Establishing a financial incentive scheme to reduce fire starts from electricity distribution assets - the F-Factor

Consultation Paper

Contents

  1. Introduction
    Structure of this paper
    Making submissions
  2. Overview of the economic regulatory regime
  3. Overview of the f-factor scheme
  4. Issues for consideration
    Definition of fire starts
    Definition of the days on which the fire incentive scheme is to apply
    Fire start benchmark or target
    Fire start incentive rate
    Role of the AER

1. Introduction

On Saturday 7 February 2009, Victoria experienced catastrophic bushfires that resulted in substantial loss of life and property.

While electricity distribution assets generally start around 1 to 1.5 per cent of bushfires, on that day up to 4 per cent of the fires appear to have been started by electricity distribution assets.

Immediately following Black Saturday, the Department of Primary Industries worked with other government departments and agencies to identify how the risk of electricity distribution assets starting bushfires could be reduced. A number of measures were identified including tightening the provisions for bushfire mitigation plans, strengthening the powers of Energy Safe Victoria and introducing a financial incentive scheme (also referred to as an f-factor scheme) to encourage improvements in the management of electricity distribution assets, so as to reduce the number of fires started by those assets.  The objective of introducing the f-factor scheme was to address a potential gap in the current economic regulatory regime.

The Victorian Parliament recently passed legislation to provide for the f-factor scheme. The legislation provides the "head of power" to authorise the Governor in Council to make an Order which may specify a range of matters relating to the f-factor scheme.

The Department of Primary Industries is now seeking comments from interested parties on key issues that need to be considered in the drafting of the Order in Council.

Structure of this paper

This paper provides an opportunity for interested parties to comment on the issues that need to be considered by the Victorian Government in the drafting of the Order in Council.

The paper sets out the following:

  • Section 2 provides an overview of relevant aspects of the economic regulatory regime
  • Section 3 provides an overview of the f-factor scheme
  • Section 4 identifies the issues on which comment is sought from stakeholders.

Making submissions

Submissions are preferred in electronic format and should be provided to the Department of Primary Industries by Friday 18 February, 2011.

Via email: mark.feather@dpi.vic.gov.au

Or mailed to:

Mark Feather
Director, National Energy Development
Energy Sector Development Division
Department of Primary Industries
GPO Box 4440
Melbourne VIC 3001

The Department encourages respondents to make their submissions available publicly. Unless certain sections of submissions are marked 'IN CONFIDENCE', all sections of the submissions will be treated as public documents and will be placed on the Department's website www.dpi.vic.gov.au for public access. Formal requests for confidentiality will be honoured; however, Freedom of Information access requirements still apply to confidential submissions.

2. Overview of the economic regulatory regime

As the electricity distribution network is a natural monopoly, the revenues earned by the electricity distributors are regulated. A national economic regulatory regime was established by agreement of the Council of Australian Governments (COAG) on advice from the Ministerial Council on Energy (MCE).  This regime is set out in the National Electricity Rules (NER), which are made under the National Electricity Law (NEL).

The economic regulator of the Victorian electricity distributors is the Australian Energy Regulator (AER). The NER, together with guidelines made by the AER under the NER, spell out the detailed processes and principles to be applied by the AER.

Service is an integral component of the economic regulatory bargain. In determining the allowed revenue for an electricity distribution business, the economic regulator must trade off service and price components – that is, customers may pay a higher price for a higher level of service or conversely may pay a lower price for a lower level of service.

To improve service, additional expenditure allowances can be provided to an electricity distribution business as part of the revenue determination. However, under this approach, there is a risk that customers will pay more but not actually receive a higher level of service. Alternatively, performance-based schemes can be used to ensure that customers only pay more if a higher level of service is delivered (or conversely pay less if a lower level of service is delivered).

The process for determining the revenue the electricity distribution business may recover is set out in Chapter 6 of the NER. Chapter 6 includes provision for various performance incentive schemes, including the Service Target Performance Incentive Scheme, that encourage electricity distributors to maintain and improve performance, for example, supply reliability. The performance incentive schemes have been provided to balance the incentive for the electricity distributors to reduce service levels in order to reduce costs and thereby increase profitability, and to ensure that customers only pay more for improved reliability when that improved reliability has actually been delivered.

Following the 2009 bushfires, concerns were raised that the existing incentive schemes do not adequately target the arrangements that electricity distribution businesses have in place to mitigate the number of fires caused by electrical assets. In addition, concerns have been raised that the current service incentive scheme provides greater incentives for electricity distributors to improve services in "high density areas" which are less exposed to bushfire risk than "low density areas".

To address this concern, the Department of Primary Industries developed the concept of a financial incentive scheme (also referred to as an f-factor scheme) to encourage improvements in the management of electricity assets to reduce the number of fires started by electricity assets.

3. Overview of the f-factor scheme

To reduce the number of fires started by electricity distribution assets, additional expenditure allowances can be provided to the electricity distributors with the risk that customers pay more but there is no reduction (or even an increase) in the number of fires started by electricity distribution assets. An alternative approach is one where customers pay more through an incentive scheme, such as the f-factor scheme, only when the number of fires is actually reduced.

Through the f-factor scheme, the electricity distributors have a financial incentive to reduce the number of fires started by electrical assets where the costs to do so are less than the benefit received through the financial incentive scheme.

It is proposed that the f-factor scheme will operate in a similar way to the current service incentive scheme. It will balance the service/price trade-off by linking annual changes in the electricity distributors' regulated revenue to the number of fires started by its electricity distribution assets each year. However, while the service incentive scheme is based on averages and is therefore skewed towards "high density areas", the f-factor scheme will be based on absolute numbers and is therefore skewed towards areas where the likelihood of fires starting is higher.

The impact of the f-factor scheme on the distributors' revenues will be a function of the number of fires started by electricity distribution assets in each year, the benchmark or targeted number of fires (ie. the annual benchmark) and the penalty rate per fire.

If the number of fires started by the assets of an electricity distributor is less than the benchmark or target, the revenue earned by that electricity distributor will increase by a pass through amount which is determined by multiplying the fire start penalty by the number of fires started below the benchmark or target. In this case, the customers of that electricity distributor will pay more as the electricity distributor can recover the revenue from customers via a pass-through.

Conversely, if the number of fire starts is greater than the benchmark or target, the revenue earned by that electricity distributor will decrease by a pass through amount determined by multiplying the fire start penalty by the number of fires started above the benchmark or target. In this case, the customers of that electricity distributor will pay less.

Under both the existing service incentive scheme and the proposed f-factor scheme, the customers that pay more may not be the beneficiaries of the scheme. Under the existing service incentive scheme, customers pay more when the average reliability for all customers in that electricity distribution area improves even if there is no improvement in reliability experienced by that customer. Similarly, the beneficiaries from a reduction in the number of fire starts will not necessarily be all customers in that electricity distribution area.

4. Issues for consideration

The Energy and Resources Legislation Amendment Bill 2010 was introduced into Parliament on 22 June 2010 and received Royal Assent on 14 September 2010. The Bill inserted a new Division into the National Electricity (Victoria) Act 2005 (the Act) to provide for the f-factor scheme. Section 16C(1) of the Act provides the "head of power" for an Order in Council to establish the f-factor scheme to reduce the risk of fire starts and reduce the loss or damage caused by fire starts.

The Act requires the economic regulator (the AER) to make two different types of determinations in relation to the f-factor scheme:

  • A determination of the parameters of the f-factor scheme that will apply in each five year regulatory period (referred to as an f-factor scheme determination)
  • An annual determination of the f-factor scheme amount that will either be paid by electricity distributors to customers if the number of fires increases, or by customers to electricity distributors if the number of fires decreases based on the parameters of the f-factor scheme (referred to as an f-factor amount determination).

Section 16C(2) of the Act specifies the type of information that may be included in the Order in Council, including:

  • How the AER is to make, publish, implement and administer these two different types of determinations
  • A definition of fire starts to be covered by an f-factor scheme determination
  • Any benchmarks, targets, incentives, rewards or penalties for inclusion in an f-factor scheme determination.

The f-factor scheme parameters are related in that a lower (higher) penalty rate will apply when the number of potential fire starts included within the scheme is greater (fewer).

It is expected that the first f-factor scheme determination would be made in 2011 and would apply during the 2011-15 regulatory period. The first f-factor amount determination is expected to be made in 2012 in respect of the 2011 performance, with the pass-through amount applied to the distributors' revenues in 2013.

The following key issues for consideration will be discussed in the remainder of this section:

  • Definition of fire starts
  • Definition of the days on which the fire incentive scheme is to apply
  • Fire start benchmark or target
  • Fire start incentive rate
  • The role of the AER.

Definition of fire starts

For the purposes of the f-factor scheme, a fire could be defined as any fires started by electricity distribution assets or, alternatively, fires that are required to be reported to Energy Safe Victoria (ESV) as serious electrical incidents. The definition also needs to consider the circumstances under which the electricity distribution assets are considered to have started the fire.

Section 3 of the Electricity Safety Act 1998 defines a serious electrical incident as "an incident involving electricity which causes or has the potential to cause the death of or injury to a person, significant damage to property or a serious risk to public safety".

Significant damage to property has been defined by an ESV Guideline as:

  • Fire damage greater than 0.3 hectares
  • Any livestock loss
  • Greater than $5,000 damage to property other than network assets
  • Damage that has potential for significant media interest
  • Damage serious enough to warrant on site action to mitigate risk to the public by Police, Ambulance Service, Melbourne Fire and Emergency Service Board (MFB), Country Fire Authority (CFA), Victorian WorkCover Authority, a statutory body or an emergency service provider.

However, the reporting of serious electrical incidents to the ESV has not been audited to the extent that would provide assurance that all fires that met the above criteria have been reported and, conversely, that fires that did not meet the criteria had not been reported.

If such a measure was to be used as the basis for the fire incentive scheme, it is likely that the definitions would be applied more forensically than in the past. This may provide a perverse incentive to not report the types of incidents that have previously been reported, by more strictly adhering to the definition in the guideline.

For the purposes of the f-factor scheme, it would be administratively simpler to define fires as all fires started by electricity distribution assets, rather than only fires that exceed some form of threshold.

Data on all fire starts can be obtained from the CFA and MFB. However, the Department understands that the data on causes of fire may not be reliable enough for the purposes of the f-factor scheme.

If the definition of fires were to include all fires started by electricity distribution assets, this provides an incentive to improve the recording of the cause of fires. However it is noted that if the definition of fires includes all fires started by electricity distribution assets, there is the potential that the baseline will be set too high for the first regulatory period after the introduction of the scheme. This infers that the penalty rate should not be set too high in the first regulatory period until there is greater confidence with the data.

Fire risk is a function of likelihood and consequence. Whilst the electricity distributors are able to manage and minimise the likelihood of electricity distribution assets starting fires, the consequences are a function of a range of factors that are outside their control. To maximise the effectiveness of the scheme, all fires could be considered, not just those that exceed a relatively arbitrary consequence threshold.

In defining a fire for the purposes of the f-factor scheme, the circumstances under which the fire started also needs to be considered. Many fires are started by vegetation, but in most cases the vegetation is outside the clearance space. As an example, the Beechworth fire was started when a tree located on Crown land fell on a powerline. For the purposes of the f-factor scheme, was this fire started by the tree or by the electricity distribution asset?

If fires initiated by vegetation falling on electricity distribution assets are included in the f-factor scheme, and the incentive rate is high, it provides an incentive for the electricity distributors to remove trees located near electricity distribution assets. This may be inconsistent with the community's interests.

Fires initiated by vegetation falling on electricity distribution assets could be included in the f-factor scheme if the incentive rate is low and they are included in both the baseline and the annual monitoring and reporting.

In some cases the cause of the fire is clear but in other cases the cause is disputed. Fires for which the cause is disputed could not be included in the f-factor scheme until any civil or criminal action is concluded, which could take several years. If all fires are included in the fire incentive scheme and the incentive rate is low, this delay would have an immaterial impact on the operation of the fire incentive scheme. However, if fires are defined more narrowly for the purposes of the f-factor scheme and the incentive rate is correspondingly high, this delay would have a material impact on the operation of the fire incentive scheme.

Comment is sought from stakeholders on the appropriate definition of fire starts for the purposes of the f-factor scheme. Should all fires started by electricity distribution assets be included, or only fires that exceed some form of threshold? Should fires started by vegetation falling on electricity distribution assets be included in the f-factor scheme?

Definition of the days on which the fire incentive scheme is to apply

The f-factor scheme could apply on:

  • all days
  • days of total fire ban in any district
  • days of total fire ban in the districts in which a total fire ban day has been declared
  • days of catastrophic fire risk (Code Red days) only.

Applying the f-factor scheme to all days is consistent with an all hazards risk approach, which is consistent with contemporary best practice. Limiting the days on which the scheme applies biases the scheme towards the perceived high risk on a relatively limited number of days, influenced by the catastrophic bushfires on 7 February 2009. However, there are a range of catastrophic scenarios that could unfold on days that are not declared as Code Red days or as a total fire ban.

Applying the f-factor scheme to all days is administratively less complex and would be more straightforward for the AER to administer. If the scheme is applied on all days, the historical data will only need to be analysed to allocate fires to electricity distribution areas.

If the scheme is limited to a certain type of day, the historical fire start data will need to be further analysed to identify only the relevant data, in addition to allocating the fires to electricity distribution areas. For example, if the scheme is limited to the days of total fire ban in the districts in which a total fire ban day has been declared, the relevant data will need to be identified by considering whether a total fire ban day was declared in the district in which the fire occurred, and the electricity distribution area in which that part of the district is located.

The system for declaring total fire ban days has been in existence for a long period of time and so there should be a good time series of data available. By contrast, the concept of Code Red days is relatively new with a likelihood that the definition will change over time. It is unclear whether there is sufficient data available to be able to assess which historical data relates to days which would now be declared as a Code Red day. Additionally it is noted that there are nine districts for the purposes of fire danger ratings and only five districts for total fire ban days, and that the boundaries of these districts do not necessarily align.

Following the Ash Wednesday fires in 1983, the former State Electricity Commission of Victoria (SECV) began reporting on the number of fires associated with SECV assets on total fire ban days. The statistics for the period from 1982-83 to 1991-92 are provided in the graph below:

Graph showing the number of fires associated with SECV assets on total fire ban days. The statistics are for the period from 1982-83 to 1991-92. The statistics are highest in 1982-83 with over 160 fires from SECV assets. Since then, the number of fires associated with SECV assets has varied between almost none and nearly 40 reported for each period.

Data source: State Electricity Commission of Victoria, Annual Report 1991-92, page 93

The graph above illustrates the volatility from year to year in the number of fire starts associated with electricity distribution assets on total fire ban days. This volatility provides the opportunity for the electricity distributors to earn windfall gains (and losses) under the f-factor scheme if it applied on total fire ban days only, depending on the climatic conditions.

It is expected that the level of volatility arising from the f-factor scheme would be greater if the fire starts for the purposes of the fire incentive scheme were limited to Code Red days and would be less if all fire starts were included.

Comment is sought from stakeholders as to the days on which the f-factor scheme should be applied. Should it be applied to fire starts on all days, on total fire ban days only or only on Code Red days?

Fire start benchmark or target

The benchmark or target number of fires could be set by having regard to the number of fires that have started historically, or alternatively, a reduction in the number of fire starts over time.

The targets for the AER's Service Target Performance Incentive Scheme, on which the design of the f-factor scheme is based, are set "based on the average performance over the past five regulatory years". A similar approach could be adopted for the f-factor scheme whereby the target is set based on the average performance over the last five years. This is a relatively transparent and objective means of setting the target.

Alternatively, a reduction in the number of fire starts could be targeted. If this approach was to be adopted, there would need to be an objective means for determining the reduction required which takes into consideration the particular circumstances of each of the electricity distributors. Importantly, an expenditure allowance would need to be provided to the electricity distributors to deliver this reduction.

Regardless of the approach adopted, the setting of the target would need to be consistent with the definition of a fire start for the purposes of the f-factor scheme.

Comment is sought from stakeholders as to how each electricity distributors' target should be set for the purposes of the f-factor scheme.

Fire start incentive rate

The fire start incentive rate could be set by having regard to the marginal cost to reduce the likelihood of electricity distribution assets starting fires or on the value that the community places on reducing the likelihood of electricity distribution assets starting fires. However, currently this data is not readily available.

Alternatively, the fire start incentive rate could be set based on the amount of revenue to be placed at risk until further information is available.

If it is assumed that fires cause around $5 billion in losses approximately every 30 years and electricity distribution assets cause between 1 and 1.5 per cent of fires, then electricity distribution assets contribute to around $50 to $75 million of losses every 30 years. This equates to around $1.7 to $2.5 million of losses per year.

If it is assumed that electricity distribution assets start around 800 fires per year, 200 fires per year that are considered to be a serious electrical incident and 20 fires on days of total fire ban, and the f-factor scheme reduces the number of fires by 10 per cent, this equates to a penalty rate of around $20,000 to $30,000 for all fires, $80,000 to $100,000 for serious fires and $0.8 to $1.0 million for fires on total fire ban days.

A single penalty rate could be set that applies for all electricity distributors on all days. Alternatively different penalty rates could be set based on the level of fire risk associated with that day, or for different electricity distributors.

A single penalty rate would be simpler to administer.

If the penalty rates were different on different types of day, there would be greater attention by the electricity distributors on days when the penalty rate is higher. This is inconsistent with an all hazards approach.

If the penalty rates were different for each electricity distributor, it could be inferred that the community places a different value on fires started in different electricity distribution areas.

The head of power provided in the legislation for the Order in Council allows the penalty rate to be varied over time. If it is considered that more data is required before introducing different penalty rates, the different penalty rates could be introduced in a later regulatory period.

Comment is sought from stakeholders on the appropriate penalty rate that should be applied under the f-factor scheme. Should there be a single penalty rate on all days for all electricity distributors, or should the penalty rate vary based on the day and the electricity distribution area?

Role of the AER

The legislation requires the AER, as the national economic regulator, to make the f-factor scheme determination and the f-factor amount determination. As this provision is included in Victorian legislation, the f-factor scheme applies only to the Victorian electricity distributors and not to electricity distributors in other jurisdictions that are regulated by the AER.

The AER will require data on the number of historic fire starts to determine the benchmark or target number of fire starts as part of the f-factor scheme determination and to fire starts in each year of the regulatory period as part of the f-factor amount determination.

Section 16G of the Act provides broad information gathering powers to the AER to request fire start data from relevant entities, in addition to its existing powers to issue regulatory information instruments to request fire start data from the electricity distributors. For the purposes of the f-factor scheme, a relevant entity is defined in section 13 of the Act to be Energy Safe Victoria, and the Victorian fire agencies (the Department of Sustainability and Environment, the Country Fire Authority and the Metropolitan Fire and Emergency Services Board).

In addition, it is proposed that the electricity distributors be required to report fire start data to the AER on an annual basis, similar to the way they currently report service data.

Consistent with good regulatory practice, it is proposed that the AER should undertake consultation in making its f-factor scheme determinations and f-factor amount determinations. For consistency, the AER could use the existing distribution consultation procedures as set out in rule 6.16 of the NER (with relevant amendments).

It is further proposed that all f-factor scheme determinations and f-factor amount determinations made by the AER should be published.

As indicated previously, it is expected that the first f-factor scheme determination would be made in 2011 and would apply during the 2011-15 regulatory period. The first f-factor amount determination is expected to be made in 2012 in respect of the 2011 performance with the pass-through amount applied to the distributors' revenues in 2013.

To allow sufficient time for the AER to consult on its first f-factor scheme determination, it is proposed that this determination be made by 30 November 2011.

The f-factor amount determinations will need to be made after the fire start data is provided by each electricity distributor to the AER for the previous year and before the network tariffs are proposed to the AER by the end of October for the subsequent year. It is therefore proposed that the electricity distributors be required to report on the number of fire starts in each year by March 31 for the preceding year and that the f-factor amount determinations be made by the end of September in that year, for application to the network tariffs in the following year.

Comment is sought from stakeholders on the role of the AER in the making, publishing, implementing and administering of the f-factor scheme. Are the timeframe, consultation procedures and publication requirements proposed appropriate? Does the AER have sufficient information gathering powers for the purposes of making the f-factor scheme determination and f-factor amount determination?

Footnotes

  1. The Energy and Resources Amendment Act 2010
  2. This may be offset by the receipt of a Guaranteed Service Level (GSL) payment if the customer receives reliability worse than the thresholds set for the GSL payments scheme.
  3. Guidelines for Reporting Electrical Incidents (Electricity Safety (Network Assets) Regulations 1999), ESV, May 2001
  4. An all hazards risk approach is one in which all hazards are considered and prioritised as part of a risk management approach.
  5. Australian Energy Regulator, Electricity distribution network service providers: Service target performance incentive scheme, November 2009, section 3.2.1(a)
  6. For example, if the scheme covers fires on all days, there are approximately 800 fires per year. If the number of fires reduces by 10% as a result of the scheme, this equates to 80 fewer fires. If the losses are approximately $1.7 to 2.5 million per year, this equates to $21,250 to $31,250 per fire, or approximately $20,000 to $30,000 per fire.
  7. Either a Regulatory Information Notice (RIN) or a Regulatory Information Order (RIO)

Download the PDF version of this document

Page last updated: 09/06/17