariffs for different consumer groups, but each individual in a group must have the
same tariff.
- The tariffs shall as far as possible reflect the costs and the physical flow in the network so that costs in the
network are paid by those customers who causes the costs.
- Network companies set their own tariffs which afterwards can be revised by the regulator.
Therefore network tariffs for consumers must be point tariffs and tariffs for producers may be either point tariffs
or physical path tariffs.

3. Metering
The electricity Act in Sweden gives third party access for all customer categories, but requires hourly metering
equipment to participate. In the Swedish debate the fairness is questioned over the fact that all other customers in
a grid region will be paying the difference between the maximum price permitted and the actual cost for new
hourly metering, since this cost can be added to the average grid fee. As a response to thisVattenfall, the largest
Swedish supplier, has introduced a new meter which can measure electricity hourly and has the ability of two
way communication solutions preparing for future communication products. Vattenfall only gives the new
meters to those who agree not to swap to a new supplier within a given period of time.

4. Load Profiles
The Swedish Government proposal has carefully considered both the area model applied in Norway, as well as
the category-model applied in Finland. It has decided to choose the area model.
The area model gives only one consumer profile per network area. It is the network owner that is responsible for
the basis for the calculation needed to establish the consumption profile and the division of the consumption
between the different suppliers. At each change of supplier of a consumer within the area, the proportions for the
division must be re-calculated. The consumption profile will be the same for all consumers within a given area,
but varies from area to area.
The category model uses curves for different types of consumers, showing the hourly consumption for different
types of consumers at different temperature levels. (Most individual houses in Finland, Norway and Sweden
have electric heating). Currently, 4 types of profiles are used in Finland. Only consumers having changed
supplier will use the load-curves. Nationally applicable load-curve-profiles must be made available for all
suppliers.
For both models, an adjustment of the cost follows when the actual metering has taken place.
The Swedish Government supports the area model, since it gives competition neutrality between different
players in the market, i.e. large and small suppliers, who both have much administration with category load
profiles. The area model is also simpler to administer and requires less IT-support than the category model. It is
underlined that the area-model is most cost-effective where the threshold for hourly metering is relatively high.
The choice of threshold and model are therefore connected.
Accordingly, the Swedish Power Grid Company and the National Energy Authority pointed out that if the
threshold becomes too high, the profiles could become questionable in areas where there are many consumers
with a high consumption. The future costs of hourly metering must also be taken into consideration when the
threshold is chosen, so that i.e. low costs should result in a lower threshold. It must therefore also be possible to
adjust the threshold at later stages, in case i.e. the cost of hourly metering falls significantly.

5. Settlement
There is broad agreement in Sweden that the responsibility for metering related consumption and change of
supplier must lie with the network operator. The system operator is responsible for the balance settlement, for
hourly metered as well as load-curve consumers, and for preliminary as well as final calculation.

51
6. Other questions resulting from choice of system
6.1. Time restraints and other conditions for changing supplier
The Swedish Government proposes a “one month in advance” deadline for change of supplier, instead of the
current “six months in advance” deadline. Some industrial, consumer and house-owner lobbies would
recommend short deadlines, some recommend a specific day as the deadline for each month. The network
companies are worried about short deadlines. Some worries are also voiced about changes of supplier around the
1st of January 2000, and the Swedish Government has limited the possibility of changes for that period. To avoid
problems in the beginning, especially related to IT, a slow move from the six-months to the one-month deadline
is proposed.

6.2. Costs resulting from changing supplier (who pays)

The Swedish Government proposes that no costs can be charged to the consumer at the change of supplier, even
if this may result in some costs. Some associations and Swedish Power Grid propose that after the second or
third change, it could be possible to charge something, but the Government recommends that there should be no
costs, to further competition. Only very few customers will change supplier often and the costs of this will be
relatively small on the whole, and should therefore be taken by thenetwork company.

6.3. Costs of metering – included in the tariff?

The costs of metering for consumers using the type of metering they are obliged to shall be included in the
tariffs. For consumers choosing another type of metering, i.e. a consumer choosing hourly metering even if
he/she is below the threshold, the additional costs can be charged. Where an agreement on hourly metering has
already been made before the current proposal, no additional costs can be charged within the full length of that
agreement. No particular indemnification shall be given to consumers having installed an hourly meter.

52
Proposed German System
1. Background information
th
According to the new energy law in Germany, which is in force since the 24 of April 1998, all users of electric
power are free to select an utility company of their choice. Based on the “Associations Agreement” from May
22nd 1998 standardised costs for usage of the electric network were published. Today a competitive market in
particular for large and medium power consumers already has developed. Not yet fully established is the power
market for users with low consumption such as private households and small enterprises.
In order to find a proper settlement between the external electricity producer or supplier, the electric network
owner and the final consumer it seems to be necessary to distinguish between the time dependent demand of the
consumer that in fact should be covered by the external electricity producer or supplier that is delivered into the
local net.
Because continuous metering of electric power consumption in particular of medium and small size consumers is
too expensive, typical and representative daily and yearly load profiles for different groups of consumers are
assumed. The integral load profile which is the basis for settlement between electricity producer and electric net
owner is finally obtained by considering the load profiles of all consumers.

2. Tariffs
The costs that have to be paid by external electricity producers or dealers (net users) to the net owner for transfer
of electric power are based on the “Associations Agreement”. In addition to the basic costs further charges to
account for stochastic differences between the assumed load profile and the actual electricity demand of the
consumer which has to be covered by the net owner are raised.
An alternative method which is currently discussed is to distribute the costs arising from deviations between
basic load profile and true electric power consumption to all external producers or suppliers ex post.

3. Metering
The basis for load profiles is quarter-hourly mean values of power consumption.

4. Load profiles
Today two different approaches for the development and usage of characteristic load profiles are discussed: these
two are the “Synthetic Model” (consumer category model) and the “Analytical Model” (area model, similar to
that applied in Norway).
The Synthetic Model uses curves for different types of consumers (e.g. private households, agriculture) which
are fixed in advance. Load profiles are obtained by scaling these curves with the estimated total consumption by
the electricity users. Addition of individual load profiles will give the integral load profile for allsmall scale
consumers which is based on a consumption of 1000 kWh per year. So it is easy to use this load profile for the
actual consumption of the individual customer.
The basis for the Analytical Model is an indirect measurement of the time dependent electricity consumption of a
small scale consumers belonging to the electric net. This is done by metering total power supply to the net and
subtracting the measured time dependent consumption of large scale users. When using the analytical model all
external producers/dealers are required to ensure in sum electricity delivery to the net according to the integral
load profile rather than being responsible only for meeting the demand of their own customers. Deviation
between indirect measured load (see above) and electric power delivery are considered as if they would have
been supplied by the net owner and are therefore charged to all external producers/dealers.
Both models mentioned above show distinct advantages as well as disadvantages. A new concept is in
discussion.
External producers/dealers obtain information about consumption characteristics of electricity users from the net
owner which enables the external electricity suppliers to predict possible deviations between consumer typical
load profile and true electricity consumption. In contrary to the simple model discussed above, each external

53
producer/dealer is required to deliver electricity according to the integral load profile of his own customers. The
net owner’s responsibility is again to balance differences between the (predicted) load schedule for all external
producers/dealers and the actual consumption. The application of this new advanced model results in a
settlement which is quite similar to that used for (large scale) consumers with quarter-hourly metering of power
consumption and later billing of the additional costs caused by fluctuations in the customers energy demand.

5. Settlement
If the Synthetic Model is applied, external producers/dealers are required to deliver electricity into the network
as given by the fixed consumer load profiles. These load profiles will be supplied by the network owner.
Settlement between external producer/dealer and network owner will be performed strictly considering those
artificial load profiles. Additionally, the external producers are charged a mean fee for falling below the required
electricity delivery (according to the artificial load profiles).
If the Analytical Model is applied, load profiles are determined indirectly from measurements and then used as
basis for settlement. Similar to contracts between external producers/dealers and large scale consumers, the
producer/dealer is responsible for deviations between delivery and “measured” consumption for which’s
compensation the electricity network owner has to be paid for.

6. Conclusions
The rapid opening of the German electricity market for different electricity producers without threshold
restrictions lead to a strong competitive market just about one year after liberalisation. It is therefore very
important to introduce very shortly an efficient and reliablesystem which is defining and regulating the technical
and financial aspects of the electricity market. The system preferred in Germany today is based on typical
standardised load profiles for different customers (Synthetic Model) and enables a quite simple and quick entry
for producers and consumers into the new competitive market.

54
New Zealand
1. Background information
1.1. Electricity sector
In New Zealand there are 4 major generation companies whom together account for around 85% of NZ
electricity generation. Smaller companies hold the rest of the generation market.
Most of the 'old' power companies have kept their lines (or network) businesses. Smaller companies hold around
50% of the lines business in New Zealand.
As regards retail supply companies, the biggest player has around 31% of the electricity retail market while
another company has retail assets from 8 other companies whom together account for 21% of the retail market.
Another 5 companies hold the rest of the retail business, while the smaller companies account only for about 3%
of the retail electricity market.
NZEM (New Zealand Electricity Market) is the wholesale pool established under multilateral contract in
October 1996 where buyers and sellers interact to establish the price for electricity. Delivery of the electricity
occurs in the pool together with the settlement between generators and purchasers.
The Rules of NZEM cover every aspect of trading - from entry criteria through to physical dispatch of electricity
- and include procedures for the receipt of bids and offers and financial settlement. Buying and selling of
electricity at a wholesale level is via a pool. Generators offer electricity into the marketplace for dispatch and
transmission through the national grid operated byTranspower. Major users of electricity buy from the pool for
their own consumption and electricity suppliers buy from the pool for supply to retail customers. This physical
spot market is supplemented by the trading of a range of forward contracts. Electricity is priced at market
clearing levels and the price is not capped. The market trades through a custom-built computerised trading and
information system (COMIT - Commodity Market Information Trading System) which links all wholesale level
sellers and purchasers of electricity.
Retail suppliers and end users have a choice of buying their electricity either off the pool (NZEM) or through
bilateral contracts with generators (MARIA - Metering and Reconciliation Information Agreement). MARIA is a
multilateral contract that facilitates the bilateral trading of electricity across the grid. Essentially, metering
quantities are gathered together and allocated to individual bilateral contracts between buyers and sellers.
The Market Administrator, Clearing Manager and Pricing Manager roles of NZEM are ensured by The
Marketplace Co. Ltd. (M-co).

1.2. Electricity Legislation

In April 1998 the New Zealand Government announced an electricity reform package designed to promote
effective competition in the NZ electricity market. Key aspects of the reform package were:
• The sale of a 40% stake of the state owned enterprise Contact Energy Ltd through a competitive bidding
process.
• Further restructuring the ECNZ (Electricity Corporation of New Zealand) into three competing state-owned
enterprises.
• Mandating the split of supply and lines businesses for all energy companies. Lines businesses are to be
owned separately from supply, generation and any other competitive businesses in the electricity sector. In
particular, power companies can either:
• By 1 April 1999, set up a separate trust-owned company for their distribution or electricity
supply/generation activities, or
• By 31 December 2003, sell their distribution or electricity supply/generation services and, between 1
April 1999 and the time of sale operate these activities as separate companies on an arms-length basis.
• Other distribution and supply reforms.
In July 1998 the Government introduced the Electricity Industry Reform Act 1998, aimed at increasing the level
of competition in the electricity sector. Up until these reforms little competition existed at the retail supply level,
with only very large electricity users able to choose their electricity supplier.

55
1.3. Degree of market opening and future thresholds
Up until 1 April 1999 household consumers in a few large cities were offered a choice of two or three electricity
suppliers, but now all consumers in New Zealand are free to choose which company they buy electricity from.
The opening up of retail competition results from a system put in place by the electricity industry and
administered by M-co. It follows the directive, issued by the government in July 1998, for the industry and M-co
to develop a low cost system to enable consumers to switch retail suppliers if they wish.

1.4. Political and regulatory background for the current Metering, Load Profiles and Settlement System
On 1 April 1999 a low cost option to enable small consumers to switch electricity supplier if they wish was put
in place by the industry. This system was developed by the MARIA Retail Competition Committee (MRCC), a
group of industry experts administered by M-co.
At the wholesale level, electricity is purchased in half-hour slots at whatever price the market has set. However,
there is no direct link between the cost of each half-hour slot of electricity purchased by a supplier and the price
paid by its consumers. Linking the two enabled retail competition and will lead to more efficient use of
electricity. There are two ways to make this linkage happen: either using a time-of-use meters or using a load
profiles system. The industry agreed that load profiling was the best solution.
The mechanism put in place by the industry means that wholesale reconciliation can occur on the basis of
profiling. Whether this translates into suppliers competing for consumers, usingprofiling as the basis for doing
so, will be determined by the suppliers themselves. Since the introduction of profiling, suppliers have begun
competing across approximately 60-65% of New Zealand's electricity consumer base.

2. Tariffs
Suppliers and local distribution (lines) companies are separate entities in New Zealand. Often suppliers bill end-
use consumers on behalf of the local distribution companies. This tends to be invoiced to the consumer as a fixed
monthly charge.

3. Metering
In general the half-hour metering of electricity customers’ loads is considerably more accurate than the
estimation of their loads through the use of load profiling. For a large number of customers, the use of load
profiling therefore represents an interim step until the benefit from installing half-hour metering outweighs the
costs.
The initial rules required any customer who wanted to participate in choice to have a half-hourly meter that met
the MARIA standards. This requirement effectively limited competition to those customers who had such meters
in place, in general the larger customers. To address this limitation and support the development of a competitive
retail market, New Zealand began implementing load profiling methods for customers who do not have half-hour
metering in place.
The consumption level at which half-hour metering must be installed in New Zealand is approximately 500,000
kWh per annum. It is intended that this will be reviewed sometime over the next 6-12 months, to see whether it
should be lowered.

4. Load Profiles
At the moment are allowed two general types of profile:
• NSP-Derived profiles – these are profiles derived through the manipulation of the residual Network Supply
Point (NSP) profile (most commonly this will be a Grid Exit Point - GXP) by the National Reconciliation
Manager (NRM).
• Load Research/Engineering Estimate profiles – these are profiles either derived from the statistical sampling
of consumers, or from engineering estimates of consumption (e.g. street lighting).
The MRCC has agreed that the NRM’s processes should allow for the use of both types of profile. The default
set of profile classes used since 1 April 1999 is:

56
1. Non-MARIA-compliant half-hour metered loads (essentially a statistically sampled profile with 100%
sampling).
2. Non-metered loads (e.g. street lighting).
3. Interval time of use meters. Typical Meters in this classification include the following:
• Day-Night Two Rate Meter,
• Night Only Meter,
• Night Only plus Afternoon Boost Meter,
• Five Rate Time of Use Meter.
4. Separately metered controlled load.
5. Non-separately metered controlled load. Typical installations include a ripple controlled water heater, but
with only one meter measuring the whole installation including the water heater. The controlled load may be
switched on and off at variable times of the day.
6. Uncontrolled load 24-hour meters using Grid Exit Point (GXP) profile.
The first two profile class types are examples of Load Research/Engineering Estimate profile classes, while the
last four are examples of NSP (GXP)-Derived profile classes. A Profile Class is a generic grouping of similar
profiles.
For part load controlled meters, engineering estimates are to be used in allocating total consumption into its
controlled and uncontrolled portions.
Profile references will be three characters. The characters will ensure uniqueness at any GXP. A Supplier may
use the same Profile reference but with different Profile shapes at different Grid Exit Points. For example a
Supplier could use X01 to refer to an interval time of use meter that switches at 11pm and 7am in one GXP.
Customers in a different GXP can also be allocated to X01 but the switching times are 10pm and 8am. X01 is
being used to represents a different profile at each GXP, and the Supplier will need to provide the NRM with the
different switching times. This simplifies the loading of ICPs (Installation Control Point – the point of
connection to the network) for Suppliers as they can use the same Profile reference across a number of GXPs but
means that a profile is only uniquely referenced whenboth the Profile reference and the GXP that it applies to
are listed.
Two Profile Codes have been predefined. These codes are:
• HHR. This is not a real Profile but indicates to the Data Administrator that the ICP should be processed as a
MARIA compliant half-hourly meter.
• GXP. This denotes uncontrolled load 24-hour meters using GXP residual Profile. This Profile is open to all
customers and does not require the Supplier to provide the NRM with any additional information to define
the Profile.
Suppliers wishing to use Profiles must:
• decide which Profiles they wish to use,
• inform the MARIA Administration Manager of these Profiles and ask the NRM for a unique Code
Reference for the Profile.
Where Load Research / Engineering Estimate Profiles are to be used, the Supplier must provide to the MARIA
Administration Manager the information it intends to use as the basis for their Profiles.
Future Profiles will need to be registered / approved by the Administration Manager. The NRM will be
responsible for allocating new Profile references and ensuring uniqueness of references.
To date, virtually all profiles approved in New Zealand have been based on the use of the GXP load shape.

57
5. Settlement method
The new functions established for the new system since 1 April 1999 are Data Administrators, the Registry and
the National Reconciliation Manager (NRM).
Data Administrators are contracted to the Supplier. They have responsibility for gathering the meter readings
from data collectors, preparing estimates in instances when no meter readings are taken, and aggregating the
metering data for provision to the NRM.
The customer switching process is driven from a national database, the Registry, which identifies every metering
installation by way of a unique identifier or ICP (Installation Control Point). Every supplier has allocatedICPs to
their customers, which will be printed on monthly power bills. The Registry will ensure that allICPs are
included in the national reconciliation process.
The process is reconciled by d-Cypha (a Transpower subsidiary), the National Reconciliation Manager (NRM).
The introduction of profiling into the electricity industry’s reconciliation processes provides an opportunity for
the industry to move from the current ‘Differencing Reconciliation’ approach to a ‘Global Reconciliation’
approach.
Under Differencing Reconciliation, independent suppliers’ metered or profiled consumption (grossed up by the
value of losses), along with embedded network consumption and embedded generation in a local network are
netted off against the total metered consumption at the Network Supply Points (NSPs) supplying that network.
The remaining quantity of energy is assigned to the incumbent supplier operating in the network. The incumbent
supplier therefore accepts all errors and inaccuracies in the metering and estimation of independent suppliers’
energy amounts. This includes inaccuracies arising from loss calculations, theft, metering errors and profiling
errors.
With Global Reconciliation, the concept of an incumbent supplier is removed. Once the loss-adjusted metered or
profiled energy quantities of each supplier operating within a local network are removed from theTranspower
metered totals, the remaining energy is allocated across all suppliers on a pro-rata basis. This approach is
therefore seen by many as inherently fairer than the Differencing Reconciliation approach, particularly in
instances when an incumbent’s market share within a local network falls.
Settlement occurs at two levels - under bilateral contracts (MARIA), and under the wholesale pool (NZEM).
Currently the NZEM Rules do not accommodate metering data based on profiling. Consequently, the
Reconciliation Manager (RM) cannot profile any non-half-hour metered information submitted under the
auspices of NZEM. MARIA is the only industry agreement enabling the reconciliation of profiled metering data.
This will lead to the withdrawal of energy trades from the NZEM. As incumbent suppliers move to global
reconciliation, which involves the elimination of their incumbent status, through the submission ofall metered
data into the reconciliation process (both half-hour metered and non-half-hour metered), increasingly large
quantities of electricity are going to be reconciled under MARIA.
The overlying principle is that NZEM trading is reconciled by the Reconciliation Manager (RM) and non-NZEM
trading by the National Reconciliation Manager.
The present settlement process has the following steps:
Collection of Metering Data
Data for all Half Hourly Meters are read (downloaded) electronically. This may be carried out through the use of
portable devices, or remotely via a recognised communications medium.
Data for all Non-Half Hourly Meters are collected in a manner as agreed between the Data Administrator (DA)
and each Supplier that it represents. A suitably qualified Data Collector collects Non-Half Hourly data.
Receive a List of ICPs from the Registry
At the beginning of each monthly reconciliation process, the DA obtains from the Registry a list of all theICPs
per Supplier, Profile and NSP that the DA represents.
The DA will validate the Registry’s list by comparing it to theDA’s own list of ICPs. If inconsistencies are
found, then the DA will produce a discrepancy report and submit it to the Registry.A discrepancy report must be
submitted by each DA to the Registry for the initial reconciliation file and all subsequent wash-ups. If no
discrepancies exist, then the data field for the report will indicate this.
Either the Supplier or the Network will submit discrepancy corrections to the Registry.

58
Application of Accruals
Consumption data for all Non-Half Hourly Meters must be recorded on a monthly basis. Where the consumption
data for a Non-Half Hourly Meter has not been collected in a given month, or the consumption data was
collected before the end of the month, the DA will need to apply a Forward Estimate for the consumption at that
ICP.
A Forward Estimate (FE) shall be used to estimate the consumption at an ICP for a given month. Forward
Estimates shall be replaced by Historical Estimates (HE) in the month following a physical Meter reading. A HE
differs from a FE in that actual metering data is used to estimate what electricity is consumed during a given
month. Once calculated, HEs will replace FEs as the estimate of a consumer’s daily consumption within a
month.
The HE accrual methodology is as follows:
- The ICP consumption between two meter reading periods is multiplied by the ratio of the total profile
consumption for the reconciliation period over the total profile consumption during the meter reading
period.
- If a part month is included in the meter reading period, then that month’s consumption is apportioned
between meter reading periods to establish the HE.

Application of Local Losses to Half-Hourly Metered Data

For each monthly reconciliation process DAs must apply Local Loss factors to all Half-Hourly metered
consumption data per Supplier per NSP. DAs will obtain Local Loss factors from the Networks, who are
responsible for determining all Local Loss factors.
Given that Networks could potentially assign Local Loss factors to different time periods (for example: half
hourly, daily, weekly or monthly), each DA must be able to accommodate different Losses for each Half Hour of
the day. This approach will enable the DA to accommodate loss factors for any time period. For example, if a
Network submitted one loss factor which applied across the whole month, then the DA would apply the same
factor over every Half Hour period in that month. Alternatively, a Network could submit different Local Loss
factors for every Half Hour in the day.
Aggregation Approach
Once all consumption data has been collected, Half Hour data has been loss adjusted, and Forward Estimate
accruals have been generated where necessary, each DA are responsible for aggregating the data and supplying it
to the NRM for reconciliation.
The aggregation is an individual sum of all the consumption data in a particular Profile per Supplier per NSP,
and a sum of all loss adjusted Half Hourly data.
DAs are responsible for aggregating consumption figures for non-metered supply (for example streetlights), for
submission to the NRM. The consumption for non-metered supply are calculated using an agreed approach. DAs
may obtain power supply times for non-metered supply from the Networks.
DAs aggregate consumption for non-metered supply per Profile, and submit it to the NRM for reconciliation.
To date only limited analysis has been carried out to see how accurate the estimated consumption for small
residential consumers is against actual consumption.

6. Other questions resulting from choice of system

6.1. Time restraints and other conditions for changing supplier
There is no any time restriction for changing the supplier. Some suppliers do not stipulate a minimum period of
supply through contracts, while others do. Practically, it should only take approximately 1-2 weeks. However,
with all the industry restructuring that has occurred lately, many suppliers are still implementing new systems.
Hence, delays are occurring in the switching of customers between suppliers.
The process for customer switching has four steps. If a Supplier (“the new Supplier”) enters into a contract for
the sale of electricity with a person that has the effect of superseding a contract for the sale of electricity that
person had with another Supplier (“the old Supplier”) then:

59
1 The new Supplier will notify the Registry. This notification will be provided theoretically within three
working days of the effective date.
2 The Registry will notify the other Supplier.
3 The Registry will notify the Distributor.
4 Both Suppliers will agree a process for a final reading or deemed final reading of the meter of the
person changing Supplier.

6.2. Costs resulting from changing supplier (who pays)

The issue of who pays for the costs of switching (e.g. metering) is still a little unclear at times - it appears that
usually the new supplier is absorbing most, if not all, of the costs.
For household consumers, it is possible that some suppliers to charge for the changing of supplier.

6.3. Costs of metering – included in the tariff?

The general principle governing the development of rules for the competitive market is that those who benefit
from a service or product feature should be those who pay for it. The question of who benefits is tied to the
question of how universal the metering requirements should be.
The metering requirement will be a MARIA requirement. As such, it is part of the terms of agreement between
the power pool and the supplier. Responsibility for ensuring MARIA-compliant metering rests with the supplier.
How much of the cost of that metering the supplier will be able to pass back to the customer will depend on
market forces.
If half-hour metering is not required at any level, butis left to be chosen by market players, those who see a
benefit from half-hour metering will pay for it. Whether customers who want new half-hourly metering are able
to convince their suppliers to pay for it, or suppliers who want the customer to have such metering are able to
convince the customer to cover the cost will be a matter of negotiation between those parties.
If half-hour metering is required for customers above a cut-off only if they switch suppliers, the metering cost
will be borne by the customer as a cost of participating in competition. Again, the extent to which the supplier
absorbs part of that cost is a matter of price negotiation between the supplier and the customer.
If half-hour metering is required for each customer above a certain cut-off, there must be a societal benefit of the
metering beyond that which would accrue directly to the customer and its supplier. Otherwise there would be no
basis for imposing the requirement. If part of the benefit accrues to other customers (via their suppliers) in the
form of reduced settlement error and uncertainty, or less specifically in the form of a more active and smoother
functioning market, all customers who receive this benefit should share in the cost.
One way to implement this cost sharing would be via the networks. However, the networks are not parties to the
retail settlement process, and do not necessarily own their customers’ meters. They also would have little
incentive to play the role of tax collector for this purpose. Another approach would be for the pool to charge each
supplier a supplemental metering fee for each customer, regardless of whether that customer required new
metering or not. New metering installations for existing customers above the cut-off would then be reimbursed
from these funds.

60
Appendix 2: Future plans for metering load profiles and settlement systems in other
countries

Ireland
ESB Data Collection Infrastructure
Load Recorders were first used by ESB in the late sixties for Load Research. By the mid-nineties, several
generations of Load Recorder technology had been employed. These included DRICO, Westinghouse and in the
nineties, PSI technology. In the early nineties, the industry standard MV-90 meter data translation and analysis
software was installed. MV-90 coupled with remote communications enhanced load research efficiencies and
capabilities considerably.
Since the mid-nineties, a significant penetration of electronicMulti Function Meters (MFM) has occurred in the
medium to large customer categories giving the benefits of higher accuracy, lower costs, complex tariffication
capability, load profile capture and remote communications to external reading devices. TheseMFM’s have
further enhanced load research capabilities and efficiencies.
ESB’s Data Collection infrastructure has now expanded to the extent that Load Research is no longer the
predominant end use. MV-90 is now used in a networked environment to remotely interrogate data from a
variety of devices and sources. The system performs Electronic Data Interchange with existing internal
settlements systems and in the near future, with Industry settlements systems. The following are key elements of
ESB’s data collection infrastructure:
- Substations for load research
- Medium and large sized customers for load research and value added services
- Alternative Energy Generators for settlements
- Generation Stations for internal settlements
When the electricity market opens in Ireland in February 2000, around 350 customers will be eligible to
participate in the new market structure and associated settlements systems. These customers will require profile
metering with remote communications to successfully participate in the new market structures.ESB’s Data
Collection infrastructure will ensure that all eligible customers can participate in the new markets in February
2000.

Slovenia
1. Electricity market in Slovenia
Electricity supply industry (ESI) is completely unbundled with regard to function and inside function as well.
There are eight production companies integrated with regard to production speciality, onetransmission company
and five distribution companies integrated with regard to geographical proximity. Thetransmission company
purchases electricity from the generators (through yearly contract) and sells it to the distribution companies and
to a few big consumers (through yearly contract). The transmission company trades all electricity for grid
supply.
In regard to ownership Slovenia has still public owned system. The transmission company and production
companies are 100% state-owned. About 95% of the distribution companies are also still state-owned while the
rest is owned by different funds (capital, development) and by private capital (of distribution’s staff). The future
share of private capital in ESI is going to become higher gradually through the development of privatisation
process.
The implementation of the competition at the moment is on the production side, but only for new generation
facilities, which can be built by independent producers. The rules for introducing competition and entity
responsible for implementation and monitoring this implementation are defined in new Energy Law which is at
the moment in the phase of second reading in the Parliament and that means a lot of things could be changed
there yet.

61
2. Tariffs
The new tariff orders introduced in December 1998 did not introduce essential modifications in order to fulfil the
requirements of the free market. At the moment there are no unbundling tariffs for the end use customers (no
separate charges for the use of the transmission and/or the distribution grid, no metering tariffs).
At the moment a lot of attention is focused on design of the transmission tariffs.

3. MLPS situation in Slovenia

All HV consumers (consumers at 110 kV) and some MV consumers (consumers supplied at 1-35 kV) have meter
registering electricity consumption by every quarter of an hour. Yearly consumption of these consumers varies
from 1100 GWh (the biggest) to 4 GWh (the smallest). Almost all HV consumers purchase electricity from the
transmission company (ELES). Only 2 of them are distribution’s consumers.
All distribution companies have meter registering electricity consumption by the hour or quarter of an hour.
HV consumers and distribution companies have classical meter system supported by the register equipment and
communication lines for the exchange of information. The costs of this system are very high and its installing is
reasonable only for the consumers with large electricity consumption.
The functions of the metering system are application of time of day tariffs, automatic meter reading, and a
possibility for adding a load management function by the customer and remote reading consumption. The
software for remote reading, exchange data and billing does exist already for this consumption.
According to the draft of the new Energy Law, all consumers with more than 41 kW power could become
eligible consumers. A lot of MV and LV consumers could be immediately included into the near future free
electricity market when legal conditions shall be introduced. At the moment they do not have suitable metering
system for hourly metering. That means that they would have to change their meters with new ones in the case
they will want to become a part of the future free electricity market. That will not be a problem probably because
the costs of an hourly meter are not so high (about 100 Euro). The main problem will be communication lines
and their costs and software for data processing.
All consumers with demand up to 41 kW will be non-measured small customers. They will be probably left
outside the free electricity market for a few first years. The main reason is that IT is not very well developed. As
we know the large number of messages must be transferred between the parties in the free market to balance
their settlements. Moreover in Slovenia there are no currency methods available for the formation of the load
curve, which is necessary for introducing new billing system for small consumers in free market.

Spain
1. Introduction
Since 1.1.1998, the Spanish electricity industry has been regulated by the new Law 54/1997 passed by
Parliament at the end of 1997.
According to this new Law, electricity generating has been liberalised thus introducing competition. A
generating market has been created where generators, distributors, independent suppliers and qualified customers
have to go to buy and sell electricity.
Currently there are two possibilities. Any agent with authorisation can go to the daily offers market and present
their offers to buy or sell electricity. A second possibility is that generators and qualified customers can sign
bilateral physical contracts.
In addition to the liberalisation of generating activity, the supply of electricity has been considered in the new
regulation as a specific activity, separated from distribution. Qualified customers can go directly to the market to
buy the electricity or can be supplied by an independent supplier.

62
The liberalisation of the supply has been introduced gradually according to the following schedule:
Threshold Clients % Market
1.01.1998 > 15 GWh/year 558 26.0%
1.01.1999 > 5 GWh/year 1,925 33.2%
1.04.1999 > 3 GWh/year 3,282 36.6%
1.07.1999 > 2 GWh/year 4,829 39.0%
1.10.1999 > 1 GWh/year 9,120 43.1%
1.07.2000 > 1 kV 65,000 53.0%
6
2007 All 20.4 x 10 100.0%
According to this new situation, non qualified customers pay an integrated tariff which includes all the tariff
components (generation, transmission, distribution, etc) and qualified customers who decide to go to the market
or to buy electricity through an independent supplier paying the price of the electricity, on the one hand, and an
access tariff, to his regional distribution company, on the other, which includes all the tariff components,
excluding that of generation.

2. Settlement process
There are two settlement processes. One carried out by the market operator, which involves the operations
related to the generation market, this means, purchases and sales of electricity made by generators, distributors,
independent suppliers and qualified customers.
A second settlement process which is the responsibility of the CNSE (Comisión Nacional del Sistema Electrico)
and whose objective is to settle the regulated activities. The tariffs in Spain, both, integrated andaccess are
standard for the whole country, and include all the components that end users have to pay. In both cases, as
qualified customers (in this case the generation cost is not included) and as regulated customers (including non
qualified and qualified customers that prefer to buy electricity at the integrated tariff). The settlementprocess
permit that the different regulating agents (Transmission company, distributors, system operator, market
operator, etc) receive the recognised costs according to the regulation.

3. Market settlement process

The market operator (OMEL), as responsible for the system’s economic management, is in charged of
calculating on hourly base, the settlement and payment rights and obligations of all the agents acting in the
generation market.
The payments to the generators are calculated based on the net production metered at the connection point of the
transmission grid of each power station.
The charges made to the distributors independent suppliers or qualified customers are calculated as follows:
• If we are dealing with a qualified customer who buy electricity either directly or through an independent
supplier, the charge is calculated based on the consumption registered on his meter increased by the
percentage of losses that correspond to his consumption level. Every year the government publish the
percentages of losses corresponding to each tariff, based on the consumption level.
• Regarding distributors, the energy charges are calculated as follows:
♦ Transmission losses are calculated as the difference between the energy injected by the generators into
the grid, and the energy metered at the different connection points between the transmission and the
distribution grids.
♦ The energy metered at each connection point between the transmission and the distribution grids is
increased by the percentage of looses as calculated in the previous paragraph.
♦ The energy metered at each connection point between the transmission and the distribution grids,
increased by the losses is reduced by the energy corresponding to qualified customers that are

63
 connected in this distribution zone. The net balance resulting at each connection point is allocated to the
corresponding distributor.
The metering equipment characteristics and conditions and the responsibilities of the different agents are
regulated by the Royal Decree 2018/1997 that came into force 1.01.1998.
According to this, all the agents that operate in the market, which means, generators, distributors and qualified
customers that buy electricity directly or through an independent supplier has to have a meter with hourly
registers installed.
When the supply to low tension consumers is liberalised, the necessity to have an hourly meter installed could be
an obstacle due to the price of the equipment. Thus the general idea is that in this case the settlement will be
made based on load curve patterns, but this is something which has not been legally development yet.

The Netherlands
1. Background information for The Netherlands
As of 1 July 1999 a new Electricity Act entered into force in The Netherlands. Gradually, the Act will give
customers and suppliers in the electricity market the power of choice; in step with this market opening electricity
will develop into a commercial commodity. Any operator is free to generate electricity, while major customers
(with connections of over 2 MW) are free in their choice of supplier. Customers having a connection value of
over 3 x 80 Amperes will have freedom of choice as of 1 January 2002. Finally all other customers will follow
suit as of 1 January 2007.
The Electricity Act creates conditions aimed at ensuring the continued proper functioning of the electricity
supply system. The electricity companies have adapted their organisational structure to this new legislation. The
networks have been separated administratively from the electricity companies into network companies that shall
designate independent network operators. Thus, the network companies will provide grid services and facilitate
the market for all parties: generators, final customers, suppliers/traders and the Amsterdam Power Exchange.
At this moment 22 regional network companies are designated and one company is responsible for the national
High Voltage-grid: TenneT, Transmission System Operatorbv.

2. Metering in the Netherlands

In accordance with the Electricity Law the joint grid owners have prepared a draft Metering Code. The regulator
has to approve this code. The metering code describes the conditions for the relation between the conduct of grid
owners versus the clients concerning the metering and the exchange of metering information.
Everybody is allowed to realise and control metering facilities if defined quality requirements can be met. The
grid owners are responsible to maintain the quality of the metering facilities and the collection of the metering
data.
The metering code demands exactness, the solution of disturbances, the control of metering facilities and the
exchange of data.
The clients who are connected to the grid and who have balance or program responsibility are divided in three
categories:
• Contracted power 1 MW or more, registration takes place in metering periods of 5 minutes. Daily, the grid
owner collects the data. The data are the basis for the settlement.
• Contracted power is between 0.1 and 1 MW, registration takes place in metering periods of 30 minutes. At
the least yearly the grid owner collects the data by reading the data buffer. The settlement is based on the
individual load profiles.
• Contracted power below 0.1 MW, the metering facilities of the connected clients are read at the least once in
a year. The settlement is based on a load profile of the category. The grid owner decides which category, but
distinguishes between a connection of at the least 3 x 80A or lower than this value.

64
Turkey
1. Electricity market
The industry was mostly centralised over the period leading up to the early 1980s, with most generation,
transmission and distribution being concentrated in the hands of a single, state - owned organisation, Turkish
Electricity Authority (TEK). In the early 1980s, distribution in two regions was transferred two independent
companies, covering the Anatolian side of Istanbul, and covering the region aroundKayseri .In addition, two
independent utilities provided generation and transmission in the areas aroundAdana and Mersin, and Antalya,
respectively.
Since 1985 legal arrangements took place in the power sector, generation, transmission, distribution and trading
of power was removed the state monopoly as BOT, BO, TOOR Autoproducers.
In 1994, TEK was split into two entities, TEAS, operating the generation and transmission, TEDAS, operating
the distribution. TEDAS, the other major power sector entity, maintains a more limited set of responsibilities
within the sector. It is responsible for delivering power to end-users and administering contracts with private
companies that have taken over operating responsibility for state-owned plants. There are 5 distribution
companies. Delivery of energy to end user customer is served by state owned company with share of 83%, by
private company with share of 17%.
For transferring of operating right of distribution facilities, the country divided into 29 regions. The evaluation
activities of the offers for the region were completed. The results were declared and offers of 5 region are
cancelled. Regions have not been transferred yet.
Nowadays ESI have been reorganised. New generation will be constructed in the private sector. Teas will be
split two companies as generation company (includes except TOOR Plant) and transmission company, which is
state-owned company. A regulator, the Energy Market Board will be established by statute, independent both of
the industry and of political interference.
Within the existing industry model, TEAS maintains responsibility for a wide variety of functions, including the
following:
- generation from owned facilities;
- purchasing from independently owned and/or operated generators; · transmission operation and
maintenance;
- system dispatching;
- system planing;
- sales to distribution companies
Commercial relationships within the existing industry model are very simple. TEAS purchases power from
generating plants that are independently owned and operated and receives payments from distribution companies
for delivery of power.
There are three primary methods of co-ordination within the industry - management directive (primary within the
vertically integrated TEAS); contracting (primarily for procurement of power from private owners / operators of
generation); and the ministry of Energy and Natural Resources (for specifying both retail and wholesale tariffs,
setting and implementing policy, and resolving disputes among industry participant.

2. Tariff structure
Today, the Transmission portion of TEAS remains essentially fully integrated with the Generation portion of
TEAS as seen by either a consumer of electricity or a customer of the high voltage grid. By this we do not mean
that the operations of Generation and Transmission are integrated -- in fact, these operations are clearly distinct
within TEAS.
Today's approach to recovery of transmission related costs is through a "bundled" tariff, combining energy
(generation) costs as well as transmission costs. In order to meet the needs of future energy markets, it will be
necessary to develop a separate, stand-alone transmission tariff. At present it appears as if TEAS currently can
identify transmission-specific costs as well as allocation of common costs (e.g., head office costs shared by both
generation and transmission). Therefore, it should be possible to relatively quickly develop stand-alone
transmission tariffs. In the future, generation and transmission activities will be physically split.

65
Specifically with respect to Transmission pricing, onlyautoproducers (accounting for a very small fraction of the
total energy produced or consumed within Turkey), and, possibly, certain BO / BOT / TOOR contracts (see
below) have access to an "unbounded" transmission tariff -- i.e., one which charges for transmission services
only, exclusive of energy. All other energy consumers in Turkey see a "bundled" tariff -- either directly from
TEAS, or through Distributor / Suppliers.
For TEAS customers, a few directly connected industrial customers receive bundled energy and transmission
services under a single and a two-part tariff, with the rest – including TEDAS and the Distributor Suppliers --
receiving energy and transmission services under a single part tariff and a two-part tariff. Separate charges are
generally levied for reactive energy.
Autoproducers engaged in "wheeling" transactions throughTEAS's network pay a tariff for use of the network
that includes a distance component and a component for energy losses. The distance component is based on the
current costs of a hypothetical connection, including transmission lines, voltage transformers, and connection
assets. This component is charged as a fraction of the energy transmitted, assuming a specific load factor (70%)
for the autoproducer. Losses are assumed to be a constant percentage of energy transmitted.

3. Metering, load profiles and settlement

Electricity trade between the parties is operated under the Energy sale agreement. Distribution companies’ and
industry customer’s meters are read the last day of each month. At that time those data are sent to the centre by
fax. The values of the each connection point are collected and bills are prepared. That process takes 5 days. After
being taken the bill, customers have to pay in 15 day. Time of the reading, billing and collection process takes
about 3 weeks. Some of the small industry customer bills are prepared at local.

66