Green taxation theory in practice

Defining taxes

While the definition of a green tax is contingent on there being a tax in the first place, it is still necessary to look at the definition of an environmental tax. The Organisation for Economic Co-operation and Development (OECD) defines a green tax as follows:

A tax whose tax base is a physical unit (or a proxy of it) that has a proven negative impact on the environment. ²⁴

The CRC has been labelled as a tax by the government, ²⁶ so it is important to consider whether it is actually a tax for the purpose of applying tax theory. A tax is distinct from a charge in the sense that a tax, unlike a charge, ²⁷ provides no direct service to the individual paying the tax. ²⁸ While revenue from the CRC is collected by the administrator through the surrender of allowances, ²⁹ it is not a charge as there is no reciprocal action from the administrator to the participant. I turn now to whether the CRC might be considered to be a tax.

Snape and de Souza review a number of tax definitions from a breadth of authorities. ³⁰ One of the clearest can be found in Re Eurig Estate, a Commonwealth case which highlights three necessary elements of a tax: (1) it is compulsory; (2) it is levied by a public body; and (3) it is intended for a public purpose. ³¹ This definition, also used by Tiley and Loutzenhiser, ³² builds upon the above dictionary definition considerably. Further illumination on the definition of a tax can also be found in UK case law, Gatt 1994/WTO jurisprudence, and European case law. ³³ In terms of English law, ‘such English law authorities as actually exist on the nature of a tax are unsatisfactory’. ³⁴ Both the UK case law and the WTO jurisprudence are considered to be ‘uninformative’. ³⁵ The European case law relates to Article 110 of the Treaty on the Functioning of the European Union (TFEU) (ex Article 90 Treaty Establishing the European Community (TEC)) ³⁶ – however, this is not relevant to the CRC, as it only applies to supply within the UK and it applies equally to all participants.

The Re Eurig Estate definition has received positive attention by the leading academics in the environmental taxation arena, ³⁷ and it is this definition that will underpin the theoretical framework of this article. Applying this to the CRC we can see that:

the CRC is compulsory once the qualification criteria are satisfied, ³⁸ with sanctions invoked for those that do not comply; ³⁹

Snape and de Souza add a fourth element (also found within Re Eurig Estate), ⁴² which is that the tax must be introduced within legislation. ⁴³ Here, the legislative authority of the CRC can be found both in the legislation creating the scheme, and the legislation that reformed the scheme in 2012. ⁴⁴

What is different about the CRC (compared to taxes such as income tax) is the fact that its purpose is to reduce carbon emissions in line with the UK’s international carbon targets, as well as raise revenue. ⁴⁵ More recently, an alternative purpose for taxes has been highlighted – the ability to create social change. ⁴⁶ These taxes, which are usually applied to behaviour that society considers to be unacceptable or damaging, are intended to change the behaviour of the taxpayer. ⁴⁷ Environmental taxes are considered an ‘obvious’ example of a behavioural tax because they attempt to change behaviour for environmental protection. ⁴⁸ Both Tiley and Loutzenhiser ⁴⁹ and Snape and de Souza raise the argument that a ‘levy is not prevented from being a tax by the fact that raising revenue is not the government’s main reason for imposing it’. ⁵⁰ However, emissions taxes should still raise revenue, ⁵¹ and while this function of emissions taxes is in decline, ⁵² I will analyse the CRC in light of both of its functions: reducing emissions and raising revenue.

Another potential difficulty in characterising the CRC as a tax comes from the fact that there is the possibility for participants to benefit under the scheme, due to the presence of two different rates of allowances: ⁵³ forecast allowances currently valued at £16.10; and compliance allowances currently valued at £16.90. ⁵⁴ The lower-priced allowances can be purchased by participants at the start of the compliance year (perhaps by participants who are confident on their future energy consumption), with the higher compliance allowances available to buy at the end of the compliance year – as such, there is the incentive to predict and stick to that energy consumption, in order to pay less. Snape and de Souza discuss how the UK ETS (a predecessor to the CRC) could be likened at times to a fee or charge in the sense that there was an incentive payment ‘purchased at a price determined by the market’. ⁵⁵ I would distinguish the CRC from this, as the two allowances are ‘fixed government sales’ during two specific periods during the compliance year and as such the prices are not determined by the market. ⁵⁶

Having established that the CRC is a tax, the remainder of this article uses literature on effective taxation to ground my analysis. The following section sets out this literature before I turn to an analysis of the 2012 reforms of the CRC.

Theoretical framework: tax theory

This section draws on Smith, Dower and Pigou as part of a critique of effective taxation, and its application to the CRC. The principles of tax were established within the original four eighteenth-century Smithian canons:

tax contributions should be proportionate;

Whilst the four canons are now dated, they are still considered to be relevant, ⁵⁸ and they form the starting point for traditional tax theory.

The nature of emissions taxation means that it aims to change behaviour as well as raise revenue. It is logical that there are further considerations that go beyond the original canons of tax. Behavioural change was not it seems a clear and present consideration in eighteenth century taxation approaches. Instead, we must look elsewhere. Dower outlines four factors of an effective emissions tax in his work for the World Resources Institute, most of which can be linked to the original canons of tax: comprehensiveness; flexibility; administrative costs; and certainty of reductions. ⁵⁹ Whilst the Smith canons of taxation were designed with conventional taxation in mind, the four factors within Dower’s work are tailored to an emissions tax. As there has been little further development in the field of general efficient emissions taxation, this article will take Dower’s four factors and apply them to the CRC reforms.

When looking at the comprehensiveness of an emissions taxation scheme, the coverage of the scheme is important. ⁶⁰ Theoretically, a tax that covers all points and causes of emission, would be completely comprehensive. However, in practical terms, the costs of achieving this would be ‘overwhelming’ and would defeat the purpose of having a tax in the first place (in that the government would raise less than the scheme would cost to implement). ⁶¹ Taxes that are too comprehensively linked to sources are costly, but so are those that do not target the sources that have the biggest carbon impact. It is important to note here that there is often a trade-off between a strong linkage and lower costs. ⁶²

Flexibility is the scope for movement and choice under the CRC scheme, and allows for participants to ‘choose the level and the method of abatement’. ⁶³ Flexibility is specifically introduced by allowing the contributors to pay the tax when it is convenient for them. ⁶⁴ It is also achieved by an emissions tax that has a low requirement for the state to gather information and monitor the tax. ⁶⁵ This requirement relates back to the coverage of the tax, because a lower number of sources lowers the volume of information to be collected by the government; this, in turn, affects the administrative costs of the emissions tax.

The third factor is the certainty of the CRC to reduce emissions. This factor is not certainty in the traditional, legal, sense of the word; rather, it is the certainty that the tax will achieve an emissions reduction, and reflects the fact that the CRC aims to change behaviour as well as raise revenue. As stated above, just because a tax does not have a primary purpose of raising revenue, does not prevent it from being a tax. ⁶⁶ As the primary purpose of the CRC is to change behaviour, this factor examines whether a tax is certain to reduce emissions.

Costs also affect an optimal emissions tax in the three following ways: (1) The fourth Smithian canon requires a tax to be cost effective; (2) if the costs are very high, the tax may not be viable at all; and (3) even where the costs are acceptable, they may still have an impact, as the designers of the tax will have to consider whether to have high cost and a potentially high pollution reduction, or low cost and a potentially lower pollution reduction. ⁶⁷ An initial point on cost that does not form part of the 2012 reforms of the CRC, yet is important to both its ability to reduce emissions and raise revenue, is the price attributed to carbon under the scheme. Dower’s factor does not appear to cover all aspects of cost, as the carbon price under the scheme is not an administrative cost. Further literature on externalities and pricing dates back to the work of Pigou, and Baumol and Oates. ⁶⁸ This older work considers how best to calculate the cost of externalities in environmental taxation which, in the case of the CRC, would be the price attributed to carbon emissions. However, it is sufficient to say that, while initially the carbon price changed fairly substantially over the first four years of the scheme, the price has now plateaued at £16/tonne and increases in line with the Retail Price Index. ⁶⁹ This automatic increase suggests that the government feels confident with the current value (that is, it is set at the right level to reduce emissions).

Weighing up the factors

Cost–benefit analysis, which focuses on the externalities of environmental regulation, ⁷⁰ is ‘an economic tool for comparing the desirable and undesirable impacts of proposed policies’. ⁷¹ It ‘defines the costs and benefits of a project in much wider terms than those included in investment appraisals’. ⁷² Due to the recent nature of the reforms, many of the economic impacts of the CRC are unknown; as such it is not possible to use this method for the purposes of the current analysis. This article will instead contribute a preliminary overview of the likely impacts of the 2012 reforms by weighing the potential advantages and disadvantages of the reforms.

Flexibility

Increased flexibility cannot be seen within all six of the proposals or within the eventual reforms of the CRC. However, a number of the proposals did have an impact on the flexibility of the scheme as will be highlighted below.

Proposal 12, sought to reduce the number of fuels caught under the scheme from 27 to four. ⁷⁴ The eventual reform went even further and reduced the number of fuels to two: electricity and gas (for heating). ⁷⁵ In terms of the number of fuels, the added flexibility brought by proposal 12 could be seen as a double-edged sword. At both stakeholder events during the 2012 CRC consultation, participants voiced their concerns that a reduction in the number of fuels would only encourage some participants to use the ‘dirtier’ fuels, for example coal. ⁷⁶ However, a degree of flexibility was included in the reforms and participants are now able to choose to use fuels that would avoid liability under the CRC scheme. This could, however, have a devastating impact on the environment:

Compared to the average air emissions from coal-fired generation, natural gas produces half as much carbon dioxide, less than a third as much nitrogen oxides, and one percent as much [sulphur] oxides at the power plant. ⁷⁷

There was little discussion in the government documents analysing the impact of proposal 18 which removes Electricity Generating Credits (EGCs) from fuels that no longer fall under the CRC. ⁸⁰ EGCs had previously allowed participants to offset some of their liability under the scheme. ⁸¹ However, when considering the potential impacts of the reform on the flexibility of the scheme, the government stated that this could potentially discourage renewable onsite generation and increase reliance on grid electricity. ⁸² If participants will no longer be rewarded for generating their own electricity (with the removal of the EGC scheme) then the choice to generate on-site electricity has become less attractive. This in turn, would remove this choice for some participants, reducing the flexibility of the CRC. However, as grid electricity is comparably very expensive: 14.2p per kWh, ⁸³ those with existing electricity-generating infrastructure would be likely to continue generating electricity (for example, coal generates a kWh of electricity for 4.67p). Additionally, electricity generation provides an opportunity for participants to earn money via Feed-in-Tariff (FIT) payments if they generate surplus electricity exported to the grid. ⁸⁴ These points, especially the cost of grid electricity, means that the reform is unlikely to affect (a) those already generating electricity; and (b) the flexibility of the scheme.

Increased flexibility can be seen in the changes in proposal 39, which moved the CRC payment date from July to September. ⁸⁵ This was to give participants more time between ordering and paying for allowances. ⁸⁶ Following the consultation, ⁸⁷ the deadline was extended further to the end of October. ⁸⁸ Participants will now have three calendar months between the payment date (when the allowances are paid for) and the surrender date (when the number of allowances equal to emissions is surrendered). It gives participants time to accurately calculate and order their allowances, and allows them to choose, during a longer period of time, when to do so. It generally permits ‘the deadlines relating to the request, payment and allocation of allowances to be staggered’, which could also have an impact on the costs of the scheme. ⁸⁹ A minor and final point is that proposal 14 gives participants the flexibility to choose to reduce gas consumption below the 2 per cent de minimis limit. This is, perhaps, a tenuous point, but the potential for excluding liability for gas under the scheme now exists.

Comprehensiveness

The comprehensiveness of a green tax is directly linked to the scheme’s ability to reduce emissions. That is to say that if a source of carbon is not taxed, the scheme cannot have an impact on reducing that source of carbon. This section will look at the proposals that have had the biggest impact on the comprehensiveness of the CRC.

Some concern had been expressed about the reduction in emissions coverage, as proposal 7 excluded all domestic electricity, and domestic gas supplies that totalled 73,200kWh or less. ⁹⁰ It was argued that the scheme would be in a weaker position to reduce emissions. ⁹¹ However, the government and the majority of respondents to the 2012 CRC consultation believed that this effect would be minimal as most of the domestic supply would have been discounted in any event at a later stage under the domestic accommodation exclusion (DAE). ⁹² This exclusion applied to ‘premises intended to be used as a permanent home’. ⁹³ However, the exclusion not only applies to private owner occupier housing, but also emergency temporary accommodation provided by local authorities (a mandatory participant in the CRC). ⁹⁴ Therefore, although the scheme may appear to be less comprehensive, the existence of DAE prior to the reform means that domestic electricity would have been excluded anyway. There is, therefore, no net change in the comprehensiveness or emissions reduction of the CRC.

The reduction in the scope of the CRC, in proposal 12, calls into question the comprehensiveness of the scheme. With 27 fuels now falling outside the scope of the scheme, there is an argument that the CRC is less comprehensive and excludes fuels that emit a high amount of carbon, such as coal. This argument, however, is less persuasive when considering the Final Impact Assessment (FIA) of the 2013 Order, which stated that the two remaining fuels (electricity and gas) account for 97 per cent of emissions under the scheme. ⁹⁵ Gas falls under the revised CRC scheme if it used for heating purposes. ⁹⁶ Consequently, gas that is used for other purposes represents a reduction in the scope of the scheme. The 2013 FIA indicated that the gas used for heating purposes, represented 90 per cent of the total gas used by participants. ⁹⁷ The current reform represents a change in the comprehensiveness of the CRC; with a 3 per cent drop in the coverage of fuels, and 10 per cent of gas now falls outside the scheme.

Proposal 14 sought to increase reporting obligations on emissions from 90 to 100 per cent. ⁹⁸ This was implemented with a 2 per cent de minimis threshold for gas. ⁹⁹ At face value, an increase to 100 per cent reporting may look like an increase in the scope of the scheme. However, the increase to 100 per cent was primarily designed to offset the loss of coverage resulting from the reduction in fuels. ¹⁰⁰ The arguments about the loss of benefits associated with proposal 12 are thus equally applicable for proposal 14. The scheme will be in a weaker position to reduce emissions at the rate it was achieving before.

Costs

One of the primary drivers behind the 2012 consultation and review of the CRC was the initial cost of the scheme. According to the Adam Smith’s fourth canon of taxation, costs should not be too high. I will now turn to some of the reforms that will have an impact on the cost of the CRC.

Excluding certain meters from the scope of the CRC (proposal 7) means that there are fewer from which to collect data. ¹⁰¹ This has a potential beneficial impact on the administrative costs of the scheme:

Government proposes that this exclusion will significantly simplify how organisations identify and exclude supplies used for domestic accommodation, without a significant emissions coverage implication. ¹⁰²

Respondents suggested that a large number of small supply points would be excluded and would result in a reduction of the administration burden. ¹⁰³

One clear advantage of proposal 12 and the smaller number of fuels is the reduction of administrative costs. ¹⁰⁵ In its report for the government, KPMG estimated that participants spent around half their time gathering information on non-core sources, which was:

… surprising as most participants should have ruled out most non-core sources that are administratively complex to gather and which, by definition, normally only represent a small proportion of their footprint. ¹⁰⁶

The main drive behind proposals 12 and 14 was the need to reduce the reporting burden of the CRC. The argument for reform of the 90 per cent rule was therefore that it removed the need for participants to provide footprint reports. ¹¹¹ Furthermore, stakeholders made the point that participants were required to gather data on 100 per cent of their emissions just so that they could work out 90 per cent of their emissions. ¹¹² Therefore, the reform will likely result in a reduction of participant staff hours put into the CRC scheme. This is counterbalanced by other stakeholders who claimed that the reform will now include sites that had previously been excluded under the 90 per cent rule. ¹¹³ On the other hand, participants will no longer have to go through the process of designating these sites when deciding which 10 per cent they will exclude from the scheme. Finally, an argument was put forward that 100 per cent reporting will place an unsustainable burden on small companies. ¹¹⁴ However, I would argue that 100 per cent of the data will be required to report upon 90 per cent of emissions, and therefore, the additional 10 per cent of reporting will be worth the ‘significant simplification’ of the scheme. ¹¹⁵

Regarding the introduction of the de minimis threshold for gas, the reduction in administrative costs is possibly not as significant as might be first thought. It is still necessary for participants to work out their gas consumption to work out whether they comply with the de minimis. ¹¹⁶ This means that all the costs that are associated with data collection will still apply. The difference, however, is that if the de minimis is achieved, the participant does not need to do anything with the raw data - there will, therefore, be no need to translate this into a report, thus there will be some reduction in costs. This is boosted by the fact that the de minimis threshold only needs to be calculated once at the start of phase two (which commenced in 2014) for a gas exemption lasting the whole of the second phase (until 2018). ¹¹⁷

The main rationale behind the reform of the EGCs in proposal 18 was the reduction in costs that this will entail. The reduction is twofold: in administrative costs and the costs associated with ‘double claims’. While this is not mentioned in the consultation process, the removal of a credits scheme also removes all the costs associated with the claims process. This has the potential to reduce staff hours for both the participants and the regulator, as well as general costs associated with data collection and the monitoring of claims. The potential for ‘double claims’ has arisen following the reduction in the number of fuels seen within proposal 12:

The proposed reduction in fuels covered by the scheme could lead to unintended consequences if participants could claim EGCs for generation where the input fuel was not within scope of the scheme. ¹¹⁸

Whilst proposal 39, and the change in payment date, was not explicitly discussed during the consultation process, it would appear logical that a longer period of time between payment and surrender would allow participants to calculate more accurately the number of allowances that they need before surrender. It would perhaps lead to a lower number of order errors such as where participants order too many or too few allowances, as they have more time to check their data and calculations. This is, however, a minor point, which is probably the reason it did not feature in the consultation process.

Finally, it is possible that the abolition of the Performance League Table (which ranked participants in order of performance under the scheme) will result in a reduction in the costs of preparing the League Tables. However, the government believed these savings to be negligible. ¹¹⁹ In effect, the reform has no real impact on the costs of the CRC because it effectively cancels out the costs of the PLT by introducing the ARPs. The cost element of this reform will be regarded by this article as unchanged following the introduction of the ARP.

Certainty to reduce emissions

The final factor, certainty, is closely linked to all the above factors. A comprehensive, flexible and affordable scheme will affect its ability to run efficiently and effectively. This section will start to discuss how the reforms could alter the CRC’s ability to change behaviour and so reduce emissions.

Before its reform, the CRC covered around 57.7 million tonnes of carbon dioxide. ¹²⁰ The reform resulted in 1 million tonnes of carbon dioxide falling outside the scope of the scheme. ¹²¹ Due to this lower coverage, it has been estimated by the government that air quality will drop by £3 million (although the qualification of a monetary figure is absent in the government documents). ¹²² The government considers any air quality impact that is less than £50 million to be small; ¹²³ by using other government figures for air quality damage costs (i.e. £0.15 per kWh/electricity as a base figure), I roughly quantify this figure as the equivalent of 9,243.8 tCO₂. ¹²⁴ Considering that the average annual emissions of participants in 2014 was 26,129 tCO₂, ¹²⁵ the drop in air quality is relatively small. However, a reduction in emissions coverage is still a reduction, and the scheme will have less impact on emissions reduction. An interesting point that was not mentioned during the progress of the 2013 Order is whether the drop to two fuels would permit participants to focus more clearly on reducing their emissions. It is arguable that a participant faced with having to potentially reduce their use of 29 fuels could be daunted - it would also be spreading their resources considerably more thinly than if they just focused on reducing their electricity and gas usage. This hypothetical would, however, need empirical evidence to support its premise.

In addition, any measure which has the ability to inhibit renewable energy generation will indirectly increase carbon emissions. However, as stated, it is unlikely that participants will stop generating electricity if they already possess the infrastructure just because of the removal of EGCs. The problem lies with future generators, as those that do not possess this infrastructure will be less inclined now to invest in renewable energy. Although the EGC scheme has now ended, the Renewable Obligation Certificates (ROC) still exist and the government predicts that as a result ‘the net impact on the scheme’s emissions coverage will be minimal’. ¹²⁶ It is difficult to predict whether renewable energy, and so the potential reduction in emissions, will be affected because of this reform. Although there is likely to be an impact on future renewable generation, this should be small considering that organisations still benefit from FIT payments and ROC subsidies.

The Performance League Table (PLT) ranked participants by performance, ¹²⁷ and provided the reputational element of the scheme. ¹²⁸ Proposal 43 proposed the abolition of the PLT. ¹²⁹ It also suggested that the reputational element be held within guidance documents (as opposed to legislation), so that concrete decisions about the ‘nature of the reputational element’ did not have to be made. ¹³⁰ The PLT was abolished, and the Environment Agency (EA) now annually publishes the raw emissions data for each participant (without any form of ranking or commentary). ¹³¹

A brief comment needs to be made on the substantial criticisms raised against the PLT. Two main criticisms were evident from the consultation process: that the PLT was not presented in a suitable format that allowed for comparison and that it over relied upon earlier data from the first year, which skewed the results. ¹³² The main stakeholders demonstrated a ‘strong support for presenting data in a sortable format so that users can filter it and make valid comparisons’. ¹³³ This was coupled with the desire for there to be an accompanying narrative which explained in qualitative terms what measures each participant was taking to reduce emissions. ¹³⁴ The reform, which states that the EA will instead simply publish the participant’s aggregated use in an annual report, provides a spreadsheet of data alongside a report to provide context. ¹³⁵

The second criticism is that the PLT did not accurately portray information. ¹³⁶ This criticism relates to complex matrixes and weightings that were used by the EA to calculate the PLT, which consequently skewed the table. ¹³⁷ By simply publishing the aggregate energy use, the EA will no longer have use for such matrixes and so this criticism has been addressed. Another comment made in relation to the data was that the PLT does not distinguish between public and private data. ¹³⁸ As such, the majority of stakeholders even felt that the PLT compared ‘apples to oranges’ in relation to this mixed public/private sector table. ¹³⁹ When considering the 2012/2013 Annual Report Publication (ARP) published by the EA, there is little to distinguish the table from the PLT, other than the fact that participants are in alphabetical order rather than ranked. ¹⁴⁰ The ARP which accompanies this table does not distinguish between public and private organisations. ¹⁴¹ However, it is clear from the report that the table of information is simply raw data – it is even explained that any further analysis of the data would now be left down to third parties. ¹⁴² The abolition of the PLT left a potential gap in the scheme’s ability to reduce emissions. However, the EA now publishes an ARP; consequently, participants are still potentially exposed to public criticism which could create a reputational driver for participants to lower their energy use. There is also the potential for third-party analysis of the raw data to pose a threat to those participants who are very high energy users. It is difficult to say at this stage whether the PLT or the new ARP has more of a motivational effect; however, the factual information now published by the EA is easily accessible and would be easy to analyse. High energy users will no longer be able to rely on any early data matrixes that previously skewed the table.

A final and more subtle point, is that proposal 39 will ‘assist participants in complying with their obligations’ under the 2013 order. ¹⁴³ This general extension in the payment date does not reduce emissions per se, but it facilitates the general running of the scheme. ¹⁴⁴ Therefore, any reform that facilitates the operation of the CRC is arguably facilitating the CRC’s end goal: an 80 per cent reduction in GHG by 2050. ¹⁴⁵