[ECONOMY] a page

15 Jul [ECONOMY] a page

Posted at 11:53h in 1Academic Level by Homework Writer

AN ECONOMIC EVALUATION | 1

Australia’s Carbon Tax: An Economic Evaluation

by Dr. Alex Robson, PhD

Department of Accounting, Finance and Economics

Griffith University, Brisbane, Australia

September 2013

AUSTRALIA’S CARBON TAX

AN ECONOMIC EVALUATION | 1

Table of Contents

Executive Summary ……………………………………………………………………………………………………….5

1. Introduction …………………………………………………………………………………………………………..11

2. History of the Carbon Tax ……………………………………………………………………………………….12

2.1. The Shergold Report (2007) ………………………………………………………………………………..12

2.2. The Garnaut Report (2008) and the Carbon Pollution Reduction Scheme (CPRS) (2009) ……………………………………………………………………….14

2.3. Carbon Taxes versus Cap and Trade Schemes: The Standard Treatment in the Literature ………………………………………………………………15

3. Policy Framework and Key Parameters……………………………………………………………………18

3.1. Development of the Tax ……………………………………………………………………………………18

3.2. Coverage: Who Pays?……………………………………………………………………………………….18

3.3. Abatement Target and Sources of Abatement ……………………………………………………..19

3.4. The Price Floor and Price Ceiling ……………………………………………………………………….19

3.5. Other Policies ………………………………………………………………………………………………….21

3.5.1. Complementary Emissions Reduction Policies…………………………………………..21

3.5.2. Household Compensation……………………………………………………………………….22

3.5.3. Free Permits …………………………………………………………………………………………22

4. The Economic Costs of Australia’s Carbon Tax ……………………………………………………….23

4.1. The Carbon Tax and Australia’s Exports ………………………………………………………………23

4.2. The Interaction Between the Carbon Tax and Other Policies ………………………………….25

4.2.1. Complementary Emissions Reduction Policies…………………………………………..25

4.2.2. The Effects of Household Assistance and Income Tax Changes ………………….25

4.3. International Linking and Restrictions on International Trade ………………………………….28

4.3.1. The Gains from International Trade in Permits …………………………………………..28

4.3.2. Constraints on Overseas Permit Purchases ………………………………………………29

AN ECONOMIC EVALUATION | 1

2 | AUSTRALIA’S CARBON TAX

4.4. Dynamic (In)efficiency ……………………………………………………………………………………….31

5. Economic and Fiscal Effects……………………………………………………………………………………32

5.1. Gross Domestic Product (GDP) Losses ………………………………………………………………32

5.2. GDP Costs per Tonne of Abatement……………………………………………………………………34

5.3. Business Costs, Profitability and Carbon Leakage………………………………………………..35

5.4. Real Wages and Unemployment…………………………………………………………………………37

5.5. Consumer Prices………………………………………………………………………………………………39

5.6. Fiscal Effects ……………………………………………………………………………………………………41

5.7. Effect on CO2-e Emissions ………………………………………………………………………………..43

5.7.1. Overall Emissions ………………………………………………………………………………….43

5.7.2. Electricity Sector Emissions ……………………………………………………………………43

6. Conclusions: Policy Lessons from the Australian Experience ………………………………….47

References……………………………………………………………………………………………………………………50

Appendices ………………………………………………………………………………………………………………….51

Endnotes ……………………………………………………………………………………………………………………..56

AN ECONOMIC EVALUATION | 3

List of Figures

Figure 2.1: Marginal Costs and Benefits of CO2-e Abatement …………………………………………….12

Figure 2.2: A Cap and Trade Scheme ………………………………………………………………………………13

Figure 2.3: Transaction Costs in a Cap and Trade Scheme ………………………………………………….13

Figure 2.4: Summary of the Legislative History of the Carbon Tax ……………………………………….15

Figure 2.5: Efficiency Losses from a Carbon Tax and a Cap and Trade Scheme When Marginal Costs are Unexpectedly High ……………………………………….16

Figure 3.1: Sources of Cumulative Abatement Relative to Business as Usual Projections, 2013-2050 ……………………………………………………………………..19

Figure 3.2: Baseline Permit Prices, Price Ceiling Path and Price Floor Path Under the Original CEF Policy ………………………………………………………………….20

Figure 3.3: Welfare Effects of a Price Floor and Price Ceiling ………………………………………………21

Figure 4.1: Peters and Hertwich (2008) Estimates of the Balance of Emissions in Trade …………24

Figure 4.2: The Costs of a Mandatory Renewable Energy Target …………………………………………25

Figure 4.3: The “Tax Interaction Effect”: The Welfare Effects of a Pigouvian Tax when there is a Pre-Existing Distortion in a Related Market ………………………………..26

Figure 4.4: International Trade in Emissions Permits …………………………………………………………..29

Figure 4.5: Domestic Emissions and Abatement with Free Trade in Permits…………………………..29

Figure 4.6: Domestic Emissions and Abatement Under Australia’s CEF Policy to 2020…………..30

Figure 4.7: Carbon Price Projections, 2012-13 to 2019-20 ………………………………………………….31

Figure 5.1: Present Value of Projected Economic Costs of Australia’s Carbon Tax to 2050 ……..33

Figure 5.2: Average and Incremental Costs of Abatement, Relative to Business as Usual Emissions…………………………………………………………..34

Figure 5.3: Sectoral Shares of Total Business Electricity Use, 2011-12 ………………………………..35

Figure 5.4: Carbon Leakage in an Import-Competing Industry …………………………………………….36

Figure 5.5: Annual Reduction in Real Wages Versus Annual Reduction in GDP Under the Carbon Tax, Relative to Baseline, 2013-2020…………………………..37

4 | AUSTRALIA’S CARBON TAX

Figure 5.6: Unemployment in Australia, July 2012 to July 2013 …………………………………………..38

Figure 5.7: Inflation-Adjusted Household Electricity Prices, 1980 to 2013 …………………………….40

Figure 5.8: Expected Cumulative Fiscal Impact of the Carbon Tax and Associated Policies, 2011-12 to 2014-15 ………………………………….42

Figure 5.9: Initial Estimates and Revisions of Carbon Tax Revenue, 2012-13 to 2016-17 ……….43

Figure 5.10: Government Projections of Australia’s Domestic CO2-e Emissions Under the Carbon Tax, 2013-2050……………………………………….44

Figure 5.11: Australia’s Electricity CO2-e Emissions, 2002-2012 …………………………………………45

Figure 5.12: Australia’s total CO2-e emissions, Seasonally Adjusted Weather Normalised, 2002-2013……………………………………………………………………46

List of Tables

Table 3.1: Allocation of Free Permits to Emissions Intensive Trade Exposed Industries …………22

Table 4.1: New Statutory Income Tax Rates, Old EMTRS and New EMTRs …………………………28

Table 5.1: Estimated GDP Costs of Policy Commitments Under the Copenhagen Accord……..32

Table 5.2: Estimated Effect of the Carbon Tax on Wholesale Electricity Prices to 2050 …………39

Table 5.3: Estimated Contribution of the Introduction of the Carbon Tax and Other Green Schemes to a Typical Annual Household Electricity Bill, Qld and NSW ………………………………………………..41

July 2012 Carbon tax beginsFebruary 2011

Government announces proposed architecture of carbon tax

August 2010 Government promises not to introduce carbon tax

April 2010 Government withdraws cap and trade legislation and commits not to introduce scheme before the end of 2012

May 2009 Cap and trade legislation introduced. Scheme to begin on July 1 2010

AN ECONOMIC EVALUATION | 5

Executive Summary

Australia has implemented a carbon tax, and it is failing to deliver any of its promised benefits. Its failures

have made the tax a highly politicized issue, and may provide lessons for other nations. The tax, which is

currently set at $24.151 is the central component of the Australian Government’s climate change policy.

The tax applies directly to around 370 Australian businesses2 and was originally designed as a precursor

to a “cap and trade” scheme, with the transition to a flexible price originally (and currently) scheduled to

take place on July 1, 2015.

This report, commissioned by the Institute for Energy Research, evaluates Australia’s carbon tax experience and draws lessons for policymakers in the United States and other jurisdictions, who may be considering following the Australian example and implementing their own carbon taxes or cap and trade schemes. The analysis establishes a number of key points, which are summarised below.

Establishing a Robust, Sustainable and Credible Carbon Tax is Politically Difficult. Policy Uncertainty and Time Inconsistency are Likely to be the Rule Rather than the Exception Figure E1 below summarises the legislative evolution of Australia’s carbon tax and shows that the policy was plagued by uncertainty well before it was formally introduced. Prior the 2010 election, neither major

political party in Australia supported a carbon tax – yet less than a year later, legislation to give effect to the tax was introduced into Parliament. In addition, the tax was subjected to a number of significant changes almost immediately after it came into effect, reducing certainty for businesses and directly negating one of the original justifications for the tax. For example, originally the proposed scheme was to have a fixed price for the first three years, followed by a floating price which would be subject to floor and ceiling prices. However, on August 28, 2012, less than two months after the scheme began, the Government announced that there would no longer be a floor price. That such significant changes were made to the scheme so soon after it began suggests that the original design contained significant flaws.

FIGURE E1: SUMMARY OF THE LEGISLATIVE HISTORY OF THE CARBON TAX

Sep 1980

Sep 1982

Sep 1984

Sep 1986

Sep 1988

Sep 1990

Sep 1992

Sep 1994

Sep 1996

Quarter

Sep 1998

Sep 2000

Sep 2002

Sep 2004

Sep 2006

Sep 2008

Sep 2010

Sep 2012

Carbon Tax Introduced

0.5

1.5

2.5

In de

x of

H ou

se ho

le E

le ct

ric ity

P ric

es (S

ep 1

98 0=

6 | AUSTRALIA’S CARBON TAX

Despite the carbon tax passing both the House of Representatives and the Senate and becoming law, political and popular support for the policy has been weak. Recently the Australian Government has proposed further major changes to the tax, announcing its desire to move earlier towards a cap and trade scheme, with the new transition taking place on July 1, 2014. However, legislation to give effect to this proposed change has not yet been introduced into Parliament; and in any case, it is unclear whether such legislation would actually be passed.3

As a result, there is still a great deal of uncertainty surrounding the future status of the carbon tax. Depending on the result of the forthcoming election, the tax may either remain in place and transition to cap and trade in 2015, or it may move to a cap and trade scheme in 2014, or it may be abolished completely.

In Assessing the Case for a Carbon Tax or Cap and Trade Scheme, the Incremental Net Benefits of All Feasible Policy Options Were Not Estimated. One reason for the lack of robustness of the carbon tax policy is that its development followed a flawed policy process. The role of climate change policy is not to assess the possible damage of climate change, but rather to focus on the incremental net benefits of possible policy options. A central tenet of good economic policymaking is that a full cost benefit analysis (CBA) should be undertaken, weighing up the gains and losses across a wide range of policy alternatives so that political decision-makers can be better informed of the economic effects of various options. Sensitivity analysis should be undertaken in order to determine the extent to which the results of such analysis depend on modelling assumptions and other inputs. If sensitivity analysis shows that a proposed policy’s estimated net benefits vary wildly

FIGURE E2: INFLATION-ADJUSTED HOUSEHOLD ELECTRICITY PRICES, 1980 TO 2013

720

700

680

660

640

620

600

580

560

540

Carbon Tax Introduced

Pe rs

on s

U ne

m pl

oy ed

(T ho

us an

ds )

July 2011

Sep 2011

Nov 2011

Jan 2012

Mar 2012

May 2012

July 2012

Sep 2012

Nov 2012

Jan 2013

Mar 2013

May 2013

July 2013

Month

AN ECONOMIC EVALUATION | 7

with assumptions, the policy should be treated with a great deal of care and probably rejected on the grounds that it is unlikely to result in net benefits.

Whilst a number of Government-commissioned reports attempted to examine the economic costs of carbon taxes and emissions trading schemes, the incremental net benefits of the policy were never assessed. In other words, costs and benefits were never compared. Instead, Government-sponsored reports purported to measure benefits by examining the possible future damage that may be caused by climate change in Australia. But estimating these costs is not the same as estimating the benefits of various policies. In particular, there was never an assessment of the incremental net benefits to Australia of limiting emissions, versus other measures such as adaptation. The Australian debate has always been framed as limiting emissions on the one hand, versus doing nothing on the other.

In addition, the Government’s quantitative modelling of the costs made a number of highly unrealistic assumptions and lacked transparency (Ergas and Robson, 2012). This made it impossible for neutral third parties to replicate and evaluate the results, or modify the assumptions to test the robustness of the results.

The Cumulative Economic Costs of Carbon Taxes or Cap and Trade Schemes are Likely to be Substantial Over the Long Term, with Lower Discount Rates Resulting in Higher Cumulative Costs in Present Value Terms Under the carbon tax, most of the abatement that Australia will take credit for over the period to 2050 will be undertaken overseas, with Australian businesses paying their foreign counterparts to reduce emissions. Nevertheless, the tax will have significant economic costs. So far the main economic effect of the tax has been to increase energy prices (particularly

FIGURE E3: UNEMPLOYMENT BEFORE AND AFTER THE INTRODUCTION OF THE CARBON TAX

Dec 2002

Jun 2003

Dec 2003

Jun 2004

Dec 2004

Jun 2005

Dec 2005

Jun 2006

Dec 2006

Jun 2007

Dec 2007

Quarter

Jun 2008

Dec 2008

Jun 2009

Dec 2009

Jun 2010

Dec 2010

Jun 2011

Dec 2011

Jun 2012

Dec 2012

125

130

135

140

145

Emissions since the introduction of the carbon tax

Q ua

rt er

ly C

O 2-

e E m

is si

on s

(M t)

8 | AUSTRALIA’S CARBON TAX

electricity costs) for households and businesses (see Figure E2). According to the Australian Industry Group (AIG), energy cost increases have averaged 14.5 per cent for businesses as a result of the carbon tax, whilst TD Securities and the Melbourne Institute found that due to the introduction of the carbon tax, the price of electricity for households rose by 14.9 per cent. The increase in household electricity prices after the carbon tax was introduced was the highest quarterly increase on record.

The Government’s own modelling (which, as the report discusses, are likely to have underestimated the costs of the tax) indicates that Australia’s Gross Domestic Product (GDP) will be lower than it otherwise would be for every year that the tax is in place. Depending on the discount rate used, the present value of these costs could be as high as 83 per cent of current Australian GDP, or $1.25 trillion. The carbon tax has been

FIGURE E4: AUSTRALIA’S TOTAL CO2-E EMISSIONS, SEASONALLY ADJUSTED WEATHER NORMALISED, 2002-2013

directly linked to a number of business closures and job losses, with overall unemployment rising significantly since the tax was introduced (see Figure E3).

Furthermore, government data shows that the tax has not reduced the level of Australia’s domestically produced CO2-e emissions (Figure E4). This is not surprising, since under the carbon tax Australia’s domestic emissions are not expected to fall below current levels until 2045.

Carbon Leakage is Likely and will Create Economic Costs with no Offsetting Environmental Benefit Overall, Australia’s exports are relatively emissions intensive. Hence a carbon tax is likely to increase the cost of exports, whose prices are largely determined on world markets. There is little opportunity for Australian export industries to pass on the increases in costs that are due to the carbon tax. In other

35,000

30,000

25,000

20,000

15,000

10,000

5,000

-5,000

-10,000

$24,470 million

$3,300 million -$15,356 million

-$9,222 million

$-3,192 million

Revenue from permit sales

Revenue from other measures

and fuel tax credit

Household Assistance

Free Permits

Overall Fiscal Impact -4,380 million

Other Programs

$ m

ill io

Measure

AN ECONOMIC EVALUATION | 9

FIGURE E5: EXPECTED CUMULATIVE FISCAL IMPACT OF THE CARBON TAX AND ASSOCIATED POLICIES, 2011-12 TO 2014-15

words, the effect of the carbon tax on Australia’s emissions-intensive, trade-exposed industries is similar to a tax on exports or a tax on import- competing industries. Providing free permits to these industries does not alter marginal incentives. Domestic emissions in these industries may fall after a carbon tax is imposed, but that cannot be counted as an environmental gain if the ultimate effect is that the businesses shut down and emissions simply rise overseas. The net effect will be a pure deadweight cost to the Australian economy.

Fiscal Impacts are Likely to be Uncertain, with both Carbon Taxes and Cap and Trade Schemes Adding to Existing Revenue Volatility Due to the structure of the carbon tax and accompanying policies, a sizeable fiscal gap has opened up between the revenues generated by the tax on the one hand, and the increases in government

spending and tax cuts that accompanied the scheme on the other. A significant proportion of compensation payments were “locked in”, whilst revenue from the tax is likely to be lower than originally anticipated. Hence the introduction of the tax, together with other policies, is likely to worsen Australia’s budget bottom line going forward, leading to higher deficits and higher public debt than would otherwise have been the case.

Attention Needs to be Paid to the Effects and Costs of “Complementary” Policies, Which Are Likely to Result in Efficiency Losses Rather than Efficiency Gains, Compounding any Negative Effects of a Carbon Tax or Cap and Trade Scheme Table E1 below shows that the carbon tax, together with other green schemes, now account for a significant portion of a typical Australian household’s electricity bills. Proponents of carbon taxes have

TABLE 2: INCREASED OUTPUT FROM OPENING FEDERAL LANDS ($ MILLIONS ANNUALLY)

QLD (2012-13) NSW (2013-14)

RENEWABLE ENERGY TARGET $102 $107

SOLAR BONUS SCHEME/OTHER SCHEMES $67 $53

CARBON TAX $190 $172

TYPICAL HOUSEHOLD BILL $1900 $2073

GREEN SCHEMES/TOTAL 19 PER CENT 16 PER CENT

10 | AUSTRALIA’S CARBON TAX

pointed to several kinds of efficiency gains that may accompany such taxes. It is often claimed, for example, that imposing a carbon tax allows policy makers to eliminate other, more costly “complementary” measures that are designed to reduce emissions, such as green subsidies (eg for solar and wind power), renewable energy targets, and so on.

However, these efficiency gains are unlikely to materialise in Australia’s case: the complementary measures have remained in Australia after the carbon tax was put in place. To make matters worse, new complementary measures have been introduced which will likely increase economic costs. Hence any hypothetical efficiency gains that may have occurred as a result of eliminating other programs remain just that: hypothetical.

TABLE E1: ESTIMATED CONTRIBUTION OF THE INTRODUCTION OF THE CARBON TAX AND OTHER GREEN SCHEMES TO A TYPICAL

ANNUAL HOUSEHOLD ELECTRICITY BILL, QLD AND NSW

The “Double Dividend” is Elusive in Theory and Difficult to Achieve in Practice Carbon tax proponents also argue that carbon tax revenue can be “recycled” and used to reduce marginal income tax rates, thus providing a “double dividend.” The report also shows how the double dividend hypothesis is a dubious proposition in theory, due to the interaction between the carbon tax and the existing tax system (particularly personal income taxes and corporate taxes). In addition, as part of the household compensation package for the carbon tax, the Australian Government lowered some average income tax rates but actually increased marginal tax rates for around 2 million taxpayers. This increase in marginal tax rates is exactly the opposite policy of what a Government would do if it were trying to capture a “double dividend” from environmental taxation. In practice, therefore, there has been no double dividend from Australia’s carbon tax.

Conclusion Poor policy processes tend to lead to poor policy outcomes. Australia’s carbon tax experience provides a number of important lessons in how not to go about implementing sensible climate change policy. Although a number of Government reports examined the possible costs of the carbon tax, none of them assessed the incremental net benefits of the policy. For a variety of reasons, it is unlikely that Australia’s carbon tax will achieve “abatement at least cost.” The most significant complementary climate change policies have remained in place after the introduction of the tax, and a range of new, costly measures were introduced to accompany the policy. These factors have weakened—perhaps fatally—the economic case for Australia’s carbon tax.

Overall, Australia’s exports are

relatively emissions intensive. Hence a

carbon tax is likely to increase the cost

of exports, whose prices are largely

determined on world markets.

AN ECONOMIC EVALUATION | 11

I. Introduction

Australia’s carbon tax, which came into effect on July 1, 2012 and is currently set at $24.15, covers a

broad range of industry sectors and categories of carbon dioxide equivalent (CO2-e) emissions. The tax is

a fixed price emissions permit system, and is legislated to move to a full “cap and trade” or flexible

emissions price scheme in 2015, with Australian firms permitted to buy permits from the European Union.

The stated purpose of the tax is to reduce Australia’s CO2-e emissions below projected “business as

usual” levels. Slightly less than half of the expected CO2-e abatement in the period to 2050 is expected to

occur as a result of domestic reductions in emissions, with the most abatement being sourced from

purchases of overseas permits.

Despite the carbon tax passing the House of Representatives and the Senate and becoming law, political and popular support for the policy has been weak. Recently the Australian Government has proposed to move earlier towards an internationally linked cap and trade scheme, in July 2014. However, legislation to give effect to this proposed change has not yet been introduced into Parliament. In any case, it is unclear whether such legislation—should it be introduced—would even pass.4

As a result, there is a great deal of uncertainty surrounding the future status of the tax. Depending on the outcome of the forthcoming Australian election (which will be held on September 7), the tax may either (i) remain in place and transition to cap and trade in 2015 (as originally planned); or (ii) it may move earlier to a cap and trade scheme in 2014; or (iii) it may be abolished altogether.

This report evaluates the carbon tax in terms of policy process, policy design, and economic outcomes. The report is structured as follows. Section 2 outlines the policy history behind the carbon tax, and some of the legislative background to the current scheme, as well as reviewing some of the basic economic arguments that have been made to justify the introduction of the tax. Section 3 outlines the key economic features of the scheme. Section 4 examines the economic costs of the tax, and explains why it is unlikely that the Australian Government’s policy will “achieve abatement at least cost.” Section 5 examines the economic and fiscal effects of the carbon tax. Section 6 concludes by outlining the main policy lessons from the Australian experience.

Despite the carbon tax passing the House of Representatives

and the Senate and becoming law, political and popular support for

the policy has been weak.

12 | AUSTRALIA’S CARBON TAX

2. History of the Carbon Tax

The political history of Australia’s carbon tax began with the Prime Ministerial Task Group on Emissions

Trading (more popularly known as the Shergold Report after its main author), which was released in mid

2007.5 This Task Group advised “on the nature and design of a workable global emissions trading system

in which Australia would be able to participate.” In response to the Shergold report, then Prime Minister

John Howard announced on July 17 that a “cap and trade” system would be introduced in Australia. The

scheme was planned to be operational possibly in 2011, and no later than 2012.6

The Shergold Report appealed to some standard analysis and results from basic welfare economics to justify its recommendation to introduce a cap and trade scheme, arguing that emissions generate external costs and that the market supplies too large a volume of emissions relative to the efficient level. Consider, for example, Figure 2.1 below, which plots the aggregate marginal costs and benefits of abatement in the case where there is complete certainty. Marginal costs here are the aggregate incremental social costs of reducing emissions, including lost output, lower living standards, the cost

of developing new technologies, the cost of geo- and bio-sequestration, and so on. Marginal benefits are the marginal social benefits that might come about as a result of the effect that a lower stock of CO2-e in the atmosphere has on global temperatures.

In this diagram q* is the socially optimal quantity of abatement—it maximises net social benefits. The standard analysis assumes that in the absence of regulation no abatement would be produced. In such circumstances conventional economic theory argues that under a simple cap and trade scheme, the Government can issue an aggregate number of emissions permits at the desired level of the cap, and then allow firms to trade them so that the total costs of abatement can be minimised.

The basic idea is to create an artificial set of legal rights and obligations and allow those rights to be traded. To illustrate this idea, consider Figure 2.2 below, which plots the marginal cost of abatement for two firms, A and B. In the absence of regulation, it is assumed that firms produce no abatement. Now suppose that the Government sets a target of QA+QB tonnes of abatement. It issues emissions permits equal to the difference between the business as usual quantity of emissions and the desired abatement. Suppose that firm B is issued with a sufficiently high amount of emissions permits that it undertakes no abatement, whilst firm A is allocated no emissions permits. In the absence of any trade, B would undertake no abatement, whilst firm A would be forced to reduce its emissions by QA+QB. Notice that at this point, the marginal cost of abatement would be higher for A than for B.

Now suppose that A and B can trade these permits. At the point OB, the cost of abatement of the last unit

2.1. The Shergold Report (2007)

MB, MC

FIGURE 2.1: MARGINAL COSTS AND

BENEFITS OF CO2-E ABATEMENT

AN ECONOMIC EVALUATION | 13

MC*

Overall Abatement

MC*

MCA

OA OB QA

TCA TCB

QB OB

$/tonne of abatement

FIGURE 2.2: A CAP AND TRADE SCHEME

of emissions in industry B is less than the same cost in industry A. Thus B could offer to reduce emissions on A’s behalf, in return for A no longer having to do so. This is accomplished by B selling a permit to A at a price less than MCA at the point OB. Firm A would willingly pay B to do this, and B would willingly accept such a payment in return for abating one tonne of emissions. In other words, at this point there are gains from trade to be exploited. In the absence of transaction costs, permit trades take place until the allocation QA QB is reached. At that point, the equilibrium permit price will be MC*.

Note that a potential problem with a cap and trade scheme in practice may be the existence of high transactions costs in the permit market, which could exist due to the cost of arranging and negotiating trades, the costs of verifying permits and monitoring abatement activity, and the costs of enforcing the law. For example, consider Figure 2.3 below. Suppose that the initial allocation of permits is at point z. If transactions costs are equal to TC or higher, then no trade will take place, because the price that B is willing to accept plus transactions costs is equal to the price that A is willing to pay at that point. No gains from trade can be exploited because of the existence of transactions costs, and a cap and trade scheme fails to minimise the costs of abatement. If a

firm has market power in the permit market then costs will also fail to be minimised, for similar reasons.

The possibility that high transaction costs or market may emerge as issues in a future Australian permit market have never been seriously considered by Australian policymakers. Instead, the Shergold Report argued that free trade in permits would minimise the total costs of abatement, but without asking what the appropriate level of abatement actually was. The Report also argued that a cap and trade scheme was preferable to a carbon tax, because cap and trade focuses “on the ultimate environmental objective— namely, reducing emissions to a point that mitigates the effects of climate change” and that there would be more opportunities to link with other markets under a cap and trade scheme.7

MCA a

OBOA q*

A q*B

FIGURE 2.3: TRANSACTION COSTS

IN A CAP AND TRADE SCHEME

14 | AUSTRALIA’S CARBON TAX

2.2. The Garnaut Report (2008) and the Carbon Pollution Reduction Scheme (CPRS) (2009)

Prior to the election of the new Labor Government in late 2007 (whose policy platform included the introduction of a cap and trade scheme), Professor Ross Garnaut of the University of Melbourne was commissioned by the then Leader of the Opposition (and current Prime Minister), Mr Kevin Rudd, to report to Australia’s Federal and State Governments on “the possible ameliorating effects of international policy reform on climate change, and the costs and benefits of various international and Australian policy interventions on Australian economic activity.”8 Professor Garnaut’s report argued that “a well- designed emissions trading scheme has important advantages over other forms of policy intervention.” However, the report also argued that a carbon tax would be “better than a heavily compromised emissions trading scheme.”9 The Garnaut Review proposed a policy similar to the one that was eventually adopted: a cap and trade scheme with a short transition phase in which emissions permits would be sold by the Government at a fixed price, rather than being freely auctioned.

In July 2008 the Australian Government released a green paper10 on a proposed “Carbon Pollution Reduction Scheme” (CPRS), which outlined the major issues surrounding the establishment of a cap and trade system in Australia. The Government responded in December 2008 with a white paper entitled “Australia’s Low Pollution Future” (ALPF).11 This report committed the Australian Government to an unconditional reduction in CO2-e emissions of at least 5 per cent below 2000 levels by 2020, as well as a long-term emissions reduction target of 60 per cent below 2000 levels by 2050. It also proposed a cap and trade scheme for Australia, to begin on July 1, 2010, and was accompanied by a summary of the results of economic modeling by the Treasury Department of some of the costs of such a scheme.12

Following this series of reports, the Government introduced the Carbon Pollution Reduction Scheme (CPRS) Bill (2009) on May 14 2009. The CPRS proposal was for an initial fixed auction price (which is effectively a carbon tax) of $10 per tonne beginning in July 2011, transitioning to a full cap and trade scheme from July 2012. The CPRS passed the House of Representatives on June 4, but failed to pass the Senate. A second CPRS Bill passed through the

House of Representatives on November 16, 2009, but again failed to pass the Senate. Finally, a third CPRS Bill was introduced on February 2, 2010 and again passed the House of Representatives. However, on April 26, 2010 Prime Minister Rudd announced that the Government would delay the introduction of any scheme until the end of 2012, and the carbon tax moved off the Government’s legislative agenda. The 2010 Bill then lapsed in the Senate due to the calling of the 2010 Australian general election.

During the 2010 election campaign the Government promised that should it win the election it would not introduce a carbon tax in its next three year term.13 Instead, it proposed a number of alternative policies including a “Citizens’ Assembly” which would spend 12 months examining the evidence on climate change, the case for action and the consequences of putting a price on CO2-e emissions.14 However, the Government soon reneged on this promise. Following the 2010 election the Australian Labor Party formed a minority government with the Greens and some independents, and the Government established a Multi-Party Climate Change Committee to investigate “options for implementing a carbon price and to help build consensus on how Australia will tackle climate change.”15 The legislative and regulatory framework of the current tax, together with its design features, emerged from this committee.

Figure 2.4 summarises the legislative history of the CPRS scheme and how it evolved to take the form of the current tax.

The Australian public opposed the carbon tax at the time of its introduction. For example, a Morgan Poll on July 19, 2011 found that:16

• A majority of Australians (62%) agreed that “The carbon tax will have no significant impact on reducing the total world-wide volume of carbon dioxide put into the atmosphere” (34% disagreed).

• An overwhelming majority of Australians (75%) disagreed that “The $23 a tonne carbon price should be higher” while only 15% agreed that it should be higher.

• A majority of Australians agreed that “We should not have carbon tax until China and the USA have a similar tax.”

AN ECONOMIC EVALUATION | 15

• A plurality of Australians (49 per cent) disagreed with the statement that “The carbon tax is a good first step towards a market-based price on carbon.”

2.3. Carbon Taxes versus Cap and Trade Schemes: The Standard Treatment in the Literature

As in earlier reports, the Government has appealed to some basic economic principles to argue its case for the carbon tax. Consider again Figure 2.2. If the Government has perfect information, a carbon tax of t* per tonne (which allows the market to determine the quantity of emissions) can in principle be used to achieve exactly the same outcome as a cap and trade scheme in which permits are auctioned (and the market determines the price of a ton of emissions). At the point OB, for example, firm B it is not abating at all, and pays a tax of t* on all units of emissions. But if B had abated one tonne of emissions, its tax bill would be reduced by the tax cost of one tonne. Since the marginal cost of abatement is less than the tax for the first tonne, it has an economic incentive to abate this first tonne. Such an abatement incentive remains

up until the point at which B reduces emissions by qB. The same argument applies to industry A. At a tax or carbon price of t*, industry A has an incentive to abate exactly qA tonnes.

However, although with perfect information a carbon tax can be equivalent to a cap and trade scheme, in reality of course policymakers do not have perfect information. Yet neither the Shergold Report nor subsequent reports discussed in any great detail why a cap and trade scheme might be preferable to a carbon tax (or vice versa), or whether either of these is preferable to direct “command and control” mechanisms. Moreover, policymakers didn’t attempt to show that their recommended “solution” was better than the status quo, because they failed to conduct an accurate assessment of the marginal benefits and costs of their proposals relative to a plausible baseline in which Australian firms and households adapted to possible future climate change.

To see the importance of the (implicit) perfect information assumption, note that in Figure 2.2, if the

July 2012 Carbon tax beginsFebruary 2011

Government announces proposed architecture of carbon tax

August 2010 Government promises not to introduce carbon tax

April 2010 Government withdraws cap and trade legislation and commits not to introduce scheme before the end of 2012

May 2009 Cap and trade legislation introduced. Scheme to begin on July 1 2010

FIGURE 2.4: SUMMARY OF THE LEGISLATIVE HISTORY OF THE CARBON TAX

16 | AUSTRALIA’S CARBON TAX

government has complete information and knows that q* is efficient, direct command and control policies can achieve an identical outcome to a tax or a cap and trade scheme. If the government knew what the individual marginal cost curves looked like, it could simply force industry A to reduce its emissions by qA units, and force industry B to reduce its emissions by qB units. In other words, under conditions of complete policy certainty and perfect knowledge, an appropriately chosen carbon tax has the same outcome and welfare properties as both command and control and a cap and trade scheme.

The textbook analytical case for preferring a carbon tax or a cap and trade scheme over alternatives therefore rests on the superiority of those instruments in an environment in which policymakers have imperfect information. However, in such an environment the optimal policy choice is far from clear—information about the marginal benefits of abatement is needed to make a full determination. In a static setting, a cap and trade scheme may indeed minimise the costs of abatement for a give target of emissions reduction, but those costs may still exceed the benefits of achieving that target. Only a full cost benefit analysis can determine which policy is appropriate. Unfortunately, such an analysis has never been completed for Australia.

A standard result in the literature states that in the absence of international permit trading, if the marginal cost of abatement curve is very steep and the marginal benefit of abatement curve is relatively flat, then a carbon tax or fixed price scheme is preferred on the grounds that it has a lower expected deadweight loss. The intuition behind this result is as follows. Consider Figure 2.5, which is based on McKibbin and Wilcoxen (2002) and which plots the marginal social costs and marginal social benefits of abatement. Suppose that marginal benefits are known but that marginal costs are unknown but are believed to be MCLow. Under this belief, the efficient quantity of abatement is Q0. Suppose that the government has two choices: a cap and trade system which either auctions Q0 emissions permits (or allocates them freely to firms); or a carbon tax of t.

If costs actually turn out to be MCLow, then both policies are equivalent and are efficient. However, if marginal costs actually turn out to be high (MCHigh) then the efficient quantity ex post turns out to be Q1. But the cap and trade system—which fixes aggregate

emissions at Q0—results in a very high price and a large welfare loss triangle of DWL1. The reason for this welfare loss is simple: if marginal costs of abatement turn out to be high then efficiency requires that less abatement actually takes place than what was initially planned—but that cannot occur under the cap and trade system in which the aggregate quantity is fixed.

On the other hand, a fixed emissions price or carbon tax performs much better in this scenario. If marginal abatement costs turn out to be high, then the tax allows less abatement to occur, which is what is required. In Figure 2.5 if the tax is set at and marginal costs turn out to be MCHigh instead of MCLow , then the aggregate emissions abatement is equal to Q2, which is lower than the efficient level. Despite this, the welfare loss is relatively small (it is equal to the small triangle marked DWL2) because large costs are avoided whilst only small benefits are foregone. Under a carbon tax it is very unlikely that a fixed target will be met—but not meeting a target is an advantage, not a disadvantage. The point is that under these conditions, the economic consequences of not meeting a target are relatively small, whereas the economic consequences of fixing a target and

Price

DWL2

DWL1

MCHigh

MCLow

Abatement Q1 Q0

FIGURE 2.5: EFFICIENCY LOSSES FROM A CARBON TAX AND A CAP AND

TRADE SCHEME WHEN MARGINAL COSTS ARE UNEXPECTEDLY HIGH

AN ECONOMIC EVALUATION | 17

meeting it no matter what the cost could be quite severe. If marginal costs are rising steeply and are uncertain, it makes little economic sense to try to hit a precise target. Indeed, the more precise the target, the most costly the scheme is likely to be.

If the marginal benefit curve is steep, then the best policy would be to implement a policy that mimics such a curve. An aggregate fixed abatement target basically looks like a vertical marginal benefit curve and so is a better instrument in this case. Under those circumstances, hitting a target is desirable on economic grounds, not just because hitting a target is a good idea in itself.

Hence, the standard results in the literature are that:

• A carbon tax or cap and trade scheme with a price ceiling is preferred in circumstances where the marginal benefit curve is relatively flat and the marginal cost curve is relatively steep.

• A cap and trade scheme is preferred in circumstances where the marginal benefit curve is relatively steep and the marginal cost curve is relatively flat.

The basic lesson of this analysis is that the policy tool which more closely resembles the actual social marginal benefit curve will tend to work the best. Knowledge of the shape of the marginal benefit of abatement curve is therefore crucial in being able to decide which instrument is optimal. Most of the literature argues that the marginal benefit curve is relatively flat, for the following reason. There are infinitely many ways in which current emissions can be reduced over time to achieve some given future target for annual emissions. Any current and future benefits of abatement are related to the stock of greenhouse gases, whereas the current and future economic costs of abatement are related to flows of emissions (or more precisely, how those flows are restricted). This means that as a general proposition, current marginal

costs of abatement are likely to be sensitive to the current rate of emissions reductions, whilst current marginal benefits are likely to be relatively insensitive to current levels of reductions.

All of this means that missing a single annual emissions target has relatively low economic costs in future climate change damages (i.e. low foregone benefits). But the consequences of rigidly fixing a flow target that can then only be achieved by having a high marginal cost of abatement (and therefore a high price under a cap and trade scheme) are potentially quite severe. Therefore, a policy which focuses on the price (e.g. a tax, subsidy or cap and trade scheme with a price ceiling) is likely to be preferable to a policy that rigidly sticks to an aggregate quantity.

It is important to note, however, that the standard textbook result depends on a critical assumption: that the tax or cap are chosen at their ex-ante efficient levels (i.e. where expected marginal benefits equal expected marginal costs). If the tax or cap differ from these levels, then the welfare ranking can be reversed when any comparison is made. Intuitively, if the tax is set above the point where expected marginal benefits equal expected marginal costs—which could happen, for example, if the tax is retained for raising revenue, as opposed to climate damage mitigation—then there will be an additional, systematic welfare loss that is not caused by imperfect information or uncertainty, and this could offset any additional expected gain that the tax brings about.

As mentioned earlier, in Australia’s case there has never been a full assessment of costs and benefits of a carbon tax or a cap and trade scheme, or indeed any demonstration that either policy is better than the status quo. In particular, there has been no assessment of the likely position or shape of the marginal benefit of abatement curve. Hence, even if policymakers had accepted the standard result in the literature regarding the ranking of the two policy instruments, they lacked the information that would have enabled them to make a judgement about which policy was more desirable, or if either policy in practice (with the attendant rent seeking and other real-world considerations such as the “tax interaction effect,” discussed later in this paper) would be better than an alternative approach relying on adaptation.

Under a carbon tax it is very unlikely

that a fixed target will be met—but not

meeting a target is an advantage, not a

disadvantage.

18 | AUSTRALIA’S CARBON TAX

The legislation sets the initial level of the carbon tax, which commenced on July 1, 2012, at $23 per tonne CO2-e. The tax increased to $24.15 on July 1, 2013, which is nearly four times the current level of the EU permit price.17 The tax is legislated to rise again to $25.40 on July 1, 2014. Under the initial policy design, the tax was always designed to transition to a flexible price or cap and trade scheme in July 2015, with permit prices fluctuating with market conditions.18

During the fixed price period, permits cannot be traded or banked for future use, but banking is permitted during the flexible price period. The Government expects the permit price to more than double by the end of the next decade, reaching $57.61 (in 2013 dollars) by 2030. The use of international permits to meet liabilities is not permitted in the fixed price period. During the flexible price period firms may use international permits, subject to certain qualitative and quantitative restrictions. Importantly, until 2020, liable parties must meet at least 50 per cent of their annual liability with domestic permits. This restriction is due to be reviewed by the Climate Change Authority in 2016. The economic effect of the 50 per cent cap on overseas permits during the cap and trade phase is analysed in section 6 below.

In addition to establishing the initial level and coverage of the carbon tax, the CEF legislation establishes two new regulatory agencies:

• The Clean Energy Regulator (CER), whose responsibilities include overseeing the administration of the carbon tax, monitoring compliance and assessing the emissions of individual firms, enforcing payments of the tax, and determining eligibility for free permits, as well as overseeing auctions of permits during the flexible price phase.

• The Climate Change Authority (CCA), whose primary role is to provide advice and recommendations to the Government on important aspects of the carbon pricing mechanism, including future emissions caps. The Clean Energy legislation contains a “poison pill” arrangement in the form of punitive default emissions caps. These default caps are automatically activated if Parliament fails to pass regulations specifying the cap.19 The CCA was also established to provide advice on the role of the price floor and price ceiling beyond the first three years of the flexible price phase (see section 3.4 below). However, the price floor was abandoned less than two months after the carbon tax became operational.

3.2. Coverage: Who Pays?

Australia’s carbon tax applies to emissions of carbon dioxide, methane, nitrous oxide and perfluorocarbons from aluminium smelting. A threshold of 25,000 tonnes of CO2-e applies for determining whether a production facility is covered by the tax. Liable firms which emit but which do not surrender a permit must pay an emissions charge. The emissions charge in the flexible price period will be double the average price of permits for that year. In terms of sectoral coverage, the Australian scheme is very comprehensive, covering emissions from stationary energy, industrial processes, fugitive emissions (other than from decommissioned coal mines) and emissions from non-legacy waste. Agricultural and

3. Policy Framework and Key Parameters

The Government announced its “Clean Energy Future” (CEF) plan on July 10, 2011. The CEF policy

consists of a complex package of 18 different pieces of legislation. Despite popular opposition to the tax,

the CEF legislation was introduced into the Australian Parliament on September 13, 2011. The bills

passed the House of Representatives (with amendments) on October 12, 2011, and passed the Senate on

November 8, 2011. The package became law soon after, receiving Royal Assent on November 18, 2011.

3.1. Development of the Tax

Although the carbon tax only directly

affects around 370 businesses, the

economic incidence is far broader than

the narrow legal incidence.

AN ECONOMIC EVALUATION | 19

forestry emissions, as well as emissions from the combustion of biofuels and biomass (including CO2-e emissions from combustion of methane from landfill facilities) are not covered by the scheme.

Household transportation (i.e. fuel for personal vehicle use) is not directly covered by the scheme.20 However, as part of the CEF package, the Government imposed an effective carbon tax in relation to off-road business use of diesel fuel by reducing the existing diesel fuel tax credit.21 The carbon tax will be extended to the fuel used in trucks on July 1, 2014.

Although the carbon tax only directly affects around 370 businesses, the economic incidence is far broader than the narrow legal incidence. As a general rule, the economic incidence of any tax depends on the elasticities of demand and supply in the affected markets, with most of the economic incidence falling on the less elastic side of the market (usually consumers in the case of the carbon tax). If the tax affects the

production of exported goods where prices are determined by conditions in world markets (such as coal-mining), then the incidence of the tax will fall entirely on domestic producers. The carbon tax will therefore adversely affect consumer prices, real wages, investment, and GDP growth. These broader economic effects are examined in section 5 below.

3.3. Abatement Target and Sources of Abatement

The CEF plan proposed a carbon tax for three years and aimed for a reduction in emissions of at least 5 per cent compared with 2000 levels by 2020, and a reduction of 80 per cent below 2000 levels by 2050. It is important to note that these emissions reductions targets do not refer purely to domestic reductions or abatement which actually take place within Australia’s borders. Under the Government’s policy, Australia will only reach its overall target if Australian firms can purchase permits from overseas—in other words, if Australian firms pay businesses in other countries to further reduce their emissions. Under the policy, cumulative abatement relative to business as usual will be 16.7 Gt CO2-e by 2050. However, 9.3 Gt or 55.7 per cent of this total abatement is sourced from overseas jurisdictions, rather than domestically (see Figure 3.1). In other words, a significant part of the CEF policy involves Australian taxpayers paying other countries to reduce their emissions. As a result, along the price path that was originally projected by the Government, the purchase of foreign permits will involve a cumulative transfer of around $75 billion from Australian taxpayers to the rest of the world to 2050.

3.4. The Price Floor and Price Ceiling

For the first three years of the flexible price period, the Government’s original policy added two important institutional features to the planned cap and trade mechanism:

• A price ceiling of $20 above the expected international price, rising annually by 5 per cent in real terms. Domestic permit prices were not allowed to rise above this price ceiling; and

• A price floor of $15 rising annually by 4 per cent in real terms. Domestic permit prices were not permitted to fall below this price floor.

The originally anticipated price path, together with the expected floor and ceiling prices, are shown in Figure

60%

50%

40%

30%

20%

10%

0% Domestic Abatement

Overseas Abatement

FIGURE 3.1: SOURCES OF CUMULATIVE ABATEMENT RELATIVE TO

BUSINESS AS USUAL PROJECTIONS, 2013-2050

SOURCE: SGLP, CHART 5.2

20 | AUSTRALIA’S CARBON TAX

3.2. The Government’s original intention was that the price floor and ceiling were to be reviewed by the Climate Change Authority in 2017.

In principle, the introduction of a price cap and floor can improve the expected outcome under a cap and trade scheme. Consider Figure 3.3 which again plots marginal social costs and marginal social benefits of abatement. In this figure the marginal cost of abatement is uncertain. The Government introduces a cap and trade scheme to equate expected marginal benefits with expected marginal costs, and this target is fixed at Q0. There is a price floor of P and a price ceiling of P. Under this system, if either the price ceiling or price floor bind, then firms abate up to the point where marginal costs equal the price. If costs turn out to be lower than expected at MCLow, then Q4 is efficient. But without a price floor there will still only be abatement of Q0, and there is a welfare loss. If the price floor binds, then abatement of Q3 is

produced and there is a welfare gain of the lower shaded area, relative to the case where there is no price floor.

Similarly, if costs turn out to be higher than expected, then Q1 is efficient and under a standard cap and trade there would be a deadweight loss. However if there is a binding price ceiling in place, then less abatement (Q2) is produced, and there is a welfare gain of the upper shaded area relative to the case where there is no price ceiling.

The Government’s position on the floor price has never been clear. In 2011 the Government stated that “the floor is designed to reduce the risk of sharp downward movements in the price, which could undermine long-term investment in clean technologies.”22 However, on August 28, 2012, less than two months after the carbon tax had taken effect, the Government announced that there would not be a

0 2012-13 2013-14 2014-15 2015-16

Year

$/ to

nn e

C O

2- e

Price Ceiling

2016-17 2017-18 2018-19 2019-20

Baseline Price Path

Price Floor

FIGURE 3.2: BASELINE PERMIT PRICES, PRICE CEILING PATH

AND PRICE FLOOR PATH UNDER THE ORIGINAL CEF POLICY

SOURCE: AUSTRALIAN GOVERNMENT, SECURING A CLEAN ENERGY FUTURE, PAGE 27

AN ECONOMIC EVALUATION | 21

price floor after 2015-16, and that there would instead be direct linking with the EU cap and trade scheme. The price ceiling remains in place but it is proposed to end before mid-2018. When the Government announced that it was not proceeding with the floor price and would instead link with the EU scheme, it stated that this would provide “investors with long term certainty on the price of carbon pollution.” In other words, establishing a floor price was supposed to lead to less risk, but not establishing a floor price was supposed to provide long term certainty.

3.5. Other Policies

3.5.1. Complementary Emissions Reduction Policies

There are a number of other policies at both the Federal and State level which accompany the carbon tax. Most of these policies are intended to achieve the same or similar policy goals as the carbon tax (i.e. reductions in CO2-e emissions below business as usual levels). In addition to the large number of subsidies to alternative energy sources (such as solar and wind) that have remained in place after the tax was introduced, the most important “complementary” policies are Australia’s Renewable Energy Target (RET), the Clean Energy Finance Corporation (CEFC), and the Australian Renewable Energy Agency

(ARENA). All of these complementary policies show that the introduction of Australia’s carbon tax was not accompanied by the phase-out of inefficient, command-and-control policies, but in fact ushered in more of them.

The RET was implemented in August 2009 well before the carbon tax was introduced, and is an extension of the previous Mandatory Renewable Energy Target (MRET), which began under the previous government in 2001. The RET requires that by 2020, 20 per cent of Australia’s electricity must come from renewable sources. As of December 2012, 11.36 per cent of Australia’s annual electricity output in the National Electricity Market23 is sourced from hydroelectric power and other renewables.

The CEFC is a wholly government-owned entity that will siphon $10 billion taxpayer funds into renewable energy projects, energy efficiency schemes, and new technologies. The purpose of the CEFC is to provide debt and equity financing to projects which would otherwise not be sufficiently commercial to borrow on their own.

ARENA, which has funding of around $3 billion, provides financial assistance for “research, development, demonstration, deployment and commercialisation of renewable energy and related technologies”, as well as “storage and sharing of knowledge and information about renewable energy technologies.”24

The textbook analysis can be used to show that in the presence of a renewable energy target, a carbon tax or a cap and trade scheme will not lead to abatement at least overall cost. Consider Figure 2.3 again, and suppose that industry A is the renewable energy industry, which due to the presence of a renewable energy target must produce at least z tonnes of abatement. Then the outcome under a cap and trade scheme will be that the renewable energy industry will produce abatement exactly equal to z, and marginal costs of abatement fail to equalise across sectors, meaning that the overall cost of abatement is not minimised. Thus, although a textbook case for an “optimal” carbon tax or cap and trade scheme may be made, at least in the case of Australia, policymakers failed to act according to the textbook. The economic effect of these other policy instruments is discussed further in section 4.2 below.

Q1 Q2 Q0Q3Q4

Price MCHigh

PHigh

PLow

P P

MCMed

MCLow

Abatement

FIGURE 3.3: WELFARE EFFECTS OF A PRICE

FLOOR AND PRICE CEILING

22 | AUSTRALIA’S CARBON TAX

3.5.2. Household Compensation

In addition to these complementary measures, the Australian Government made a number of changes to Australia’s personal income tax system in an attempt to compensate households for increases in the cost of living caused by the carbon tax. The Government also increased payments, including pensions and family tax benefits. Although this compensation scheme involved lowering marginal tax rates for some taxpayers, to claw back revenue the Government had to increase marginal rates for around 2 million taxpayers. Furthermore, income tax cuts that were originally scheduled for 2015-16 were subsequently rescinded by the Government. Again, one of the chief arguments in favour of a carbon tax—that its revenues will be used to flatten and simplify the income tax system—did not come to fruition in the case of Australia’s carbon tax. The economics of these income tax changes is discussed further in section 4.2.2 below.

3.5.3. Free Permits

The other major component of the carbon tax is the Government’s “Jobs and Competitiveness” program, which allocates free carbon permits to businesses involved in emissions-intensive, trade-exposed industries (EITIs) such as aluminium production, steel manufacturing, pulp and paper manufacturing, glass making, cement production and petroleum refining.

Under this program, the allocation of free permits is determined as follows:

• Step 1: Determine whether an entity is trade exposed and emissions intensive

Under the carbon tax, the extent to which an industry is “trade-exposed” is determined by whether exports or imports as a share of the value of domestic production was greater than 10 per cent in either 2004-05, 2005-06, 2006-07 or 2007-08, or if there is a “demonstrated lack of capacity to pass-through costs due to the potential for international competition.” “Emissions intensity” is determined by whether the industry-wide weighted average emissions intensity of an activity is above a threshold of either 1,000 tonnes CO2-e per million dollars of revenue or 3,000 tonnes CO2-e per million dollars of value added.

• Step 2: Determine an “allocative baseline” for each firm.

Allocative baselines are determined by regulation, and take into account historic emissions and production information regarding emissions and production levels in 2006-07 and 2007-08. Baselines will not be updated over time as emissions intensities change.

• Step 3: Determine the share of the baseline each firm will receive as free permits.

Under the carbon tax, free permits are allocated according to Table 3.1.

These initial rates will be reduced by 1.3 per cent a year, and are not adjusted for future emissions levels. The economic effect of free permits, effective carbon prices and carbon leakage is discussed in section 5.3 below.

TABLE 3.1: ALLOCATION OF FREE PERMITS TO EMISSIONS INTENSIVE TRADE EXPOSED INDUSTRIES

EMISSIONS INTENSITY FREE PERMITS (% OF ALLOCATIVE BASELINE) ≥2,000 TONNES OF CO2-E/MILLION DOLLARS OF REVENUE OR ≥6,000 TONNES 94.5 OF CO2 E/MILLION DOLLARS OF VALUE ADDED

1,000-1,999 TONNES CO2 E/MILLION DOLLARS OF REVENUE OR 3,000- 5,999 TONNES OF 66 CO2 E/MILLION DOLLARS OF VALUE ADDED

AN ECONOMIC EVALUATION | 23

4. The Economic Costs of Australia’s Carbon Tax

The incremental costs to Australia per tonne of CO2-e abatement will likely exceed those of many other

advanced economies. One of the contributing factors to this higher cost burden is the fact that exports

account for a relatively high share of Australian’s GDP, and that these exports are relatively emissions

intensive.

To see how the emissions intensity of exports is likely to affect overall costs, note that under standard production-based approaches to measuring emissions reductions, CO2-e emissions that are produced within a country (rather than consumed within a country) are counted as part of that country’s emissions target under international agreements. Hence, emissions that are created in the process of producing exports are attributed to the exporting country, rather than the importing country.

The consequences for the costs of emissions reductions and trade are straightforward. Countries tend to export goods in which they have a comparative advantage. This means that countries export goods which can be produced at lower opportunity cost, and import goods which they can only produce at relatively high opportunity cost. If a country’s exports have a relatively high intensity of emissions, this means that the country has a comparative advantage in production of goods which are relatively carbon intensive.

It follows that in such a country, the opportunity cost of producing low emissions goods must be relatively high. Furthermore, reducing domestic emissions requires reducing production of goods that are currently exported, and switching production to less CO2-e emissions intensive goods. The costs of such a switch are likely to be greater for Australia compared to countries which enjoy a comparative advantage in the production of goods which are less emissions intensive.

There have been a number of recent empirical studies of the carbon intensity of production, consumption, exports and imports. The literature has used two basic methodologies:

• Emissions averaging approach: This approach takes the emissions intensity of each economy

overall, then examines the exports of each country and assumes that emissions intensity of exports is the same as the rest of the economy. Similarly, the studies examine the country-breakdown of imports from other countries and compute the emissions intensity of those imports.

• Input-Output Analysis (IOA) approach: This approach uses input-output analysis at the sectoral level. Input-output analysis is a way of tracking the inputs used by various sectors in the production of final goods, using fixed input coefficients. The IOA approach uses this method to directly compute the inputs used in the production of exports, and then applies an assumed emissions intensity coefficient to those inputs. This estimate is then used to obtain estimates of the emission intensity of exports (EEE), which is computed by dividing the emissions embodied in a country’s exports by total emissions produced. The same is done for imports, to obtain the emissions embodied in imports (EEI), which is computed by dividing the emissions embodied in a country’s imports by total emissions produced. Finally, the balance of emissions embodied in trade (BEET) is defined as difference between the EEE and the EEI.

Both approaches have advantages and disadvantages. The emissions averaging approach is relatively straightforward, but implicitly assumes that a

4.1. The Carbon Tax and Australia’s Exports

The consequences for the costs

of emissions reductions and trade are

straightforward. Countries tend to

export goods in which they have a

comparative advantage.

24 | AUSTRALIA’S CARBON TAX

country’s exports have the same emissions intensity as the rest of the economy, which is not the case in general. The IOA approach is more rigorous and detailed, and does not make the restrictive assumption of the emissions averaging approach. However, it is a less transparent method, and is less straightforward to compute estimates using this approach.

Estimates of the emissions intensity of exports and imports using the emissions averaging approach have been published for some countries, but not for Australia. On the other hand, there are some recent estimates in the literature of the emissions intensity of

exports and imports using the IOA approach. Figure 4.1 below, for example, reports estimates of the BEET computed by Peters and Hertwich (2008) for a range of countries. A positive BEET indicates that a country’s exports are relatively more intensive than its imports. In the Peters and Hertwich study, Australia’s BEET is measured as:

BEET = EEE – EEI = 31.4 – 14.9 = 16.5

Peters and Hertwich find that Australia has the highest BEET in the OECD (the next highest OECD country is Poland, with a BEET of 9.4).

-10

-20

-30

-40

-50

So ut

h A

fr ic

a R

us si

an F

ed er

at io

n Ve

ne zu

el a

M al

ay si

a In

do ne

si a

C hi

na A

us tr

al ia

C ze

ch R

ep ub

lic Th

ai la

nd B

el ar

us /U

kr ai

ne Po

la nd

Fi nl

an d

In di

a Tu

rk ey

Ta iw

an C

an ad

a A

rg en

tin a

B ra

zi l

M ex

ic o

G re

ec e

U ni

te d

St at

es Sp

ai n

K or

ea D

en m

ar k

Po rt

ug al

Ja pa

n Ita

ly G

er m

an y

Fr an

ce U

ni te

d K

in gd

om N

et he

rla nd

s Sw

ed en

B el

gi um

Country

16.5

B A

LA N

C E

O F

EM IS

SI O

N S

EM B

O D

IE D

IN T

R A

D E

FIGURE 4.1: PETERS AND HERTWICH (2008) ESTIMATES OF

THE BALANCE OF EMISSIONS IN TRADE

SOURCE: PETERS AND HERTWICH (2008), PAGE 1404.

AN ECONOMIC EVALUATION | 25

Peters and Hertwich (2008) use version 6 of the GTAP modelling database to derive their estimates. Davis and Caldeira (2010) use similar methods and an updated version of the GTAP database to derive estimates for a larger group of countries, including Australia. They find a BEET of 2.1 for Australia, ranking Australia fifth in the OECD behind Poland (9.7), Estonia (8.9), Canada (4.3) and Slovakia (3.2), but well ahead of larger OECD economies such as the US (-12.1), the UK (-45.6), Germany (-28.3), Japan (-21.6) and France (-43.4).

It is often claimed that Australia’s carbon tax will “achieve abatement at least cost.” The argument for the superiority of a cap and trade scheme or carbon tax over more direct abatement mechanisms in a static setting relies on the theoretical analysis presented earlier. However, there has been no direct evidence to demonstrate that this is the case. In reality, there are a number of reasons why it is unlikely that Australia’s carbon tax will achieve this objective. The remainder of this section examines these reasons.

4.2. The Interaction Between the Carbon Tax and Other Policies

4.2.1. Complementary Emissions Reduction Policies

As discussed in section 2.1 above, the main argument for a carbon tax or cap and trade scheme is that such instruments in principle allow the marginal costs of abatement to be equalised across firms and across sectors, which means that overall abatement is achieved at least cost. Under a carbon tax, equality of marginal costs of abatement is achieved by levying the same tax on each tonne of emissions, no matter where it is produced. Under a cap and trade scheme, equality of marginal costs is achieved by permitting free trade in permits.

The presence of complementary measures such as wind and solar subsidies, the RET, and the Clean Energy Finance Corporation means that achieving such equality of marginal costs is unlikely, if not impossible. Consider, for example, the RET which was discussed in section 3.5.1 above. To understand the costs of this scheme, consider Figure 4.2, where abatement is plotted on the horizontal axis and can be produced in two sectors, A and B. To reach the emissions target, QA+QB tonnes of abatement are required. Sector B is the renewable energy sector. Under an undistorted cap and trade scheme, permits

would be traded and the renewable energy sector would produce at the point where marginal abatement costs are equalised. Under a mandatory renewable energy target, however, sector B must produce QB tonnes of abatement. The marginal costs of achieving the last unit of abatement in B exceed the marginal costs of achieving the last unit in A. Hence the marginal costs of abatement are not equalised and abatement is not achieved at least cost. There is a welfare loss equal to the shaded triangle in the diagram. The only way such a scheme could achieve a different outcome is if it either (i) achieved more than QA+QB units of abatement, in which case the permit price would be zero (and indeed there would be no need for a cap and trade scheme) or (ii) if transaction costs or some kind of market failure in the permit market prevented trade in permits from achieving the efficient outcome (in which case a cap and trade scheme may also not be desirable).25

4.2.2. The Effects of Household Assistance and Income Tax Changes

Another common argument for introducing a carbon tax (or a cap and trade scheme in which permits are auctioned by the government) is that tax or permit revenues can be used to reduce existing distortionary taxes, such as personal income taxes. The “double dividend hypothesis” refers to the idea that there may

Target for Sector B

MCA

MCB

QA OBOA QB

MCB

$/tonne of abatement

FIGURE 4.2: THE COSTS OF A MANDATORY RENEWABLE ENERGY TARGET

26 | AUSTRALIA’S CARBON TAX

actually be two benefits from environmental taxation: the usual welfare gain that a Pigouvian tax brings about by reducing external costs, and an additional gain which comes about from the reduction in the welfare losses associated with existing taxes. However, it is unlikely that there is a double dividend in the case of Australia’s carbon tax for two reasons: a theoretical reason, and an empirical reason.

To understand why this is the case, we consider Figure 4.3, which models a simple situation in which there are two related markets, A and B, and assumed that A and B are complements. There is an existing distortionary tax of tAin market A, and consumption in market B causes a negative externality. Initial consumption and production levels in each market are Q1A and Q1B.

Now introduce a Pigouvian tax in market B. Since the goods are complements, introducing this tax leads to a reduction in the consumption of both goods. Consumption and production fall to Q2A and Q2B, and the tax in B creates a benefit equal to the shaded area in B. However, it also exacerbates the negative welfare effects of the existing tax in market A, leading

to a fall in welfare in that market equal to the shaded rectangle in market A. This is usually referred to in the literature as the “tax interaction effect” and it can partially (or even completely) offset any welfare gain in market B.

A reduction in the tax in market A will simply reduce this exacerbating effect, and so there is no sense in which there are “two” gains from introducing the Pigouvian tax.

The analysis also shows that if there are existing distortions in other markets and the goods are complements, then the optimal Pigouvian tax should be set at a level that is lower than the marginal external harm. More formally, letting be the change in welfare as a result of imposing the Pigouvian tax tB, MSC be the marginal social cost of activity B, and letting be the initial price in B, we have that the change in welfare in market B is:

0 Bp

PA 0

QA 2 QA

Market A

+ tA 0

DA’ DA

2 1

Begin at point “1” in each panel. There is an existing tax in market A, and no tax in market B

But activity declines in market A because we lose some revenue without reducing the tax. Overall welfare could even fall!

FIGURE 4.3: THE “TAX INTERACTION EFFECT”: THE WELFARE EFFECTS

OF A PIGOUVIAN TAX WHEN THERE IS A PRE-EXISTING DISTORTION IN A RELATED MARKET

PB 0

QB 1 QB

2 QB 1

MSC

+ tB 0 2

Now levy a tax in market B. We get the usual welfare gain from a Pigouvian tax in B.

Market A

( )0 BB B B B

QW MSC p t t

= +

AN ECONOMIC EVALUATION | 27

where is the slope of the demand curve for good B.

On the other hand, the change in welfare in market A is the change in revenue that occurs in that market, which is not offset by any gain in market A:

where is the shift in the demand for A when the

price of B rises.

The optimal (welfare maximising) Pigouvian tax is the one where the marginal welfare gain in market B just equals the marginal welfare loss in market A:

Rearranging this expression gives:

This expression tells us that:

• If there are no distortions in other markets, then the usual Pigouvian rule applies (set tax = marginal external harm);

• If there are existing distortions in other markets and the goods are complements, then the optimal Pigouvian tax should be set at a level that is lower than the marginal external harm; and

• If there are existing distortions in other markets and the goods are substitutes, then the optimal Pigouvian tax should be set at a level that is higher than the marginal external harm.

For Australia’s carbon tax the case of complements is the most empirically relevant one—the income tax is the most significant pre-existing tax in the Australian tax system, and activities that the carbon tax effects (such as electricity, fuel, and so on) are

Q t

A A A

QW t t

Q t

( )0 B AABB B B

Q QMSC p t t t t

+ =

External harm

Additional term

A A

B B B

Qt tt MSC p Q t

complementary to labour. Hence the likely effect of the carbon tax is that it will decrease labour supply, exacerbating the distortionary effects of the existing income tax system.

Whatever one thinks about the theory of the double dividend hypothesis, actually implementing the idea in practice requires marginal income tax rates to be reduced. As part of the carbon tax policy, the Australian Government made a number of changes to the personal income tax system to compensate for increases in electricity prices and other household costs. However, as section 5.1 below shows, since the carbon tax will reduce GDP below the level it otherwise would have been in every year that it is in place, full compensation for all taxpayers is not possible. Hence average income tax rates remained unchanged for a large number of taxpayers.

Table 4.1 below summarises the changes to statutory and effective marginal tax rates for Australian taxpayers as a result of the changes.26 As a result of the changes, effective marginal tax rates (EMTRs) fell for those on incomes between $16,001 and $20,542 (around 560,000 taxpayers), but increased for those on incomes between $20,543 and $30,000, as well as those on incomes between $37,001 and $67,000 (a total of around 2.2 million taxpayers). In other words, regardless of the textbook theory, in actual practice whilst average income tax rates for many Australians declined, more Australians saw their marginal income tax rates go up, rather than down, by a ratio of almost 4-to-1.

As Williams (2011) explains, this increase in marginal rates was needed as part of the compensation package because of the revenue cost of the Government’s decision to increase the tax free

Since the efficiency costs of taxation

increase with the square of the tax rate,

these changes are likely to have led to

significant overall efficiency costs, even

taking into account the fact that

marginal rates were reduced for some

taxpayers.

28 | AUSTRALIA’S CARBON TAX

threshold, which reduced marginal rates for some low income earners. To claw back some of the foregone revenue, marginal rates for those on higher incomes had to be increased. Under the carbon tax policy, marginal income rates were increased for those already facing reasonably high tax rates. Since the efficiency costs of taxation increase with the square of the tax rate, these changes are likely to have led to significant overall efficiency costs, even taking into account the fact that marginal rates were reduced for some taxpayers.

The crucial point is that that this increase in effective marginal tax rates is the exact opposite of what the double dividend theory recommends. In other words, instead of using the tax system to offset some of the negative welfare effects caused by the interaction between the carbon tax and the personal income tax system, the most likely effect of the Australian Government’s income tax changes is that they have actually made things worse and exacerbated the negative effects of those interactions. Moreover, none of these additional costs were taken into account in the Government’s modelling of the effects of the tax.

TABLE 4.1: NEW STATUTORY INCOME TAX RATES, OLD EMTRS AND NEW EMTRS

APPROXIMATE NEW CHANGE NUMBER OF

OLD STATUTORY NEW IN TAXPAYERS INCOME LEVEL EMTR RATE EMTR EMTR (MILLION)

$0-$16,000 0 0 0 NO CHANGE 0.28

$16,001-$18,201 0.15 0 0 FALL BY 0.15 0.28

$18,201- $20,542 0.15 0.19 0 FALL BY 0.15 0.28

$20,543-$30,000 0.15 0.19 0.19 RISE BY 0.04 1.38

$30,001- $37,000 0.19 0.19 0.19 NO CHANGE 1.19

$37,001- $67,001 0.34 0.325 0.34 NO CHANGE 3.58

$67,001- $80,000 0.3 0.325 0.325 RISE BY 0.025 0.83

$80,001- $180,000 0.37 0.37 0.37 NO CHANGE 1.19

>$180,001 0.45 0.45 0.45 NO CHANGE 0.18

SOURCE: WILLIAMS (2011).

4.3. International Linking and Restrictions on International Trade

4.3.1. The Gains from International Trade in Permits

The Australian Government has recently proposed moving to a flexible price “cap and trade” scheme a year earlier than originally intended, in 2014. In principle, this may help to achieve abatement at less cost. To see this, consider Figure 4.4 below, which analyses the economic effects of opening up the economy to trade in permits.

The figure plots the marginal private benefits and costs of emissions. Before trade, the domestic tax is tD and domestic emissions are E. Now the economy is permitted to trade permits with the rest of the world, at the world price of tW < tD. The overall emissions target is still E, but at the world price of tW domestic emissions now increase to ED. Domestic firms purchase ED – E from the rest of the world.

Note that the external costs of emissions are irrelevant here, due to the assumption that the target is fixed and so global emissions are the same before and after the economy opens up to trade. The economic effect

AN ECONOMIC EVALUATION | 29

is therefore similar to lowering an ordinary tax, taking into account the fact that part of the revenue is now transferred to foreign firms. The domestic economy gains the area a + b + c + d + e, which is the usual welfare gain from lowering a tax. There is a transfer to the rest of the world of b + d. Hence the overall welfare effect is a gain of a + c + e. Note that by similar triangles, b + c = c + e, and so the welfare gain

is simply , which is

one half multiplied by the difference between the domestic and foreign price, multiplied by the number of foreign permits purchased. Australia’s carbon tax in 2014 is scheduled to be $25.40. In the Government’s modelling, there are 15 million tonnes of overseas abatement in 2014. At an EU price of $6, using the above estimate we obtain a welfare gain from an early move to a link with the EU of $145.5 million, or less than 0.01 per cent of Australia’s expected GDP in that year.

Any estimate of the gains from allowing international trade in permits depends on the assumption that the European Union keeps the same number of permits in circulation. If instead the EU meets the increase in demand for permits by increasing supply, then the welfare gain for Australia would be much lower

because there would be less global abatement. If that is the most likely scenario, the optimal policy is simply to lower the carbon tax to the current EU price and not bother linking. Domestic firms emit the same amount as they would under the link to the EU, and global emissions would rise by the same amount as they would if the EU issued more permits. But the Australian government would earn additional revenue.

4.3.2. Constraints on Overseas Permit Purchases

Although there may be gains to Australia from linking to the European scheme, there are a number of reasons why such gains are likely to be smaller than those identified in the previous section. The primary reason is that under the Government’s policy, liable entities must meet at least 50 per cent of their annual liability with domestic carbon units. This section examines the effect of this constraint.

To understand the effects of this restriction, we simplify Figure 4.4 in Figure 4.5 by assuming that the marginal private cost of emissions are zero. In this diagram, Dpermits is the Australian demand for emissions permits and is derived from the marginal cost of abatement curve. Throughout we assume that the world permit price is pW and that this is lower than

MPC+tW

b c

MPC

MPC+tD

E EED

MC, P

FIGURE 4.4: INTERNATIONAL TRADE IN

EMISSIONS PERMITS

( )( )1 2 D D W

a b c E E t t+ + =

Price

Domestic Emissions

E0EW

Our website has a team of professional writers who can help you write any of your homework. They will write your papers from scratch. We also have a team of editors just to make sure all papers are of HIGH QUALITY & PLAGIARISM FREE. To make an Order you only need to click Ask A Question and we will direct you to our Order Page at WriteDemy. Then fill Our Order Form with all your assignment instructions. Select your deadline and pay for your paper. You will get it few hours before your set deadline.

Fill in all the assignment paper details that are required in the order form with the standard information being the page count, deadline, academic level and type of paper. It is advisable to have this information at hand so that you can quickly fill in the necessary information needed in the form for the essay writer to be immediately assigned to your writing project. Make payment for the custom essay order to enable us to assign a suitable writer to your order. Payments are made through Paypal on a secured billing page. Finally, sit back and relax.

Do you need an answer to this or any other questions?

About Writedemy

We are a professional paper writing website. If you have searched a question and bumped into our website just know you are in the right place to get help in your coursework. We offer HIGH QUALITY & PLAGIARISM FREE Papers.

How It Works

To make an Order you only need to click on “Order Now” and we will direct you to our Order Page. Fill Our Order Form with all your assignment instructions. Select your deadline and pay for your paper. You will get it few hours before your set deadline.

Are there Discounts?

All new clients are eligible for 20% off in their first Order. Our payment method is safe and secure.

Hire a tutor today CLICK HERE to make your first order

Print page

15 Jul [ECONOMY] a page

About Writedemy

We are a professional paper writing website. If you have searched a question and bumped into our website just know you are in the right place to get help in your coursework. We offer HIGH QUALITY & PLAGIARISM FREE Papers.

How It Works

Are there Discounts?

Hire a tutor today CLICK HERE to make your first order

Study Help

About Us