Climate Change Authority

You are here

Design choices

Chapter 5 sets out the Authority’s analysis of the design of a light vehicle emissions standard for Australia. This appendix outlines the underlying analysis and evaluation of the policy design options:

  • C.1—Coverage and liability
  • C.2—Standard design and measurement
  • C.3—Timing and compliance.

As outlined in Section 5.1, the Authority used a simple framework to evaluate the design options:

Environmental effectiveness—the standard should ensure that the emissions intensity of new light and commercial vehicles is reduced. The standard should contribute to the overall reduction of transport emissions intensity.

Administrative and regulatory burden—the standard should be low cost, and simple for government to administer and for industry to comply with. It should draw on existing governance and regulatory structures where possible.

Equity—the standard should ensure, to the extent possible, equity in the compliance burden placed on manufacturers with a diverse product mix.

Policy stability and credibility—the standard should minimise opportunities for gaming, avoidance and market distortions. Participants and the wider public should have confidence in the standard.

C.1 Coverage and liability

C.1.1 Coverage

There are three key design questions about the application of the emissions standard to the light vehicles class:

  • Will the standard apply to all light vehicles or only passenger vehicles?
  • If it applies to all light vehicles, will there be a single standard or split standards for passenger vehicles and light commercial vehicles?
  • Will the standard cover second-hand vehicles imported into Australia?

C.1.1.1 Type of light vehicles

As discussed in Section 2.3, the light vehicles class includes both passenger vehicles (cars, sports utility vehicles, light buses) and light commercial (goods-carrying) vehicles (utilities, light trucks, vans).

Passenger vehicles account for 80 per cent of new light vehicles in Australia and are responsible for the majority of CO2 emissions (see Chapter 2). Light commercial vehicles comprise the remaining 20 per cent of new vehicle sales. Light commercial vehicles travel greater distances than passenger vehicles, estimated to be 28 per cent more on average (ABS 2013), with vehicle kilometres travelled projected to grow more than twice as fast as passenger vehicles to 2020 (BITRE 2009).

Light commercial vehicles typically represent a larger, heavier and more powerful segment of the vehicle market and have, on average, higher rates of fuel consumption than passenger vehicles. In some circumstances, light commercial vehicles may face greater challenges to deliver better fuel economy and lower emissions than passenger vehicles. For example, the functional requirements of light commercial vehicles (particularly light trucks) may limit the incorporation of fuel-saving technologies such as drag reduction. Converting vehicles to diesel is an emissions reduction opportunity but most light commercial vehicles already use diesel—82 per cent of light commercial vehicles sold in Australia in 2012 were diesel (FCAI 2013).

All light vehicle emissions standards applied in other countries cover passenger vehicles at the minimum, and most also cover light commercial vehicles. International evidence suggests that the most effective vehicle emissions standards have broad coverage (ICCT 2011a).

Limiting standards to passenger vehicles alone would reduce environmental effectiveness, compared to a standard with wider coverage. While light commercial vehicles comprise a significantly smaller proportion of new vehicle sales, their higher emissions profile and travel distances suggest that they should be covered by a standard.

There are no obvious barriers to implementing a light vehicle emissions standard for both passenger and light commercial vehicles. Emissions data on all new light vehicles—passenger and light commercial—is currently collected under the Australian Design Rule (ADR) 81/02 Fuel Consumption Labelling for Light Vehicles.

Conclusion

The standard should cover both passenger and light commercial vehicles.

C.1.1.2 Single or split standard

If all light vehicles are covered, a secondary question is whether to set a single standard encompassing all light vehicles, or to split the standard into two parts, with separate levels applying to passenger and light commercial vehicles.

In the EU, the US, China, Japan, Mexico and Canada, separate standards apply to passenger vehicles and light commercial vehicles, with the latter category having a higher (less stringent) numerical standard to meet than passenger vehicles. In part, this split is due to the history of the introduction of standards in these jurisdictions, where standards were initially applied to the largest group (passenger vehicles) and only later applied to light commercial vehicles.

The strongest argument for setting split standards for passenger and light commercial vehicles is to avoid a disproportionate burden on light commercial vehicle manufacturers. The weight of this burden largely depends on the product mix provided by the manufacturer, as the standard is for the average level of performance across all of a manufacturer’s sales mix of new vehicles.

  • The manufacturers of the 10 highest-selling Australian pick-up truck models in 2012, and of the five van or light truck models selling over 1,000 vehicles a year, all produce a significant number of passenger vehicles (NTC 2014). They can therefore meet a single standard by varying the relative mix of passenger and light commercial vehicles sold, as well as by improving performance of their light commercial vehicles.
  • The choice of size thresholds for the application of the standard will also have an influence (see C.1.2.2). In 2013, there was only one specialised light commercial vehicle manufacturer that sold over 2,500 vehicles that did not also manufacture passenger vehicles (NTC 2014).

This suggests that, with the proposed standard, light commercial vehicle manufacturers in the current Australian market would not be unduly burdened.

On the other hand, split standards increase administrative complexity (especially as the majority of manufacturers would have to meet two separate standards, rather than one), and potentially create an incentive for manufacturers to market light commercial vehicles (subject to a less stringent standard) into the passenger vehicle market.

A 2012 analysis recommended the continued separation of standards for passenger vehicles and light commercial vehicles in Europe for a range of reasons, but acknowledged that the split does increase the risk of manufacturers ‘gaming’ the system by marketing certain light commercial vehicles as passenger vehicle substitutes (TNO et al. 2012).

In responses to the 2011 DIT discussion paper, most respondents (including the vehicle industry) favoured a single standard covering all vehicles (DIT 2011b). The FCAI (2011a) noted that there is some substitution between commercial and passenger vehicles, that a single standard provides a consistent policy objective for both and that separate standards could increase the regulatory burden.

Conclusion

A single standard for both passenger and light commercial vehicles should apply.

C.1.1.3 Second-hand imports

A third question is whether standards should extend to imports of second-hand vehicles.

Second-hand imports are currently a very small segment of the Australian ‘new’ vehicle market, estimated to be less than three per cent of new vehicle sales (DIRD 2014d). Situations in which vehicles may be imported into Australia are tightly prescribed to ensure that vehicles meet safety and environmental standards, and current legislation appears to prevent large-scale imports. The Motor Vehicle Standards Act provides that applications for licence plates, or to supply an imported vehicle without licence plates, can only be made for a single used imported vehicle (sections 13C(2), 16(3)). Importers of second-hand vehicles thus tend to be individuals and small businesses, licensed automotive workshops restricted to fewer than 100 cars annually, and immigrants and returning expatriates importing personal vehicles.

No other vehicle efficiency standards currently apply to second-hand imports. This may reflect relatively limited importation of second hand cars in most jurisdictions. New Zealand includes second-hand imports in its fuel efficiency-labelling scheme.

Applying vehicle emissions standards to second-hand imports would marginally increase environmental effectiveness through increased coverage but could also significantly increase administration and compliance costs. In part, this is because existing fuel efficiency values for those imported cars that are derived from non-European test cycles could not simply be adopted into an Australian standard. Calculated fuel efficiency per kilometre differs between standards in separate jurisdictions, reflecting the different testing methodologies used. In New Zealand, where a significant proportion of imported vehicles are second-hand, a conversion formula is applied to second-hand Japanese imports that are pre-2008 models for the purpose of vehicle fuel efficiency labelling.

Even if second-hand imports were covered under the standard, it is very unlikely that individual suppliers would be liable due to low numbers of annual sales under the current import restrictions (see C.1.2.2 on threshold for liability).

The Productivity Commission has suggested that restrictions on large-scale second-hand imports be removed (PC 2014, pp. 100–102). If adopted, this change could potentially lead to a large increase in second-hand vehicle imports. In that case, both equity across suppliers and environmental effectiveness would suggest second-hand vehicles should be covered.

On balance, the very small increase in coverage from including second-hand imports does not appear to warrant the extra administrative costs of including them in the scheme at this stage. In the event that circumstances change and there is a significant increase in the quantity of vehicles imported, this issue should be reconsidered. Coverage could also be reassessed as part of the proposed 2021 review (section 5.3), taking account of market developments.

Conclusion

Second-hand imports should not be covered under the standard at this stage.

C.1.2 Liability

The liable entity is responsible for compliance with the light vehicle emissions standard, including reporting performance and paying any penalties for non-compliance.

The key design questions for determining the point of liability are:

  • Where in the vehicle supply chain (from manufacturer to retailer) should liability be placed?
  • What size threshold (defined by annual sales) for imposing liability should be applied?

C.1.2.1 Choice of liable entity

The Australian new vehicle market is dominated by a relatively small number of large vehicle manufacturers, with the top 10 manufacturers responsible for approximately 80 per cent of new vehicle sales in 2012. A further 37 manufacturers accounted for the remaining 20 per cent of vehicle sales (see Figure C.1). As noted in Chapter 3, it is expected that there will be no vehicle manufacturing operations in Australia by 2018, with all new cars imported. As only around 10 per cent of vehicles sold are currently domestically manufactured, there is no reason to expect that this will fundamentally alter the broad market structure.

Figure C.1: Light motor vehicle sales by manufacturer 2012

Figure C.1 is a pie chart of light vehicle sales in Australia by supplier in 2012. It shows that the new light vehicle market was dominated by a relatively small number of large vehicle suppliers, with the top 10 suppliers responsible for approximately 80 per cent of new vehicle sales. The supplier with the largest share is Toyota (with 20% of sales), followed by Holden (11%), Mazda (10%) and Hyundai (8%). A further 37 suppliers accounted for the remaining 20 per cent of vehicle sales.

Source: Climate Change Authority based on NTC 2013

The larger vehicle manufacturers responsible for the bulk of vehicle sales use integrated supply chains, which encompass manufacture, import and retail sale. Business models for other vehicle suppliers are more varied. In some instances, authorised importers supply independent retailers; in others, independent retailers directly import vehicles. All importers must comply with government requirements under the Motor Vehicle Standards Act.

Practical considerations are important. The liable entity should be able to comply with the standard reporting requirements (and be penalised in the case of non-compliance) and be the entity able to respond to standards by altering its vehicle mix. For this reason, retailers are clearly not an appropriate point of liability—they are far more numerous and varied in structure than large manufacturers and, importantly, less able to respond to the standard by controlling product mix. Accordingly, manufacturers (or their importing agents) are likely to be a better choice for point of liability.

Several submissions to the 2011 DIT discussion paper suggested the entity responsible for certifying new vehicles for the Australian market under the Motor Vehicle Standards Act (the MVSA certifying entity) should be the point of liability (see, for instance, the Australian Automobile Association 2011, Ford Australia 2011 and Honda Australia 2011). This is broadly consistent with US and EU standards, which both hold either the domestic manufacturer or a licensed importer responsible for ensuring compliance with relevant environmental and safety regulations.

The MVSA certifying entity already has a legal relationship with the Commonwealth, and is required to submit detailed technical information on vehicle design and safety as part of the approval process for new vehicles entering the Australian market. In the case of larger manufacturers, the certifying entity is likely to be either the manufacturer or closely related to the manufacturer, so obligations could be effectively passed through by contractual or other arrangements.

Smaller manufacturers may contract independent agents for certification, and have less of a business presence in Australia, making it less clear that the licensing entity will be able to influence the vehicle sales mix. These arrangements could be considered further as part of a RIS process.

At this stage, the MVSA certifying entity appears to be an appropriate point of liability for vehicle emissions standards. This should lead to a relatively small number of liable entities, which in most cases will be closely related to the vehicle manufacturer and have the technical capacity to comply with reporting obligations.

Conclusion

Subject to further consultation with industry, the liable entity under the standard should be the same entity responsible for Australian certification of a vehicle under the Motor Vehicle Standards Act 1989 (Cth).

C.1.2.2 Threshold for liability

A threshold for liability is an important design feature to reduce compliance and administration costs. Some form of size threshold is used in all overseas schemes; direct comparisons are complicated by differences in the overall size of new vehicle markets. The US applies less stringent transitional standards to manufacturers with fewer than 50,000 annual sales, and manufacturers with fewer than 5,000 sales worldwide can apply for firm-specific standards. The EU also applies several threshold levels, with those with under 1,000 annual sales exempted altogether, and those between 1,000 and 10,000 able to apply for firm-specific standards.

Selecting a size threshold requires balancing the improved environmental effectiveness and improved equity of a lower threshold against the increased regulatory burden of imposing liability on more and smaller entities. Any threshold will invariably raise boundary issues, with the potential for entities near the threshold to alter activity levels to avoid liability. A more limited liability could be imposed on smaller entities to reduce costs and risks of gaming, although this would increase regulatory complexity.

An important consideration is the market structure of liable entities and how the point of liability is determined (discussed in C.1.2.1). Many vehicle brands may be linked into a larger corporate group, and may or may not operate as distinct legal entities. The practical implications of the threshold level therefore interact with selection of the point of liability.

Figure C.2 sets out the distribution of Australian car sales by make under 40,000 annual vehicle sales and Table C.1 sets out the implications of different thresholds, based on 2012 sales volumes.

Figure C.2: Light vehicle sales under 40,000 vehicles by make in 2012

Figure C.2 is a bar chart showing sales figures by make for supplierswith relatively low sales. It shows there is a long ‘tail’ of smaller makes that accounted for fewer than 2,500 sales in 2012, including Saab, Lotus, Ferrari and Porsche. Makes such as Volvo sold about 5,000 vehicles; Mercedes-Benz about 20,000; and Honda about 35,000.

Source: Climate Change Authority based on NTC 2013

Table C.1: Implications of selected thresholds (based on 2012 light vehicle sales)

Threshold (vehicles sold) Number of vehicles not covered Percentage of vehicles not covered Number of uncovered makes (out of 47 total)
100 524 >0.01 9
500 923 0.09 12
1,000 3,784 0.4 16
2,500 7,521 1.2 23
5,000 23,317 2.1 25
10,000 56,313 5.2 30

Note: This assumes each make operates as a separate liable entity. In practice, some small makes may be part of a larger corporate group; this would reduce the number and percentage of vehicles excluded by the threshold.
Source: Climate Change Authority based on NTC 2013

Figure C.2 shows that there is a long ‘tail’ of smaller makes that account for a very small proportion of sales. As set out in Table C.1, thresholds could be set to eliminate a large number of makes with minimal effect on coverage. A 500-vehicle threshold would exclude about a quarter of makes from liability while diminishing coverage by less than a 10th of one per cent; a 1,000 vehicle threshold would exclude about one-third of makes at the expense of 0.4 per cent of coverage; a 2,500 vehicle threshold would exclude about half of the makes and reduce coverage by 1.2 per cent.

Ideally, the threshold would be set at a level that minimises the number of makes near to the threshold (which might offer an incentive to ‘game’ it to avoid liability, including through disaggregating brands covered by a corporate group that might otherwise operate as a single entity for the purposes of the standards). The distribution of sales is, however, relatively uniform with no obvious gaps.

On balance, a threshold of 2,500 vehicles would appear to provide an appropriate balance between compliance costs and coverage. Based on current sales, this threshold would exclude about half of the makes but only reduce coverage by about 1.2 per cent. This threshold could be reviewed when considering the second phase of the standard, to address any distortionary market responses if they emerge.

Conclusion

Subject to further consultation and consideration of how the point of liability will be determined, the threshold for liability should be annual sales of 2,500 vehicles.

C.2 Standard design and measurement

Determining how the standard applies to manufacturers of new light vehicles raises two related design choices:

  • Should a flat or attribute-based standard be applied?
  • If an attribute-based standard is favoured, which is the most appropriate attribute to adopt?

This section also discusses a number of measurement and scope issues for how emissions will be measured under the standard:

  • whether the standard should be based on fuel consumption or CO2 emissions
  • what test procedure should be used
  • whether multipliers, which recognise specific low-emissions technologies or fuels, should be part of the scheme
  • whether off-cycle credits, which recognise emissions reductions not captured by the standard test, should be part of the scheme.

C.2.1 Form of standard

The most common forms of light vehicle emissions standards that have been evaluated internationally are:

  • a flat standard for the fleet (or sections of it), usually an absolute cap or uniform percentage reduction of emissions intensity, which applies to every manufacturer
  • an attribute-based fleet-average standard, where the level of the standard varies with an attribute of the vehicle (typically vehicle mass or size).

In determining the best option, a reasonable starting point is to consider the simplest model possible that delivers significant emissions reductions, is cost-effective to administer and is equitable across manufacturers. The selected approach also needs to be objective and transparent so that liable entities clearly understand their obligations.

The simplest approach—to set a flat (absolute) target(s) for the fleet, or categories of the fleet, or to apply a uniform percentage reduction on emissions—imposes the same requirements on every manufacturer, regardless of their mix of vehicles. While this may appear fair, a ‘one-size-fits-all’ approach can disadvantage manufacturers at both ends of the emissions spectrum and reduce consumer choice. Different manufacturers produce a heterogeneous mix of models and have different starting positions linked to previous investments in fuel economy and reducing emissions. Applying a uniform percentage reduction target to a manufacturer who has already invested heavily in emissions or fuel consumption reductions will put it at a competitive disadvantage relative to a company that has not previously focused on this aspect. Conversely, applying the same flat standard to all manufacturers could force one with larger vehicles sitting above the standard to remove certain models from its range (even if such models are relatively efficient for their size or are important for their commercial viability and are strongly favoured by consumers).

The alternative approach is to implement a sales-weighted fleet-average standard. The target emission level varies across manufacturers, in light of their product mix. The standard is defined by the relationship between the CO2 emissions or fuel consumption of a vehicle and an objective attribute of the vehicle such as mass or size. Attribute-based standards enable manufacturers to supply vehicles above the target level of the standard, provided they are offset by sufficient sales of vehicles that are below the target. The International Council on Clean Transportation (ICCT) notes that such standards enable a manufacturer to market vehicles that ‘… remain diverse in terms of vehicle shape, size and functionality and to improve efficiency without compromising vehicle functionality’ (Mock 2011).

International assessments in both the US and EU have strongly favoured attribute-based standards. Benefits include encouraging emissions improvements across the full range of vehicle types, spreading the regulatory burden across all manufacturers and respecting consumer choice (US EPA and NHSTA 2011b). A useful overview of the EU’s assessment of the various approaches is summarised in the 2011 DIT discussion paper (DIT 2011a). Similarly, a US EPA and NHTSA evaluation in support of the US 2017–25 standards also identified multiple benefits from attribute-based standards (compared to class-based caps or uniform percentage reductions).

All countries that have adopted mandatory fuel consumption or CO2 standards have included an attribute adjustment, but not all have included fleet-averaging. For example, China applies the averaging across specified categories of vehicles, not the fleet as a whole. However, soon all four major markets (the US, the EU, China and Japan) will take a flexible corporate-average approach to standards, with Japan switching to fleet-averaging for its 2020 target. In submissions to the 2011 DIT discussion paper, there was overwhelming support for attribute-based fleet-average standards, including from the vehicle industry.

If an attribute-based standard is favoured, a decision needs to be made on the most appropriate attribute to adopt. To date, the attributes used internationally are mass or vehicle size, usually measured as the ‘footprint’ of the vehicle (the size of the vehicle determined by the product of the vehicle track width and the wheelbase—which is the distance between the two axles).

Footprint is used in the US, Canada and Mexico, with mass adopted in the EU, China, Japan and Korea. International assessments conclude both can work effectively and impose similar costs but, on balance, the evidence favours footprint as the best option (TNO 2011; German & Lutsey 2011; US EPA and NHSTA 2011b).

The key advantage of footprint is that it encourages manufacturers to improve efficiency by reducing vehicle mass (‘light weighting’). Light weighting is a major emissions reduction strategy in new vehicle design, and vehicles can be light-weighted without compromising vehicle functionality from the consumer’s perspective. Mass-based standards discourage light weighting, as they require lighter vehicles to meet more stringent average emissions targets. A recent ICCT assessment (2011a) argues strongly against mass-based standards, which typically shift fleets towards bigger or heavier models. In addition, size-based standards tend to encourage better safety design than weight-based standards. This was one of the key factors in NHTSA’s decision to adopt footprint-based standards instead of weight-based standards for the US 2008–11 light truck Corporate Average Fuel Economy rule (US EPA and NHTSA 2011a).

In submissions to the 2011 DIT discussion paper, there was broad support for footprint-based standards. However, vehicle manufacturers at the time were split on the issue, with most favouring mass-based standards because they were used more frequently in the standards applying in source countries.

There is good evidence, and widespread support, for the adoption of attribute-based fleet-average standards compared to any alternative approach. Such standards maximise equity and flexibility for manufacturers and preserve consumer choice. While attribute-based standards using either mass or footprint can be effective, the balance of the evidence supports a footprint-based standard. There is no evidence to indicate that this would disadvantage suppliers from markets where mass-based standards (including the EU, Japan, Korea and China) are in place.

Conclusion

The standard should apply to fleet-average emissions and be based on vehicle footprint.

C.2.2 Measurement and scope

C.2.2.1 Basis for measurement—fuel consumption or emissions

Standards can be based on fuel consumption per kilometre travelled or on CO2 emissions per kilometre travelled. These metrics are directly related; combustion of fuel leads to emissions of CO2 (noting that different types of fuel have different emissions profiles). Reducing consumption of fuel per kilometre will therefore lead to corresponding decreases in emissions per kilometre.

Internationally, the US and Republic of Korea use both fuel economy and CO2 emissions standards. The EU uses CO2 emissions standards. Japan and China use fuel economy. In Australia, the existing ADR81/02 collects both CO2 emissions and fuel consumption data at a model-specific level (see C.2.2.2).

An emissions-based standard is preferable where the primary objective is to reduce emissions. Fuel economy improvements do not always give an equal emissions intensity improvement, as emissions rates from different fuels vary. For instance, depending on driving conditions and engine performance, diesel engine vehicles can be up to 30 per cent more fuel-efficient than comparable petrol vehicles. However, diesel consumption results in about 15 per cent more CO2 being emitted per litre of fuel (calculated based on NTC 2012, p. 3). Overall, a shift to diesel reduces emissions by about 15 per cent relative to petrol.

However, consumers could more easily understand a fuel economy standard than an emissions-based standard, and it provides a better basis for aiding consumer purchasing decisions. Vehicle fuel economy labelling schemes, including in both the US and EU, tend to include both emissions and fuel economy data.

A further question is whether to base the standard on CO2 emissions only or to include emissions of other greenhouse gases from vehicles. These could include nitrous oxide exhaust emissions from the combustion of fuel and emissions of hydrofluorocarbons (HFCs) from vehicle air conditioning systems. Overseas schemes do not generally include these emissions directly, although they may be considered in the calculation of off-cycle credits (see C.2.2.4).

The primary objective of the standard suggests that it should be based on CO2 emissions rather than fuel economy. Emissions of other greenhouse gases are very small compared with CO2 emissions from a vehicle over its lifetime, and are unlikely to warrant the extra effort and complexity of inclusion. In addition, the inclusion of other gases would require every vehicle model to undergo additional Australia-specific testing (CO2 is the only greenhouse gas directly measured in the standard emissions test that underpins ADR81/02).

Conclusion

The standard should be based on CO2 emissions. Other greenhouse gases should not be included in the standard.

C.2.2.2 Test procedure

To minimise administrative complexity and reduce compliance costs, existing testing procedures, if appropriate, should be used wherever possible. In Australia, the CO2 emissions value for each vehicle model (and its variants) is already collected under ADR81/02 as part of the vehicle type approval certification process under the Motor Vehicle Standards Act. This applies to all new vehicles to be sold in the Australian market. This test can supply the basic data for calculating the CO2 emissions targets under vehicle emissions standards—the tailpipe CO2 emissions produced by the vehicle (in grams of CO2 per kilometre travelled) is combined with annual sales data to determine the liable entity’s compliance requirement.

The ADR81/02 data does not represent all ‘real-world’ driving and does not take into account the non-road CO2 emissions from the production and supply of various transport fuels (including electricity for electric vehicles). International research on testing procedures that provide robust data for life-cycle emissions for all fuel types is currently underway, but remains at an early stage.

The data currently collected under ADR81/02 is robust, verifiable and comparable, and is the only such data available at the individual model or variant level for all light vehicles. It is internationally recognised and already used for Australia’s CO2 labelling requirements. The CO2 standard would not require any additional vehicle testing if ADR81/02 is accepted as the data source.

There was broad support for this approach in submissions to the 2011 DIT discussion paper (DIT 2011a).

Australia is committed to matching United Nations regulations, through which a new harmonised testing procedure is being developed. This test may help to reduce the gap between real-world and tested performance, and is likely to be adopted in the EU from 2020 onwards (ICCT 2013a). Developments in the adoption of this procedure will need to be monitored and any transition arrangements considered in setting the first phase of the standard.

Conclusion

The standard should use the CO2 emissions value collected under ADR81/02 Fuel Consumption Labelling for Light Vehicles.

C.2.2.3 Multipliers

Many countries allow manufacturers to reduce their reported average emissions by using ‘multipliers’. Multipliers can be awarded to vehicles that satisfy low-emissions benchmarks or utilise specific technologies or fuels claimed to reduce CO2 emissions relative to conventional vehicles.

The principal rationale for multipliers is to encourage innovation and early deployment of advanced (often high-cost) low-emissions technologies such as electric vehicles (EVs). Multipliers can act as an additional incentive to innovate and outperform standards.

Crediting arrangements such as multipliers can contribute to the environmental effectiveness of the scheme over the medium term. While multipliers will lead to an increase in overall fleet CO2 emissions in the short term (as more credits are awarded for the same amount of emissions reductions), if carefully designed they may have beneficial effects in the longer term as lower emissions technologies are more rapidly deployed. Multipliers are likely to involve a minor increase in administrative effort to design as an element of the scheme and, on an ongoing basis, to assess for each liable entity choosing to utilise them.

Multipliers can be implemented in two ways. The first is to provide multipliers for specific technologies or fuels that are claimed to reduce CO2 emissions relative to conventional vehicles. This requires governments to choose particular technologies or fuels for eligibility at a certain point in time.

The second approach provides for more equitable treatment of technologies by setting an emissions performance benchmark. This provides multipliers for vehicles below a specific emissions intensity level, regardless of the technology or fuel used. Benchmarks of 50 g/km and 100 g/km were raised by industry in response to the 2011 DIT discussion paper (DIT 2011a).

On balance, multipliers are not considered a necessary option to pursue at this stage. A fleet-average standard already creates a direct incentive for innovation—producing very low-emissions vehicles makes it easier for manufacturers to meet the fleet average. Other options to encourage innovation and performance beyond the standard are discussed in C.3. This could be considered for later phases of the standard.

Conclusion

Multipliers are not necessary to the functioning of a vehicle emissions standard and are not proposed at this time.

C.2.2.4 Off-cycle credits

Off-cycle credits are primarily designed to recognise technologies that can deliver actual on-road CO2 emissions reductions but are not ‘captured’ by the standard tailpipe emissions test used for compliance. Such credits may include, for example, measures for the more efficient operation of vehicle air conditioners, which are standard on most new vehicles but turned off during the standard test cycle.

Off-cycle credits could contribute to the overall reduction of transport emissions by providing additional incentives to reduce all emissions associated with real-world vehicle use. If credited, they may also provide a more cost-effective way for manufacturers to reduce emissions than measures that are directly assessed through standard tailpipe emissions testing. However, off-cycle credits would increase the administrative and regulatory complexity of the scheme

A central issue with off-cycle credits is the design of objective, repeatable methodologies and processes to determine and validate the claimed additional CO2 benefits. Internationally, both the US and EU have attempted to recognise and quantify the CO2 benefits from off-cycle technologies within their standards. Both place the burden for demonstrating off-cycle credits with the liable entities.

The EU provides procedures for approving and certifying ‘innovative technologies’ not captured by the standard test cycle and assessing their CO2 emissions benefits. In the US, the EPA and NHTSA have undertaken an extensive analysis covering air conditioners and a broad range of off-cycle technologies including high-efficiency lighting and engine heat recovery.

The US work has developed a ‘menu’ of technologies assessed as providing real-world CO2 benefits, which assigns default CO2/mile credit values for each. This approach reduces the need for extensive testing, and uses analysis and simulations rather than full vehicle testing as much as possible. Both the US and EU apply a cap on the maximum overall fleet benefit a manufacturer can claim for innovative or off-cycle technologies. This recognises, in part, the inherent uncertainty of a general assessment of off-cycle performance as opposed to testing the individual vehicle models (US EPA and NHTSA 2011a) (EU Regulation 725/2011).

The EU adopts a stricter approach in only considering off-cycle technologies that are deemed innovative and are intrinsic to the transport function of the vehicle (excluding accessory functions) (EU Regulation 725/2011). In practice, while a small number of eco-innovations have been approved, feedback from stakeholders indicates that implementation has proven difficult.

The FCAI submission to the 2011 DIT discussion paper (FCAI 2011c) acknowledges the fundamental burden of demonstrating additional off-cycle benefits should rest with individual manufacturers and that there is a need for a system of rigorous assessment and validation. While pre-existing international methods could be adapted for the Australian context, the emissions reductions recognised by off-cycle credits vary as a function of driving behaviour, congestion, road infrastructure, speed limits and ambient temperature, and therefore differ significantly from country to country. Adapting methods would require significant effort and involve a considerable administrative burden in both design and ongoing assessment for the standards.

On balance, off-cycle credits are not considered a necessary option to pursue at this stage, due to the significant administrative and regulatory burden they would impose to design and implement. However, there may be merit to off-cycle credits and their inclusion in later phases of the scheme could be considered as part of the proposed 2021 review.

Conclusion

Off-cycle credits are not necessary to the functioning of a vehicle emissions standard and are not proposed at this time.

C.3 Timing and compliance

C.3.1 Timing

Introducing a light vehicle emissions standard to Australia will require time for detailed policy development, stakeholder consultation, and the establishment of monitoring and reporting processes. The key timing decisions are:

  • start year
  • length of first phase.

 

C.3.1.1 Start year

The first timing decision is the appropriate start year for standards. Greater environmental and economic benefits will be achieved by introducing light vehicle emissions standards early. This needs to be balanced against providing an appropriate lead time to allow for industry consultation, the consideration and development (if required) of an appropriate legislative framework, and the establishment of monitoring and reporting processes.

Early introduction of vehicle standards would increase the fuel savings and emission reductions available to Australia. Results of the modelling conducted by the CSIRO for the Authority (discussed in Appendix B) show timing is a key determinant of benefits. Over the period to 2050, lenient standards introduced in 2018 are projected to deliver greater emission reductions than stringent standards introduced in 2025 (CCA 2014a). Early adoption of a standard maximises the benefits—it takes time for changes to new vehicles to improve the fleet overall. A strong standard starting in 2018 generates the greatest emissions reductions and the greatest financial benefits to Australian motorists.

As discussed in Chapter 1, the introduction of a light vehicles emissions standard is not a new concept in the Australian context, and significant work has already been done to both explore an appropriate design and consult with industry. In addition, as discussed in Chapter 4, the proposed target for Australia will not be more stringent than in other key economies. Australia will import all its new vehicles by 2018, and currently lags behind other major markets (including most of our major suppliers). This means that the required adjustment for manufacturers will be a choice about which models and model variants are supplied to the Australian market, rather than necessitating fundamental design and product changes. It suggests that lead times for a standard could be relatively short.

Internationally, lead times of about three years for the initial introduction of vehicle emissions standards are common. The EU 2012–15 standards and target for 2020 were announced in April 2009 (European Council 2009), while the US 2012–16 vehicle standards were announced in May 2009. Both have large domestic car manufacturing industries that would need to have made adjustments to comply with the new standards.

Best practice would suggest that two years is sufficient for policy planning and development (IEA 2012b). As outlined below, Australia has much of the required measurement and reporting in place already, so is well placed for rapid implementation.

A start year of no later than 2018 should therefore provide for adequate consultation and an orderly phase-in of new light vehicle emission standards in Australia. This provides a three-year lead time if a policy decision is taken in 2015.

Conclusion

The new light vehicle emissions standard should commence no later than 2018.

C.3.1.2 Length of first phase

The second timing decision is the length of the first phase of standards (phase one). The period needs to be long enough to allow liable entities to adjust their business operations, but short enough to avoid ‘locking in’ standards that prove inappropriate due to technology developments, market changes or other factors.

Internationally, compliance periods have tended to range between four and seven years. The EU and US currently have targets out to 2020 and 2025 in place under their emissions standards, and China has proposed a target to 2020.

Aligning the first Australian standards to major jurisdictions could assist with future global harmonisation, possibly simplifying the compliance process for manufacturers. With a proposed start year of 2018, ending phase one in 2020 would appear too short. A 2025 end date, however, provides a reasonable first phase (2018–25) of eight years.

This would not set a binding precedent for future phases. For example, the time span for phase two could be shorter (such as 2026–30).

Conclusion

The first phase for the new light vehicle emissions standard should be 2018 to 2025.

C.3.2 Compliance options

The key design options for compliance with the standard are:

  • whether compliance is required on an annual or a periodic basis
  • what flexibility mechanisms should be allowed to enable liable entities to cost-effectively comply with the standard
  • the frequency and start date for reporting obligations
  • what form of penalties should apply for non-compliance.

C.3.2.1 Annual or periodic compliance

Standards can require annual (frequent) or periodic (infrequent) compliance. The choice of approach has a strong relationship to what flexibility mechanisms, such as banking, borrowing or trading, are allowed within the scheme (discussed in C.3.2.2).

Annual compliance requires manufacturers to meet a set target each year, and thus drives early and progressive efforts to reduce emissions. Internationally, the US is the only major market that has mandatory annual targets. While this may increase compliance costs, mechanisms such as banking and borrowing that allow for normal business ebbs and flows can enhance flexibility for manufacturers and minimise any costs. The US allows for banking, borrowing and trading.

Periodic compliance entails manufacturers meeting a set target by a fixed future year and does not mandate an annual rate of improvement. Internationally, the EU and Japan have targets for 2015 and 2020 with no interim targets (ICCT 2014).

Annual compliance is likely to drive greater environmental effectiveness as it ensures fleet performance improves each year. Periodic compliance will have a lower administrative burden, but runs the risk of liable entities only striving to improve the emissions of vehicles sold in the final year of the period. Further concerns with periodic compliance include a risk of suppliers lobbying for target revisions, which, if successful, would unfairly disadvantage competitors who have already taken action to meet the standard; and the possibility of ‘fringing’ effects of new entrants and suppliers leaving the market and avoiding compliance altogether. Either of these could compromise policy stability and credibility and environmental effectiveness.

The extent of the administrative burden posed by annual compliance depends on the new reporting obligations introduced by the standard. As discussed in C.3.2.3, the proposed standard involves only a modest additional reporting obligation.

On balance, annual compliance is the preferred approach. It would encourage progressive improvement in fleet performance, and guard against lobbying and fringing effects. The additional administrative burden is likely to be very small.

Conclusion

The standard should set annual compliance obligations for liable entities.

C.3.2.2 Flexible compliance mechanisms—banking, borrowing and trading

Flexible compliance mechanisms provide liable entities with a range of options to cost-effectively comply with a given standard. They can improve the pace of progress in meeting given standards and assist in driving emissions improvements, while allowing flexibility in year-to-year performance.

Banking allows liable entities to save credits generated by overachieving against their targets and use them for future compliance, either within a compliance period (for example, phase one) or between periods (for example, between phase one and phase two). Similarly, borrowing allows liable entities to use credits from future periods to meet current compliance obligations. Trading allows for the movement of credits between liable parties.

In determining whether to allow one or more of these options, the starting point, as outlined in Chapter 5, has been to consider the vehicle emissions standard design for Australia that maximises emissions reductions while ensuring the cost-effectiveness, equity and credibility of the scheme for consumers and manufacturers.

As discussed in C.3.2.1, decisions on banking, borrowing and trading are closely linked with decisions on the type of compliance and the compliance period. For example, banking and borrowing can reduce the costs of annual compliance by allowing year-on-year flexibility to account for normal business ebbs and flows. They are less relevant in a scheme with periodic compliance.

Flexibility mechanisms operating within a compliance period; for example, phase one (2018–25), and between phases (pre- and post-2025) may have different impacts on the integrity of the scheme. If the Australian standard was set at a level significantly less stringent than other markets, especially in the first phase, and banking or trading across phases was allowed, liable entities could potentially establish large volumes of credits in the early years with relatively little effort. This would significantly reduce the need to act in later phases as the standards get tighter, thereby diluting the environmental effectiveness of future standards. If, however, Australian standards are on par with international standards, this is less of a concern.

Flexibility within phase one does not create the same risks to environmental effectiveness, as entities would be obliged to meet the given standard within the time frame specified.

Borrowing within phase one increases flexibility but could create risks of non-compliance in future years, or incentives to lobby to weaken the standard. This would reduce the credibility of the scheme. Borrowing from future phases could exacerbate credibility concerns. These risks can be managed by imposing limits on borrowing and restricting it to within the phase.

Trading increases flexibility and provides an incentive for performance beyond the standard. While it is difficult to predict the likely market for trade of excess credits between liable entities in the Australian context, international experience and feedback from domestic stakeholders suggests uptake may be limited. This suggests the benefits of a bespoke trading mechanism within the standard are unlikely to justify the associated administrative complexity and cost.

There may be scope to encourage performance beyond the standard through other mechanisms. For example, if a methodology could be developed to estimate and credit emissions reductions achieved through superior performance by a supplier, the ERF may be a suitable vehicle.

On balance, to give liable entities flexibility to meet their compliance requirements, banking and limited borrowing should be allowed within phase one. While trading could improve the cost-effectiveness of a standard, it does not appear warranted at this time.

Conclusions

Banking and limited borrowing should be allowed within phase one of the standard.

Trading is not necessary to the functioning of a vehicle emissions standard.

C.3.2.3 Timing of reporting obligations

Key considerations for reporting obligations are the start date and frequency of reporting.

Early introduction of reporting obligations (prior to the commencement of the standard) is likely to bring benefits. It will help liable entities track their position prior to facing formal compliance obligations, and make any necessary changes to their business operations. It also allows entities and administrators to test and refine reporting and monitoring systems.

Overall, with a proposed start date of no later than 2018, it would be worthwhile for reporting to begin two years prior, in 2016. In the event of delay, a one-year lead time for reporting would still be beneficial. Testing and refining reporting and monitoring systems could be prioritised in the policy development process, if necessary, to enable reporting to commence in 2016.

Annual compliance would clearly require annual reporting. Even if periodic compliance was preferred, annual reporting from 2016 would still be desirable as it would help identify suppliers above and below the minimum threshold for liability. Annual reporting would enable regulators, industry and policy makers to monitor performance against the standard and provide the necessary data to underpin any banking and borrowing provisions.

As discussed in Box 5.1, CO2 emissions, fuel consumption and other data is already legally required for all new vehicles entering the Australian market under the ADR81/02. The government does not, however, currently collect annual vehicle sales or footprint data, which will be required to determine individual emissions targets for liable entities. As vehicle suppliers already hold this data, requiring this to be reported to government is only likely to be a small additional burden for industry. It will also help policy makers to monitor the target and assess compliance.

In the interest of policy credibility and transparency, the government should consider making non-commercially sensitive data collected to assess compliance with the standards publicly available. This is recommended by the IEA and is currently undertaken by the EU (IEA 2012a, p. 71) and US.

Conclusion

By 2016, liable entities should be required to report annually on sales and vehicle data needed to underpin the standard.

C.3.2.4 Penalties

Penalties are a critical component of any regulatory scheme. The form and level of a penalty for non-compliance must be sufficient to encourage manufacturers to meet the required standard.

Financial penalties are commonly used, including in both the US and EU. In the US, a US$5.50 fine applies for each 10th of a mile per gallon of each new vehicle sold above the target. In the EU, a €95 fine for every gram of emissions of each new vehicle sold above the target is charged. In Japan, a smaller financial penalty applies and firms must make a public announcement of their non-compliance.

Non-compliance over a period can also be accounted for through a make-good provision at the end of a phase. Make-good provisions are easier to administer if trading is allowed.

Financial penalties seem appropriate for Australia. Further analysis by government would be required to determine the appropriate penalty level.

Conclusion

A financial penalty should apply to liable entities who do not comply with the standard.