Evaluating the Policy Trade-Offs in ARB's Cap-and-Trade Program

Evaluating the Policy Trade-Offs in ARB’s Cap-and-Trade Program M AC TAy lo r • l e g i s l A T i v e A n A l y s T • Fe b r uA r y 9, 2012 An LAO RepORt 2 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt ExECuTivE SummARy The Global Warming Solutions Act of 2006 (Chapter 488, Statutes of 2006 [AB 32, Núñez/ Pavley]), commonly referred to as AB 32, established the goal of reducing greenhouse gas (GHG) emissions statewide to 1990 levels by 2020. In order to help achieve this goal, the California Air Resources Board (ARB) recently adopted regulations to establish a new “cap-and-trade” program that places a “cap” on the aggregate GHG emissions from entities responsible for roughly 80 percent of the state’s GHG emissions. The ARB will issue carbon allowances that these entities will, in turn, be able to “trade” (buy and sell) on the open market. A cap-and-trade program offers the potential to reduce GHG emissions more cost-effectively than traditional “command-and-control” regulations. Key Trade-Offs Inherent in Designing a Cap-and-Trade Program. In this report, we analyze the design of the cap-and-trade program as adopted by ARB and the important policy choices inherent in this design that have broad environmental, fiscal, and policy implications. The ARB has made these policy choices in the context of AB 32’s competing and potentially conflicting goals and requirements. In general, our analysis indicates that ARB has made a reasonable effort to balance various policy trade-offs, such as those involving (1) efforts to prevent the unintentional increase in GHG emissions outside of California (referred to as “emissions leakage”), (2) the use of offset credits, (3) actions to reduce volatility in the price of allowances and offset credits, (4) auction and market oversight, and (5) enforcement of cap-and-trade requirements. As we demonstrate in this report, there is no one “right” way to design such a complex program. Thus, the Legislature will want to carefully consider both potential changes to the design of the cap-and-trade program, as well as possible alternatives, depending on its priorities. We present various options for program changes that could be adopted to meet various legislative priorities. Significant State Revenues Planned to Be Raised. The ARB plans to auction (rather than give away for free) an increasing portion of carbon allowances over time. Annual revenues from the planned auctions will average in the billions of dollars. We discuss the legal constraints on the use of these revenues and the Legislature’s prerogative to determine the use of these revenues through its appropriation authority. Recommendations to Change Design and Operation of Cap-and-Trade. In this report, we identify some program design changes that would improve the cap-and-trade program, have relatively little downside from a policy standpoint, and would be consistent with the overall goals set forth in AB 32. Thus, if the Legislature determines that it wishes to proceed with a cap-and-trade program, we would recommend that the Legislature seriously consider the following modifications to ARB’s design of the program: (1) make producers of offset credits liable for offset project failures, (2) eliminate holding limits to improve the way the carbon market functions, and (3) reduce uncer- tainty about how and if the cap-and-trade program would operate after 2020. Potential Alternatives to Cap-and-Trade. If the Legislature decided not to proceed with the cap-and-trade program, it would need to look at alternatives for achieving the state’s goals under AB 32. We find that there are two main alternatives for achieving the GHG emissions reductions assumed in ARB’s cap-and-trade program: making changes or additions to direct command-and- control regulations that apply to GHG emitters and the imposition of some form of carbon tax. www.lao.ca.gov Legislative Analyst’s Office 3 An LAO RepORt inTROduCTiOn As part of a larger, legislatively mandated there is still an opportunity for the Legislature to effort to reduce emissions of GHGs, the ARB weigh in on these important decisions. recently adopted regulations to establish a Methodology. In preparing this report, we new program, known widely as cap-and-trade, reviewed the ARB rulemaking documents for that relies on market-based mechanisms to its cap-and-trade regulation, related analyses help reduce GHG emissions to 1990 levels in (including public comments on the rules), and California. Cap-and-trade constitutes one of the academic literature on market mechanisms. the most wide-ranging and complex regulatory We communicated with staff of the ARB; the efforts in the history of the state. As we will California Public Utilities Commission (CPUC); discuss in this report, the particular design of the the Office of Legislative Counsel; the South program chosen by the ARB involves a number of Coast Air Quality Management District (which important policy choices that have broad environ- operates an air quality-related market trading mental, fiscal, and economic policy implications. program); federal regulatory bodies, including Given the importance of the policy issues the Securities and Exchange Commission and the involved and what are likely to be the deep and Commodities and Futures Trading Commission; long-lasting effects of this new regulatory scheme, the Congressional Budget Office; and staff of this report examines in detail the specific policy market-based climate change initiatives, including choices made by the ARB in the design of the the Regional Greenhouse Gas Initiative and program, some specific policy trade-offs inherent the Western Climate Initiative (WCI). We also in those decisions, and alternative policies the communicated with various academics, industry Legislature may wish to consider before implemen- associations, and financial market trading tation of the cap-and-trade program. In our view, participants. BACkgROund global Warming and gHgs are commonly referred to as global warming or climate change. Greenhouse gases are gases that trap heat from the sun within the earth’s atmosphere, California’s Climate Change goals thereby increasing the earth’s temperature. Both The Global Warming Solutions Act of 2006, natural phenomena (mainly the evaporation of referred to as AB 32, established the goal of water) and human activities (principally burning reducing GHG emissions statewide to 1990 levels fossil fuels) produce GHGs. Scientific experts by 2020. Among various other requirements, have voiced concerns that higher concentrations the legislation directed ARB to develop a plan of GHGs resulting from human activities are by January 1, 2009, which encompasses a set of increasing global temperatures, and that such measures that, taken together, would enable the global temperature rises will eventually cause state to achieve its 2020 GHG reduction target. significant problems. Global temperature increases 4 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt Commonly referred to as the AB 32 Scoping Plan, on GHG emissions as well as a trading component it is required by law to meet a complex, and at whereby sources of GHG emissions may buy and times competing, set of requirements. The plan sell carbon allowances in order to comply with is to minimize costs and maximize benefits for the regulation. In adopting any market-based California’s economy, improve and modernize compliance mechanism, the ARB must consider California’s energy infrastructure and maintain the potential local impacts on communities that electric system reliability, maximize additional are already adversely impacted by air pollution. The environmental benefits, and complement the state’s ARB is required to design the market-based mecha- efforts to improve air quality. The law also requires nisms to prevent any increase in emission of toxic that regulations developed pursuant to AB 32 air contaminants or other air pollutants as well as minimize so-called emissions “leakage”—increases to maximize environmental and economic benefits in emissions of GHGs outside of the state that of such an approach for California. result from efforts to reduce emissions of GHGs Two Types of gHg-Related within the state—and not disproportionately market Based mechanisms impact low-income communities in California. A final Scoping Plan was approved by the ARB in Two types of GHG market-based mechanisms December 2008. are commonly discussed in academic literature: a carbon tax and a cap-and-trade program. While AB 32 Authorizes use of AB 32 only authorizes the use of a cap-and-trade market-Based Compliance mechanisms program, both of these types of market-based Traditionally, California has relied upon direct mechanisms have been used in other jurisdictions regulatory measures to achieve emissions reduc- to achieve GHG emissions reductions. Below, we tions and meet other environmental goals. Such briefly define the two types of market mechanisms, regulations, commonly referred to as command- explain the economic theory behind them, and and-control measures, typically require specific compare the theoretical benefits and costs of actions on the part of emissions sources to achieve these two mechanisms with each other and with the desired emissions reductions or other goals. For command-and-control programs. example, a direct regulatory measure may require Carbon Tax. A carbon tax amounts to a tax that a building meet a specified energy efficiency on each ton of carbon dioxide emitted, thereby performance standard as a means to reduce placing a new cost on emitting GHGs. Under emissions. In contrast, market-based mechanisms a tax, the regulator does not directly limit the provide economic incentives to achieve emissions amount of emissions that any emissions source reductions, without specifying how emissions may emit. Rather, the regulator would set the tax sources are to achieve those reductions. schedule such that, overall, the resulting amount In addition to a traditional regulatory of emissions would not be expected to exceed approach, AB 32 also authorizes, but does not targets. Thus, an emissions source would generally require, the ARB to include market-based experience greater costs, as a result of the tax, the compliance mechanisms as part of its portfolio of greater its emissions. Those sources that can reduce measures to meet AB 32 goals. Assembly Bill 32 emissions will presumably do so as long as the cost defines a market-based compliance mechanism as of making such reductions is less than the cost of a system that includes an annually declining limit paying the tax on those emissions. If that is not the www.lao.ca.gov Legislative Analyst’s Office 5 An LAO RepORt case, they would pay the tax. The overall level of emissions reductions and allowance purchases that emissions reductions can be achieved, in theory, at minimize their costs. the least cost possible because the tax provides an Each Approach Inherently Has Its Advantages economic incentive to all emissions sources subject and Disadvantages. There are major differences to the tax to find the mix of emissions reductions between a carbon tax and a cap-and-trade program and tax payments that minimizes their costs. regarding (1) the level of certainty provided to Cap-and-Trade. The second common type of regulated emissions sources about the cost of market-based mechanism that has been used to compliance and (2) the level of certainty for the reduce emissions is a cap-and-trade program. As regulator that the planned reduction in GHG with a carbon tax, a cap-and-trade program does emissions will actually be achieved. not directly require an individual emissions source A carbon tax provides relative certainty about to reduce its emissions. However, under a cap-and- the cost of compliance because the per-ton cost of trade program, the regulator issues one “allowance” emitting CO e gases is, by definition, the dollar 2 for each ton of carbon dioxide equivalent (CO e) amount of the per-ton emissions tax. However, 2 emissions permissible within the regulated area. there is less certainty with the imposition of An emissions source regulated under the program a carbon tax about the quantity of emissions must possess an allowance (or equivalent thereof) reductions that will result. Should regulators set for each ton of CO e emissions it produces in order the emissions tax too low, emissions may exceed 2 to comply with the regulation. Because the amount regulatory targets. If regulators set the emissions of allowances issued is less than the amount of tax too high, then regulated emissions sources may emissions that would otherwise be produced, the act to reduce emissions beyond what is required to effect of the program is to lower overall emissions. meet the targets. A cap-and-trade program differs from a In contrast to a carbon tax, a cap-and-trade carbon tax in that the cost of emitting each ton program provides relative certainty to the regulator of CO e is not decided by the regulator. Rather, about the reduction in GHG emissions that will 2 the cost is determined, in effect, by the emissions be achieved. By definition, the total number of sources themselves through trading of emissions tons of CO e emitted by regulated sources cannot 2 allowances. The supply and demand of allow- exceed the amount of emissions allowances issued ances in a trading market determine the price by the regulator. However, because the price of an of an allowance. Parties that can reduce their allowance is determined by market forces, the cost emissions are likely to do so as long as it is cheaper of compliance for an emitter is less certain under a than buying allowances at current prices. (When cap-and-trade program. emissions reductions result in a party holding more Command-and-Control Measures Usually allowances than it needs for compliance, excess Less Cost-Effective Than Market Mechanisms. allowances can be traded with others who find it Economic theory indicates that either a carbon tax less costly to buy allowances rather than reduce or a cap-and-trade approach has lower compliance their emissions.) As with the carbon tax, the level costs for emitters collectively than direct regulatory of overall emissions reductions is achieved, in measures. Figure 1 provides a graphic illustration theory, at the least cost possible. This is because the of the theoretical potential of a carbon tax and allowance price provides an economic incentive to a cap-and-trade program to achieve the same all regulated emissions sources to find the mix of emissions reduction at potentially lower total 6 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt Figure 1 The Theoretical Potential of Market-Based Mechanisms: Three Ways to Cut Emissions in Half A Command and Control Policy Firm A: High Cost ($4 per ton to reduce emissions) Firm B: Low Cost ($2 per ton to reduce emissions) 4 tons of emissions 4 tons of emissions Government requires each firm to cut emissions in half. Total Cost of Reduction: $12 Reduces emissions by 2 tons. Reduces emissions by 2 tons. Total Remaining Emissions: 4 Tons 2 tons of emissions 2 tons of emissions Cost of reduction: $8 Cost of reduction: $4 Payment to government: $0 Payment to government: $0 Market-Based Mechanism #1: A Carbon Tax 4 tons of emissions 4 tons of emissions Government sets a tax of $3 per ton. Total Cost of Reduction: $8 $3 tax is less than $4 cost to reduce: $3 tax is more than $2 cost to reduce: pays tax, does not reduce emissions. pays no tax, eliminates emissions. Total Remaining Emissions: 4 Tons Total Payment to Government: $12 4 tons of emissions 0 tons of emissions Cost of reduction: $0 Cost of reduction: $8 Payment to government: $12 Payment to government: $0 Market-Based Mechanism #2: A Cap-and-Trade Program 4 tons of emissions 4 tons of emissions Government introduces a fixed quantity of allowances. Market allows buying and selling of allowances. Price of $3 per allowance results from buying and selling. Allowance price is less than Allowance price is greater than $4 cost of reduction: buys 4 allowances, $2 cost of reduction: buys no allowances, Total Cost of Reduction: $8 does not reduce emissions. eliminates emissions. Total Remaining Emissions: 4 Tons a Total Payment to Government: Varies Allowance 4 tons of emissions 0 tons of emissions Cost of reduction: $0 Cost of reduction: $8 Payment for allowances: $12 Payment for allowances: $0 a This depends on whether the allowances were initially auctioned or given away for free. The auction would result in payments to the government. www.lao.ca.gov Legislative Analyst’s Office 7 An LAO RepORt cost than a command-and-control program. The emissions can potentially minimize their costs by potential for lower total compliance costs under choosing not to reduce their emissions, instead market mechanisms stems from the fact that the deciding to buy allowances (under cap-and-trade) regulated emissions sources can be expected in or pay the tax (under the carbon tax). On the other many cases to have better information about which hand, emissions sources that can reduce their compliance strategies minimize costs for them than emissions relatively cheaply are given an economic even the best-informed regulator could. Emissions incentive to do so, as an alternative to buying sources facing relatively high costs to reduce allowances or paying the tax. THE CAP-And-TRAdE PROgRAm AS dESignEd By ARB ARB’s Scoping Plan includes a downward adjustments were made to the emissions Cap-and-Trade Program reductions estimated for each of the Scoping Plan measures. As AB 32 did not authorize a carbon tax, Below, we discuss the major features of the ARB included a cap-and-trade program in the cap-and-trade program as designed by ARB and AB 32 Scoping Plan as a market-based compliance how the program is intended to work. mechanism. This is in addition to various direct regulatory measures referenced in the Scoping Plan, The Concept of the Cap such as regulations reducing the carbon content of Cap-and-Trade Places Emissions Cap on fuels sold in California or requiring generators to Certain Sectors of the Economy. The ARB’s increase the amount of the electricity supplies they cap-and-trade program is designed to limit or receive from renewable sources to 33 percent of their cap the aggregate amount of GHGs emitted from total. As a package, the Scoping Plan measures are emissions sources that collectively represent intended to collectively lower the state’s GHG levels roughly 80 percent of the state’s total GHG in 2020 from what they otherwise would be (often emissions. While they are not assigned an referred to as the “business-as-usual” scenario) to individual emissions reduction target, entities that the 1990 level. The ARB has estimated the 1990 emit at least 25,000 metric tons or more of CO e per level to be 427 million metric tons of carbon dioxide 2 year are subject to the cap-and-trade regulation and equivalents (MMTCOe). This number is therefore 2 are therefore considered to be a “covered” entity. the 2020 emissions limit—an aggregate statewide When the program is fully operational, approxi- limit. The difference between the estimate of 2020 mately 350 of the state’s largest emission sources business-as-usual emissions and the 2020 emissions will be covered entities, including oil producers, limit is therefore the emissions reduction target of refiners, and electricity generators, as well as other the Scoping Plan. In the 2008 Scoping Plan, the total large industrial entities. Covered entities and emissions reduction target was 174 MMTCOe. In 2 their customers within California are collectively 2010, the ARB adjusted this number downward to referred to as “capped sectors” or “the capped 80 MMTCOe, in part to account for the economic 2 economy.” The remaining 20 percent of GHG downturn’s impact on emissions levels. Similar 8 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt emissions come from entities in other economic allowances and offset credits are referred to as sectors such as agriculture and forestry. These compliance instruments.) The ARB intends to phase sectors are referred to as the uncapped sectors and in the sectors of the economy that are covered under are not subject to the cap-and-trade regulation. the cap-and-trade regulation and ultimately reduce As noted earlier, the overall goal of AB 32 is emissions by reducing the annual supply of new to reduce GHG emissions from both the capped allowances over the course of the program. and uncapped sectors by 2020 to the 1990 level Allocation of Allowances and of 427 MMTCO e. As the capped sectors emit 2 use of Compliance instruments 80 percent of the state’s GHG emissions, ARB intends these sectors to reduce their GHG Some Allowances Auctioned, Some Given emissions to around 340 MMTCO e by 2020 Away for Free. An essential component of the 2 in order to meet the AB 32 emissions target. As cap-and-trade program design is deciding how to discussed further below, the cap on the aggregate place allowances into circulation so they can be emissions of covered entities is designed to decline acquired by those who will need to use them for over time to result in the planned level of aggregate compliance. Generally speaking, allowances could emissions from the capped economy in 2020 to be allocated in one of three ways: (1) they could be meet AB 32’s emissions target. given away for free, (2) they could be auctioned, or (3) some portion could be freely allocated while the High-Level Overview of other portion is auctioned. All of these approaches Compliance With the Cap yield the same programmatic results in terms of As a step toward enforcing compliance with GHG reductions. the cap-and-trade program, the ARB will require The ARB intends to do a combination of covered entities to report their GHG emissions auction and free allocation of allowances. Initially, annually based on mandatory reporting require- a majority of allowances will be allocated for ments. Covered entities within the capped sectors free. Between now and 2020, the ARB estimates can comply with the regulation by obtaining one that it will give away approximately 430 million allowance for each ton of COe that it emits during allowances (each allowance is for one ton of CO e) 2 2 a particular compliance period. The covered entity to some industrial sources in order to reduce the must turn in allowances to ARB that match the level competitive disadvantage to those sectors that are of its reported emissions for the compliance period. subject to the cap-and-trade regulation. The intent The first opportunity to obtain allowances will either is to reduce what is called economic leakage—the be through ARB’s free allocation or through ARB’s decision by firms to relocate outside of California as allowance auction. After the initial auction, covered the result of a perceived competitive disadvantage entities will have the opportunity to obtain allow- imposed by the cap-and-trade policy. In addition, ances in the carbon market (discussed in more detail the ARB will provide electricity distribution later). In addition, covered entities will be allowed to utilities free allowances to reduce the cost burden use a relatively small portion of offset credits—which on electricity users from electricity price increases are derived from GHG emission reduction projects expected to result from the implementation of the that are undertaken by emissions sources not subject cap-and-trade program. Also, while no estimates to the cap-and-trade program’s GHG emissions have yet been provided, the ARB has indicated that cap—to comply with the regulation. (Collectively, there will be some consideration given to providing www.lao.ca.gov Legislative Analyst’s Office 9 An LAO RepORt free allowances to the natural gas distribution on energy in the production or distribution sector once these entities are phased into the of their products as well as their exposure to program in 2015. out-of-state competition. It then classified covered In total, between 2012 and 2020, the ARB entities as either high, medium, or low risk of will make available up to 2.5 billion allowances, leakage. According to the ARB, the sectors that it with roughly 50 percent auctioned and 50 percent determined are at high risk of competitive disad- given away for free. The amount of allowances that vantage include oil and gas extraction, cardboard ARB puts into circulation is controlled by ARB manufacturing, and the manufacturing of certain over time to move the state towards AB 32’s 2020 chemicals such as fertilizers. Medium-risk sectors emissions target. Please see the nearby box for include food processing, sawmills, and petroleum ARB’s plans for auctions and allowance allocations product manufacturing. Low-risk sectors include in 2012-13. pharmaceutical, medicine, and aircraft manufac- Determination of Free Allowance Allocation turing. The provision of free allowances will to Leakage-Prone Industries. As mentioned above, continue longer, and at higher levels, for entities the ARB has chosen to allocate allowances for free in sectors determined to be at higher risk. The to some covered entities to reduce a competitive allocation of free allowances is also based on an disadvantage to them potentially resulting from entity’s prior output. Within those sectors that the cap-and-trade program. In order to determine receive free allowances, the more of a product, which industries may be competitively disadvan- such as cement, that an entity produces, the more taged as a result of the cap-and-trade program allowances it will generally get for free. The ARB and therefore may be at risk of leakage, the ARB has committed to adjust its free allocation policies evaluated covered entities’ degree of reliance based on its ongoing judgments about such issues The Air Resources Board’s (ARB’s) Auction and Allowance Allocation Plans for 2012-13 2012-13 Allowance Auctions. The ARB intends to hold quarterly auctions of a set number of allowances beginning in 2012. In August of this year, it plans to auction 20 million allowances for use in 2015 or beyond (“vintage 2015” allowances). A similar auction will be held in November in which 20 million vintage 2015 allowances will be made available. By auctioning these future- year allowances, ARB intends to provide greater transparency to the market regarding potential future prices in order to provide covered entities more information to use in planning for future compliance with the regulation. In February 2013, ARB plans to auction 3 million current-year allowances as well as an additional 10 million vintage 2016 allowances. In May 2013, ARB plans to hold a similar auction where another 3 million current-year allowances as well as an additional 10 million vintage 2016 allowances will be offered. 2012-13 Free Allowance Allocation. In 2012-13, ARB plans to allocate approximately 150 million free allowances to some sectors of the economy, including electric utilities and some large industrial emitters, in part to minimize leakage. Specifically in 2012-13, electricity distribution utilities will receive almost 100 million allowances. Of this number, 65 million allowances will be given to the state’s Investor Owned Utilities, which must then sell their allowances at auction. 10 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt as the acceptable impacts of the cap-and-trade volatility in compliance instrument prices. If an program on business competitiveness. unexpected shortage in compliance instruments Offset Credits Can Be Used in Lieu of caused prices to spike, parties that had banked Emissions Reductions. The ARB’s cap-and-trade these instruments would have a strong incentive to regulation allows the use of offset credits as a sell the ones they had set aside, putting downward means to comply with the cap on emissions. pressure on prices. Banking also potentially Offsets refer to GHG emissions reductions from provides an incentive for covered entities to make projects that are undertaken by emissions sources early reductions in their GHG emissions. Firms not subject to the cap-and-trade program’s GHG that believe that compliance instruments will emissions cap. These projects are developed in lieu become more valuable in the future will be more of direct emissions reductions by sources subject likely to invest in making emissions reductions to the cap. For example, the owners of a power today, thereby freeing up compliance instruments it plant with emissions covered by the cap-and-trade could sell to others down the line. program may pay a dairy (which is not otherwise Banking and the use of banked compliance regulated) to reduce its emission of GHGs. Under instruments thus give entities important flexibility the ARB’s offset program, the owners of the power regarding the speed at which they reduce their plant would be credited for the GHG emissions emissions. Banking is limited to some degree, reductions realized by the dairy. The power however, by ARB’s limit on the number of allow- plant owners would, in effect, pay for the dairy ances that any one entity can hold for trading to offset its emissions if that would be cheaper purposes, as discussed later in this report. than reducing their own GHG emissions. Thus, ARB’s Plan includes Phased-in Approach the use of offset credits allows parties subject to With gradually declining Cap the cap-and-trade program to lower their cost to comply with the regulation. The ARB will use a phased-in approach to The ARB’s rules currently allow for offset its cap-and-trade program in which there will be projects in four areas: forestry, urban forestry, dairy three distinct compliance periods between now and methane digesters, and prevention of the release 2020. The three compliance periods will encompass of ozone-depleting substances (such as refrig- 2013-14, 2015-17, and 2018-20, respectively. erants) into the atmosphere. The cap-and-trade In the first compliance period, only electricity regulation allows for offset projects anywhere in generators and large industrial sources will be the United States. However, the ARB’s regulation subject to the cap-and-trade program. Beginning allows no more than 8 percent of a covered entity’s with the second compliance period, fuel suppliers compliance obligation within each compliance will be added to the entities that are subject to the period to be met with offset credits. (The remainder regulation. As shown in Figure 2 (see next page), must be met with allowances.) from 2015 (when the cap-and-trade program is Compliance Instruments Can Be “Banked” fully phased in) to 2020, the amount of aggregate for Later Use. The ARB allows some banking—the annual emissions allowed from covered entities (the carryover of compliance instruments from one cap) gradually declines from just over 400 million compliance period to any future compliance tons to 341 tons.As the cap declines, allowances are period—as part of its cap-and-trade program. The likely to become more scarce which, in turn, will ability to bank compliance instruments may limit likely increase the cost of allowances. www.lao.ca.gov Legislative Analyst’s Office 11 An LAO RepORt Figure 2 Annual “Caps” on Emissions of Capped Economy (in MMTCO e) 2 Approximate Forecast BAU Emissions Cap as Percentage Annual Capa From Capped Economy of BAU Emissions 2015 407 407 100% Compliance Period 2b 2016 398 407 98 2017 386 408 95 2018 363 408 89 Compliance Period 3 2019 352 408 86 2020 341 409 83 a For each year, reflects allowances first available for use, assumes maximum allowable use of offset credits, and assumes a particular limited amount of allowances purchased from the reserve. b Compliance Period 1 not shown. Entire capped economy is first covered in Compliance Period 2. MMTCO2e = million metric tons of carbon dioxide equivalent gases and BAU = Business-as-Usual scenario. Cap-and-Trade Serves as a Backstop The actual emissions reductions achieved for gHg Emission Levels under the cap-and-trade program, however, could be significantly different than shown in this figure. In Figure 3, we show how the mix of measures That is because the ARB has designed the cap-and- in the Scoping Plan—including both direct trade program to serve as a “backstop” to achieve regulatory measures and cap-and-trade—are GHG emissions targets in the covered sectors. Any intended to achieve the aggregate emission underperformance of direct regulatory measures at reduction target by 2020. At full implementation of reducing emissions in effect will result in additional the Scoping Plan, cap-and-trade is now expected to reductions under the cap-and-trade program. In contribute the equivalent of 18 MMTCO e in reduc- 2 other words, the cap of the cap-and-trade program tions in GHG emissions annually by 2020 compared serves as a backstop for GHG emissions, regardless with 62 MMTCO e from direct regulatory 2 of the performance of the direct regulatory measures measures. As the figure shows, the sectors covered in achieving their estimated emissions reductions. under the cap-and-trade program (referred to in the For example, in the electricity sector, electricity figure as the capped economy) are also subject to consumption and GHG emissions are supposed various direct regulatory measures. Figure 3 Scoping Plan Forecast of GHG Emissions (in MMTCO e) 2 Emissions in 2020 Planned Reductions in Emissions in 2020 Direct Business-As-Usual AB 32 Target Regulatory Scenario (1990 Emission Levels) Totals Measuresa Cap-and-Tradeb Capped Economy 409 341 68 50 18 Uncapped Economy 98 86 12 12 — Totals 507 427 80 62 18 a For the capped economy, includes measures such as the 33 percent Renewables Portfolio Standard, the Low Carbon Fuel Standard, and energy efficiency programs. For the uncapped economy, includes measures to address high global warming potential pollutants and sustainable forest practices, among others. b If the direct regulatory measures do not result in these planned emissions reductions from the capped economy, total emissions reductions will still be achieved due to the cap- and-trade program. GHG = greenhouse gas and MMTCO2e = million metric tons of carbon dioxide equivalent gases. 12 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt to be reduced through the implementation of the The 2012-13 Governor’s Budget assumes that energy efficiency programs included in the Scoping cap-and-trade auctions will generate $1 billion in Plan. If, however, these programs fail to meet their state revenues in 2012-13. Under the administra- planned emissions targets, electricity generators or tion’s plan, these revenues would be invested importers would have to either take additional steps in (1) clean and efficient energy, (2) low-carbon to reduce their emissions or purchase additional transportation, (3) natural resource protection, compliance instruments to meet their cap-and- and (4) sustainable infrastructure development. trade compliance obligations. The budget also assumes that $500 million of the If in the future it appeared that the number of revenues will be used to offset General Fund costs allowances ARB plans to introduce into the carbon of existing programs. According to the adminis- market would likely allow emissions in 2020 to tration, since actual cap-and-trade revenues will exceed the AB 32 target, ARB would have to take not be known until late in 2012-13, the planned corrective actions to ensure the target will be met. expenditures are not specified by program in the For example, the size of allowance auctions or proposed budget. Rather, the administration pGlanrsa phic Sign Off giveaways in future years could be reduced. to submit an expenditure plan to the Legislature Secretary after the first cap-and-trade auction—which would Auction Revenues Analyst be after the 2012-13 budget is enacted—and allocate Director Billions of dollars in revenues from the auction funds to specific programs not sooner than 30 days Deputy of allowances will become available as a result of after submitting this plan. the ARB’s cap-and-trade program. The amount of revenues could range greatly, as shown in Figure 4, depending upon the cost of directly reducing GHG Figure 4 emissions, the state of Significant, But Greatly Varying, the economy overall, and Auction Revenues Possible other factors. The range 2012 Dollars (In Billions) of revenues shown in the $16 figure is based on ARB’s plan to auction a certain 14 portion of allowances 12 rather than giving them away for free, as well as 10 Possible range of revenues ARB’s targeted price range expected under current regulationsa 8 for allowances. In 2012 6 and 2013, the ARB targets a price range between $10 4 and $50 per allowance. 2 These price targets are adjusted upwards over 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20 time. a Range reflects the essentially fixed quantities of allowances the Air Resources Board will auction each year and ARB’s targeted lower and upper prices for allowances. www.lao.ca.gov Legislative Analyst’s Office 13 ARTWORK #110569/Figure 4 - Significant Auction Revenues Possible and Possible Range is Large An LAO RepORt Linkage With Other Jurisdictions’ rules for compliance. In an unharmonized world, Cap-and-Trade Systems regulated entities would pick and choose whichever jurisdiction’s rules best serves their economic In developing the cap-and-trade regulation, interests, even if this goes against the design ARB indicated that it plans to link with other decisions and policy priorities of their “home state.” cap-and-trade programs, namely those in the Without such harmonization, unintended adverse WCI—a consortium of Western states, Canadian impacts, economic or otherwise, may result. provinces, and Mexican states. Linkage would For example, if California’s and Quebec’s offset mean that compliance instruments certified or protocols are not harmonized, the ability of offsets issued by any linked jurisdiction would be accepted to serve as a cost-containment mechanism for the for compliance purposes by all linked jurisdictions. program may be diluted and emission reductions While many of the members of WCI have either from offset projects may be less certain. As another postponed or are further behind the regulatory example, to the extent that Quebec’s cap on its development process, Quebec is one member of covered entities is more stringent than California’s, WCI that is on track to link with California prior this may increase the scarcity of allowances, which to the first auction, which is scheduled for August would serve to increase overall allowance prices 2012. Linking with other jurisdictions’ cap-and- for all covered entities and potentially increase the trade programs could serve to contain aggregate compliance cost for California’s covered entities. program compliance costs by providing more Legislative Oversight of ARB’s Linkage Plans opportunities for low-cost emission reductions. Will Be Important. To the extent that linking It does, however, potentially raise both legal and would expand the market and provide a greater economic questions. number of opportunities for low-cost emissions First, in order to formally link with another reductions, harmonizing and linking with other jurisdiction, ARB must first go through the jurisdictions’ cap-and-trade programs could serve formal rulemaking process on a jurisdiction-by- to contain overall compliance costs. However, we jurisdiction basis. Under state law, such rulemaking will not know how well programs proposed to be is subject to public hearing and notice requirements linked are harmonized nor will we have a clear as well as requirements for an economic impact understanding of the potential economic and other analysis. While it has not officially filed a notice impacts of such action until an analysis has been of proposed rulemaking to link with Quebec, conducted of the board’s particular proposals to ARB has indicated that it intends to open such a link California with cap-and-trade programs in proceeding in the spring of 2012. Legal questions other jurisdictions. have been raised regarding California’s ability to As ARB must include an economic impact legally enter into a compact or agreement with the analysis in its initial statement of reasons for any province of another country. formal rulemaking on linking, the Legislature Second, in order to effectively link California’s will have an opportunity to evaluate such analysis, cap-and-trade program with another jurisdic- determine if linking with another jurisdiction is tion’s program, California’s cap-and-trade rules indeed in the state’s best interest, and provide any should be harmonized with the rules of the other necessary policy direction to ARB on this issue. jurisdiction, ensuring that covered entities in both jurisdictions are subject to equally stringent 14 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt Cap-and-Trade Program to participate, all interested parties must be registered give Rise to multiple Carbon markets with ARB. While ARB has set rules governing market The introduction of emission allowances and participation, it will not directly operate the offset credits that are designed to be tradable gives trading markets. Rather, these trades will take place rise to what is known as a carbon market. The through privately operated exchanges or in “over- carbon market will consist of a number of distinct the-counter” trading directly between parties. The but interrelated markets. The ARB’s allocation or ARB will, however, require that information on a auction of emission allowances, as well as the ARB’s trade in these markets be reported to it for input development and certification of offset credits, into a tracking system before the trade can be will take place in what is commonly referred to completed. And, while ARB will share an oversight as the “primary market.” There are also so-called role, the bulk of the oversight responsibility will fall “trading markets” where trading activity related to third parties with whom ARB will contract. We to compliance instruments will take place. These discuss this approach to oversight in more detail include the secondary market (where compliance later in this report. instruments are traded directly) and the derivatives The carbon market will play a pivotal role in market (which involves the trading of financial the cap-and-trade program. As previously noted, contracts, primarily for hedging and investment, one advantage of a market mechanism is its the value of which depends on the market behavior potential to reduce emissions at a lower cost than of compliance instruments). a traditional regulatory approach. However, in The ARB has set rules regarding who may order to facilitate lower economy-wide compliance participate in auctions and in the trading markets, costs, the carbon market must function well. A with the exception of the derivatives market. (The well-functioning carbon market is one that allows ARB is of the view that it lacks the authority to for broad participation and allows participants govern participation in the derivatives market.) As to easily buy and sell compliance instruments at noted earlier, market participation is not limited sufficiently predictable prices that accurately reflect to covered entities. Non-covered entities and costs of abatement in the capped economy. As we other interested parties are generally permitted to will discuss later, to help its carbon market function participate as well. Some parties may participate in well, the ARB has established rules and processes order to reduce the level of allowable emissions by to help prevent abuse and to allow the punishment buying compliance instruments and making them of fraudulent activity. These rules and processes unavailable for use. Only entities with a potential include efforts to establish clear legal jurisdiction conflict of interest, such as third-party verifiers over market participants, ban entities with market (entities and individuals who are responsible for oversight or offset verification roles from trading, auditing and verifying emissions reductions), are and punish entities that violate market rules in not allowed to participate in the market. In order to various ways. www.lao.ca.gov Legislative Analyst’s Office 15 An LAO RepORt kEy TRAdE-OffS inHEREnT in dESigning A CAP-And-TRAdE PROgRAm The ARB Made Reasonable Choices . . . As plan involving: efforts to prevent leakage, the use of we noted earlier, AB 32 establishes a number of auction revenues, the use of offset credits, actions different and potentially competing requirements. to reduce volatility in the price of allowances and In addition to the main purpose of reducing GHG offset credits, auction and market oversight, and emissions, the plan must take into account impacts enforcement of cap-and-trade requirements. on local air quality, impacts on state revenues, There is no one right way to design a cap-and- cost impacts on regulated parties, and impacts on trade program. The Legislature could, however, the overall state economy. The specific design of a modify some features of the ARB’s cap-and-trade cap-and-trade program thus inherently involves program before compliance is required to reflect a making a number of key policy choices. For different set of policy choices among the competing virtually every feature of its plan, the ARB had to and conflicting goals inherent in AB 32. weigh the perceived policy benefits of a particular T -O I e rade ffs nvOlvIng ffOrTsTO approach in light of the potential trade-offs of r l educe eakage pursuing its chosen course of action. There is no way to know for sure exactly how Potential for Leakage the ARB’s program and its related carbon markets will ultimately work out because of its complexity California Policies Can Increase Economic and scale. Our analysis indicates that, for the Activity—and Emissions—Outside California. most part, the ARB has made a reasonable effort While any form of California climate policy could to balance these various policy trade-offs in the directly reduce California emissions, it could particular design of the cap-and-trade program also unintentionally increase emissions outside it has adopted in its regulations. Based on our of California. Such increases are referred to as economic and policy analysis of the ARB’s package, emissions leakage. For example, under cap-and- for example, we believe the cap-and-trade program trade, the new costs of reducing emissions and the would likely function fairly effectively in terms new costs of covering any remaining emissions of achieving the targeted level of GHG emissions with compliance instruments could put businesses reductions required under AB 32. in California at a competitive disadvantage relative . . . But Alternative Choices Are Possible. Our to businesses in places without analogous costs. If analysis further suggests, however, that the reduc- a California firm reduced its activities due to these tions in GHG emissions contemplated by the ARB costs, out-of-state competitors might increase their would probably not be achieved as efficiently, from activities to serve the California market, with the an economic perspective, as might be possible with possible result that their emissions would increase. a different design involving different policy choices. Alternatively, a California business might relocate A number of the features of the ARB’s plan involve outside of California due to competitive pressures, significant policy trade-offs that warrant policy again increasing out-of-state GHG emissions. review and discussion by the Legislature. In this There is another type of emissions leakage section, we discuss a number of policy choices and commonly referred to as reshuffling. A utility trade-offs made by the ARB in its cap-and-trade within California that imports electricity or 16 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt fuel might switch to importing a less emissions- $24 billion. The broad range of allowance value given intensive product, such as renewable energy, to away for free reflects ARB’s targeted range of prices reduce its cap-and-trade obligations. However, for allowances, discussed below. The level of leakage this would free up “dirtier” resources or electrical risk that will remain with ARB’s policies in place is generation capacity for purchase by utilities unknown and would be challenging to quantify. outside California. Thus, GHG emissions Trade-Offs. If ARB gave away more of the associated with those out-of-state markets might allowances for free than currently is planned, there increase. For example, staff at CPUC estimate would be three kinds of impacts, some positive that importers of electricity into California could and some negative. First, if the ARB gave more reduce their cap-and-trade obligations by up to allowances away to sectors at risk of leakage, the 15 to 27 MMTCO e in this way in 2013 without resulting risks of leakage, and costs to covered 2 reducing aggregate GHG emissions. Thus, this entities receiving free allowances, would be lower. type of leakage would lessen the efficacy of the This would further reduce the competitive disad- cap-and-trade program in directly reducing global vantages those sectors face. Second, because the emissions. ARB’s policy will essentially reduce the number of compliance instruments certain entities will need Portion of Allowances to Be given to purchase to cover the GHGs they emit, these Away for free to Reduce Leakage entities would emit more than if they had to pay Allowances to Be Given Away Will Reduce for the allowances. For example, it is possible that Costs of Covering Emissions. In its design of the ARB’s free allowances might allow an aging factory cap-and-trade program, the ARB chose to reduce to continue to operate, emitting both GHGs and leakage risks to a certain degree by giving allow- other types of pollutants that potentially degrade ances away to certain sectors. The ARB’s policy air quality. Third, if the ARB gave more allowances will essentially reduce the number of compliance away, then fewer allowances would be available for instruments these sectors will need to purchase other parties to buy at auction. This would lower to cover the GHGs they emit. This will therefore allowance auction revenues and could affect the reduce their compliance costs. With lower carbon market and other markets in the state. For compliance costs, these sectors will experience example, parties that would have participated in fewer competitive disadvantages relative to auctions might have a harder time finding other businesses in places without analogous regulations. parties to buy allowances from if auctions were (If covered entities receive more allowances for smaller. This would make the carbon market less free than they need, they will be able to sell excess efficient and allowance prices potentially higher. allowances to other parties.) If, on the other hand, ARB gave away fewer of We note that ARB will not be able to use the allowances for free than currently is planned, giveaways to eliminate all compliance costs because there would the same kinds of impacts just the supply of allowances—the cap—shrinks over mentioned, but in the opposite directions. First, time. the risks of leakage, and costs to covered entities, To achieve its desired reduction in compliance would be greater. Second, local air quality could be costs, ARB estimates that, by 2020, it will have improved. Third, auction revenues would be higher given away approximately 430 million allowances and allowance prices might be somewhat lower valued (in 2012 dollars) between about $4 billion and because of improved market efficiency. www.lao.ca.gov Legislative Analyst’s Office 17 An LAO RepORt T rade -O ffs f rOm d IrecTIng The decide on the appropriate use of these revenues by u se Of c erTaIn a ucTIOn r evenues the IOUs. While the CPUC has not completed this proceeding, it has indicated that it expects that the Cap-and-Trade Program Will majority of the proceeds will be used by the IOUs impact Electricity users in ways intended to benefit their ratepayers—such as by increased investments in energy efficiency Because the cap-and-trade program will in that would reduce energy consumption in the state effect incorporate a carbon price into goods and and thus indirectly reduce the cost burden of the services produced in the state, this will have the cap-and-trade program on California electricity effect of making many things in the economy ratepayers. more expensive, including electricity. Under a Trade-Offs. As with the approach to reducing cap-and-trade approach, allowing prices of more leakage, giving away allowances to electricity emissions-intensive goods and services to rise over distributors would reduce the revenues that time is intended to effectively motivate people and would otherwise potentially accrue to the state businesses to change their behavior in order to from auctions. While the planned uses of these reduce GHG emissions. revenues—for energy efficiency and renewable Provision of free Allowances to programs—may have merit, using the revenues for Electricity distributors these purposes comes at the cost of not making them available for other state purposes that may Reducing Impacts on Electricity Users. The better align with legislative priorities. ARB and the CPUC have jointly agreed that ARB will allocate free allowances to electricity T -O r rade ffs elaTedTOThe distributors (as opposed to generators), including u O c seOf ffseT redITs both investor-owned utilities (IOUs) and publicly owned utilities, through 2020. The purpose of Allowing use of Offset Credits to these free allocations is to reduce the cost burden Reduce Compliance Costs on electricity users from electricity price increases As discussed earlier, while the cap-and-trade expected to result from the implementation of the program focuses on reducing the emissions of the cap-and-trade program. capped economy, the ARB would allow certain In the first year of the cap-and-trade program, emissions reductions from offset projects anywhere the ARB plans to give electricity distributors in the United States to count toward compliance allowances equivalent to almost 100 MMTCO e. 2 with the cap-and-trade program. Accepting offset The amount will decline slightly after that. The credits for compliance is a way to allow parties in allowances given to the IOUs will then be sold on the “uncapped economy” to help meet emissions their behalf by the ARB in its quarterly allowance reduction goals. auctions. We estimate the approximate revenues Accepting offset credits for compliance is also from the auction of these allowances on behalf of a way to reduce the costs to the state economy of the utilities between now and 2020 will be from reducing GHG emissions. The theory behind using $8 billion to $41 billion (in 2012 dollars), depending offsets as a cost-containment mechanism is that, on the prices of allowances auctioned. In because entities in the uncapped economy may March 2011, the CPUC began a formal process to not have had relatively strong incentives to reduce 18 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt emissions, relatively low-cost options to reduce reductions from offset projects to be counted emissions may exist there. In the capped economy, towards meeting the AB 32 goal of reducing GHG in contrast, years of regulation and rising energy emissions. The standards are intended to ensure prices have already forced entities to undertake that the projects result in real and permanent many lower-cost options so additional abatement reductions in GHG emissions (see nearby box). could be more expensive. To the degree that the However, the strictness of the standards affects the cost of obtaining offset credits is less expensive cost of the offset projects and thus the quantity and than abatement within the capped economy, price of offset credits available. allowing the use of offset credits would reduce The ARB will rely on verifiers it will accredit compliance costs and lead offset credit producers to establish that offset projects meet the statutory to make emissions reductions that covered entities criteria. Verifiers must demonstrate competence, would otherwise have to make. assess and mitigate any conflicts of interest, and be subject to audits and strict performance evalu- Regulating the Quality of Offset Credits ations. The ARB is relying on these private parties Legislation Sets Standards Applicable to for this activity because its own staff does not Offset Credits. Assembly Bill 32 sets a number of currently have this expertise and because private standards that must be met in order for emissions verifiers are already carrying out similar functions. Criteria for Offset Credits and Projects Offset credits and projects must meet these emissions reductions criteria required by Chapter 488, Statutes of 2006 (AB 32, Núñez/Pavley): ➢ Real. Offset credits should result from demonstrable actions based on appropriate, accurate, and conservative methodologies. This should include accounting for leakage. For example, the accounting for emissions reductions from a carbon sequestration project by not harvesting timber somewhere in the United States should account for leakage, such as increased emissions due to timber harvests increasing elsewhere. ➢ Permanent. Emissions reductions from offset projects should not be reversible. ➢ Quantifiable. Emissions reductions from offset projects should be able to be accurately measured and calculated in a reliable and replicable manner. ➢ Verifiable. Emissions reductions from offset projects should be well-documented and lend themselves to an objective review by an accredited verifier. ➢ Enforceable. Some party should be able to be held liable by the Air Resources Board (ARB) in respect to an offset project, and ARB must be able to take appropriate action if the cap-and-trade regulation is violated. ➢ Additional to What Would Otherwise Occur. Offsets cannot be counted if the emissions reductions were already required or would otherwise have occurred on the natural. www.lao.ca.gov Legislative Analyst’s Office 19 An LAO RepORt Trade-Offs. There are policy trade-offs to use therefore could limit the unexpected emissions consider with regard to ARB’s specific approach to from failed offset projects. Second, a limit on offset allowing offset projects to meet the GHG emissions credit use could also benefit the environment and cap. The most fundamental trade-off is that their society in other ways. Restricting offset credit use intended benefit of lowering the cost of compliance would increase the emissions reductions required comes at the cost of some loss of certainty about of covered entities collectively. Because emissions of how much emissions will actually be reduced. This GHGs are in some cases associated with the release is because an offset credit’s quality often cannot of gases that harm public health, a limit on offset be proven definitively. The emissions of covered credit use could thus result in a greater improvement entities are relatively easier to verify as they are in air quality for communities living near covered based on reported emissions that have already entities than might otherwise be the case. occurred. However, verifying that an offset credit To the degree that offset credits are less is legitimate involves estimating how much GHG expensive than direct abatement of GHG emissions emissions are with the offset project and would by covered entities, allowing their use by covered have been without the offset project. If offset credit entities would reduce compliance costs. The quality were low, offset projects would result in ARB’s 8 percent limit thus constitutes a somewhat more emissions than expected. arbitrary limit on the use of offset credits in order to limit the emissions that could result from failed Limitations on the use of Offset Credits offset projects and to make it more likely that air 8 Percent Limit on Use of Offset Credits. quality near covered entities improves. The cap-and-trade program designed by the ARB T -O r a r rade ffs elaTedTO cTIOnsTO educe limits each covered entity’s use of offset credits c I p v OmplIance nsTrumenT rIce OlaTIlITy to at most 8 percent of its compliance obligations per compliance period. In other words, as an Excessive volatility in Prices alternative to complying by having an emissions Could Weaken Trading Program allowance for each ton of CO e it emits or by 2 reducing its emissions, the covered entity could In any market—but particularly in new, use offset credits to cover up to 8 percent of its untested ones—price stability can be important emissions. to its proper functioning. Several factors could Trade-Offs. To the extent that offset quality contribute to volatile prices in the cap-and- is high, the use of offset credits would help meet trade market. Prices could spike, for example, if emissions reduction goals at lower aggregate compliance instruments became scarce relative program compliance costs and ultimately decrease to the demand for them. For example, scarcity the program’s potential negative economic impact. could result from a surge in economic activity that On the other hand, because the use of offset credits increased emissions and therefore demand for presents some problems, there may be reasons compliance instruments. Likewise, prices could to limit their use. First, the emissions reductions “crash” if emissions-producing sectors suffered associated with offset credits can be uncertain and from an economic downturn or if compliance the offset projects themselves have the potential to instruments became plentiful relative to demand. fail. For example, a forest being used to sequester Excessive volatility in the prices of compliance carbon could burn down. Restricting offset credit instruments is a potential concern for the operation 20 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt of the cap-and-trade program. If compliance then grow at 5 percent per year in real terms. This instrument prices were particularly volatile, some minimum bid will generally function as a floor on carbon market participants would likely respond compliance instrument prices. by trading less than they would in a more stable While both of these mechanisms apply only to pricing environment. Reduced trading, in turn, the price of allowances, they are likely to also have would impede the ability of the cap-and-trade an indirect effect on the prices of offset credits. program to reduce overall compliance costs. Trade-Offs. As discussed above, the market price of compliance instruments under a cap-and- mechanisms to Reduce volatility trade program should ideally give each covered Allowance Reserves and Minimum Bid entity a signal regarding the degree to which Requirements. The ARB has designed its it should reduce emissions before it turns to cap-and-trade program to limit the volatility of compliance instruments to meet its obligations to compliance instrument prices. The ARB’s plan to the ARB. A covered entity, for example, generally keep allowance prices from spiking too high is to would abate more if its cost of doing so was below sell a limited number of allowances from a reserve the market price of allowances and credits. To the of allowances that it plans to establish. These degree that the ARB’s mechanisms provide greater allowances will be available to covered entities certainty regarding the appropriate level of direct in case there is an unexpectedly short supply abatement, these mechanisms will potentially that could otherwise drive allowances prices up reduce price volatility because covered entities and to high levels. The size of the reserve would be offset producers will have more information to limited so that, even if the entire reserve were determine the appropriate level of investment. As sold, the emissions reduction targets for cap-and- a result, covered entities will be able to plan better trade would still be met. (The reserve reflects the and the markets will generally function better in set-aside of 4 percent of total allowances.) The the long run. prices of these allowances are set at $40, $45, The ARB’s choice of the price ceiling and price and $50 per ton of CO e in 2013—with the least floor, however, may have costs. It is possible that 2 expensive allowances that are still available to keeping prices artificially higher or lower than they be sold first. These prices will grow at 5 percent would be otherwise will distort decisions about per year in addition to taking account further investments in abatement. Because of the targeted adjustments for inflation. These prices function as price floor, for example, a covered entity might ceilings on compliance instrument prices because, abate more than it would in the absence of the price so long as the ARB is selling allowances at these floor. This additional abatement would be unneces- prices, market prices are unlikely to go higher. sarily expensive and could be more than would If the reserve were ever exhausted, however, the be needed to meet the 2020 emissions level target. ARB’s cap-and-trade regulations would not limit Because of the targeted price ceilings, the developer how high compliance prices could go. of an abatement technology that would reduce The ARB’s plan to keep prices from falling too emissions at a cost above the price ceilings might be low is to require a minimum bid amount in all unable to find investors. This failure to invest could of its allowance auctions. The minimum bid will limit the options available to reduce emissions and be $10 per ton of CO e in 2012 and 2013 and will therefore increase compliance costs. 2 www.lao.ca.gov Legislative Analyst’s Office 21 An LAO RepORt T rade -O ffs I nvOlvIng academics with expertise in market development a ucTIOn and m arkeT O versIghT and oversight. The U.S. Commodity Futures Trading Commission and other regulators will also The Potential for gaming have important oversight roles. Recently, WCI Inc.—a nonprofit corporation—has been formed to provide Oversight of cap-and-trade auctions and administrative and technical services to support trading markets is important because of the the implementation of GHG trading programs. potential for “gaming”—manipulation through Officials from California, Quebec, and British collusion or fraud. Such activities tend to distort Columbia are on the initial board of directors. market price signals with potentially significant The WCI Inc. plans to conduct market monitoring consequences. If prices for allowances and offset of allowance auctions and market trading of credits were artificially high as a result of market compliance instruments. manipulation, for example, covered entities would Effective oversight of carbon markets will be spend more on abatement than needed. If prices important. However, it could also be a challenging were artificially low, some lower-cost abatement and potentially expensive effort. strategies that would be effective in reducing GHG emissions might not be implemented because Limits on Holding Allowances allowances and credits were less costly options. Basic Limit on Holding More Than 2.5 Percent Gaming could also lower confidence in the of the Market. The ARB cap-and-trade program carbon market, decrease its liquidity, and reduce is designed in a way that imposes several limita- the overall economic efficiency of the market. tions on participants in allowance auctions and For example, if gaming were common or serious in compliance instruments markets intended to enough that market participants did not trust prevent abuses. In particular, the ARB’s regulations each other, it might be difficult for a buyer to find limit the number of allowances that participants a potential seller of compliance instruments or for can hold for buying and selling. These are known parties in a potential trading transaction to agree as holding limits or position limits. (Allowances on a sales price, again potentially leading to covered held only to be surrendered for compliance will entities making unnecessary and expensive invest- not be limited. Also, offset credits will not be ments in abatement. subject to holding limits.) The basic holding limit market Oversight Program is set at 2.5 percent of the number of allowances Relies in Part on Private Third Parties scheduled to be auctioned off or given away for free in that year, with various complex adjustments. Oversight Provisions. The ARB’s regulations Trades that would violate holding limits will not be include several components intended to address allowed and could be reversed by ARB, which can the potential gaming of its cap-and-trade program. impose penalties on violators. The ARB is in the process of contracting with Trade-Offs. Holding limits are a low-cost an independent market monitoring service to way—in theory—to limit market power, including detect potential market manipulation as well as the power to manipulate carbon market prices. issuing a contract for the training of ARB staff For example, a market participant who wanted to market monitors. The ARB plans to assemble a drive prices up by holding some allowances out of “Market Surveillance Committee” composed of circulation would be unable to do so if the holding 22 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt limit were tight enough. That said, the potential T rade -O ffs r elaTed TO effectiveness of holding limits at reducing manipu- e nfOrcemenT p rOvIsIOns lation depends on having sufficient oversight mechanisms in place to detect and penalize such Lax Enforcement of Cap-and-Trade Rules conduct. Also, there are reasons to be wary of Could Weaken Program holding limits. If the limits were too restrictive, Under a cap-and-trade approach, it is possible participants could be unable to hold or use desired that at least some entities would fail to surrender amounts of compliance instruments for legitimate in a timely fashion sufficient compliance instru- business purposes, potentially weakening the ments to cover the emissions they reported for program in several important ways. Participants a period. This could be the result of intentional might need to find multiple buyers or sellers if they actions or could be inadvertent on the part of wanted to sell or buy many compliance instruments covered entities. In any event, ensuring compliance because any single party would be limited in what with these requirements is critical to the success they could hold. Because participants could not of constraining GHG emissions under a cap-and- buy—or hold and sell—as many compliance instru- trade program. Failure to address such issues would ments as they may want, their abilities to “correct,” undermine the goals of the program. through their trading transactions, for prices that they thought were too high or too low, including Penalties faced by Entities price changes due to price manipulation, would not Covering Their Reported Emissions be limited. Also, the establishment of holding Quadruple Penalties for Noncompliance. limits might prompt some participants to try to Under the ARB cap-and-trade regulations, a circumvent them, thus making market oversight covered entity would generally have to surrender more difficult. For example, an entity that wanted four times more compliance instruments than will to circumvent holding limits could create an entity otherwise be required if it failed to comply with that registers as a separate market participant program deadlines. For example, if a covered entity but that they actually control. If ARB did not had 100 tons of CO e emissions but only surren- 2 know the controlling entity was linked to the new dered compliance instruments covering 90 tons participant, ARB would not know to limit their of emissions on time, as a penalty it would have to joint holdings. surrender compliance instruments covering 40 tons By their nature, holding limits are somewhat of emissions. If emissions were not subsequently arbitrary and inflexible. Moreover, it is possible covered by compliance instruments as required, that the risk of carbon market manipulation may the ARB regulations indicate that other penalties be overstated. Other types of markets involving would be possible. the trading of commodities function well without Trade-Offs. As noted above, establishing holding limits. In summary, the cap-and-trade penalties for noncompliance is essential to the program designed by the ARB relies on holding success of the cap-and-trade program. Excessive limits to attempt to reduce various risks associated penalties for late compliance, however, could force with detrimental market behavior at the potential covered entities to hold an excessive number of cost of creating less efficient markets and higher compliance instruments as insurance against overall compliance costs. emissions spikes or compliance instrument price www.lao.ca.gov Legislative Analyst’s Office 23 An LAO RepORt spikes. Such increased holding could reduce market program, the ARB will be allowed to invalidate an liquidity because it reduces the availability of offset credit up to eight years after its issuance. For allowances for trading purposes and thus might example, offset credits would be subject to invali- increase compliance costs. In summary, ARB’s dation if the offset project violated a local, state, or penalty structure for late compliance potentially federal regulation, or was being counted “twice” comes at the cost of less efficient markets and as an offset credit for another program. The ARB increased compliance costs due to entities holding could in effect put a hold on offset credits while it compliance instruments as insurance against investigates potential problems with their validity. future compliance instrument price increases. While investigations occurred, those offset credits could not be used for trading. Final decisions Responses to Smaller Errors to invalidate offset credits would be subject to in Reported Emissions appeal to the ARB’s Executive Officer. Under No Corrective Actions Required for Smaller some circumstances, if a party was found to have Errors. Under ARB’s rules, underreported held or used the invalidated credit for compliance emissions of less than 5 percent of a firm’s total in purposes, penalties could be avoided if the invali- any year are allowed. In such cases, no corrective dated credit were replaced with a valid compliance actions would be required and no penalties would instrument within 90 days of notification. be imposed. Trade-Offs. The ARB’s policies of (1) prohib- Trade-Offs. By not penalizing small under- iting the trading of offset credits that are under reportings of emissions, ARB’s approach would investigation and (2) seizing invalidated offset save covered entities the costs of ensuring greater credits from whomever holds them place the accuracy in their reporting. In some cases, this potential costs of failed offset projects on users of lack of accuracy could result in lower costs for offset credits. In effect, they would bear the costs compliance with the cap-and-trade program. On of invalidation rather than the producers of the the other hand, the establishment of such a “safe projects or the ARB. This would provide users with harbor” for underreporting of emissions may a strong incentive to try to ensure that offset credits prompt some covered entities to deliberately do so. meet the criteria outlined earlier in this report. This policy would therefore allow a given level of Placing the potential costs of failed offset unaccounted-for emissions. In summary, the ARB’s projects on users of offset credits, however, raises approach of allowing some underreporting of concerns because offset producers are in a better emissions without penalties will keep compliance position to manage the risks of invalidation. For costs lower for covered entities at the cost of some example, an offset credit producer should know uncertainty about the amount of emissions reduc- if it sold two offset credits based on the same tions that would be achieved by the cap-and-trade one-ton reduction in emissions. Moreover, the risk program. of invalidation of offset credits could make offset credits worth less in general than allowances. For Responses to Offset Project failures example, an allowance might be worth $20 while Invalidation of Offset Credits Possible for an offset credit might trade at a discounted price of Up to Eight Years. As discussed above, the actual $15 because there would be a chance that it would reduction in emissions associated with each offset be seized by ARB if invalidated. To reduce the risks credit is uncertain. Under the cap-and-trade associated with offset credit invalidation, parties 24 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt entering into transactions involving offset credits main reason for allowing offset credits in the first might write more complex contracts, making place. carbon market transactions harder to understand In summary, ARB provides covered entities and oversee. If the risks associated with offset with strong incentives to try to ensure the validity credit invalidation were thought to be large enough, of offset projects and their associated offset credits market participants might avoid using significant at the cost of making markets more complex, amounts of offset credits. The risks of invalidation making the use of offset credits less attractive in the could thus limit the potential for offset credits to carbon market, and potentially increasing cap-and- reduce entities’ compliance costs, which was the trade compliance costs. CHAngES OR ALTERnATivES TO ARB’S CAP-And-TRAdE APPROACH In the preceding section, we discuss the relative cost-effectiveness of each of the measures policy choices and the associated policy trade-offs included in the Scoping Plan. Without such an involved in the ARB’s design of the cap-and-trade evaluation, the state cannot be assured that the mix program. Because striking the right balance in of measures, as well as the extent to which any one addressing these trade-offs is so important to measure is used, results in the most cost-effective the success of a cap-and-trade program and the approach to reducing the state’s GHG emissions. achievement of the overall goals set forth in AB 32, It is possible, for instance, that a larger role for we believe that the Legislature should carefully cap-and-trade relative to the direct regulatory examine at this time (1) whether the particular measures currently in the Scoping Plan could be design choices that the ARB has made for the a more cost-effective means of achieving the goals program are the best ones and (2) whether alterna- of AB 32. We recognize that many of the direct tives to establishing cap-and-trade should also be regulatory measures included in the Scoping considered. In the first of the next two sections, Plan were developed to address other policy goals we offer the Legislature options for changing the not directly associated with a reduction of GHG cap-and-trade program, such as to reduce overall emissions. Therefore, their repeal to accommodate compliance costs or to increase the certainty that a larger role for cap-and-trade in the Scoping Plan GHG emissions levels will be reduced to targeted could run counter to the Legislature’s policy prior- levels, that it may want to adopt depending on its ities. However, the Legislature may nevertheless priorities for AB 32 implementation. In the second want to balance these other policy goals against the of the following two sections, we discuss potential potential to reduce the economic impact of AB 32 alternatives to the cap-and-trade program as a by expanding cap-and-trade’s role. means to meet AB 32’s goals. These alternatives m c d akIng hangesTOThe esIgnand include both traditional direct regulatory measures O c - -T peraTIOnOf ap and rade as well as the other type of market mechanism— the carbon tax. In this section, we discuss potential alterna- A Larger Role for Cap-and-Trade in the tives to the cap-and-trade program’s current design Scoping Plan? The ARB has not evaluated the and potential uses for allowance auction revenues. www.lao.ca.gov Legislative Analyst’s Office 25 An LAO RepORt The program design options discussed all relate to ARB has in regard to the split between auctions the policy choices made by ARB and their inherent and giveaways of allowances. For example, the trade-offs discussed above. The options are Legislature could enact legislation to provide organized according to policy goals the Legislature additional revenues to the state from larger auctions may have, such as increasing certainty that the but still permit the ARB to give away allowances emissions target will be met. Most of these options sufficient to provide a meaningful response to the would have to be considered in light of their own problem of leakage. (We are advised by Legislative trade-offs. We conclude this section, however, Counsel that new legislation mandating that ARB with three options we recommend the Legislature auction all allowances would be subject to the adopt, as in our view these options have relatively two-thirds legislative voting requirement estab- little downside from a policy standpoint and would lished in Proposition 26, which we discuss further improve the program. below.) On the other hand, it may be a priority for the Legislature to reduce the economic burden on Changing the Overall Level of covered entities of the cap-and-trade program and, Revenues That Would Be Raised to this end, it could direct ARB to give a greater The cap-and-trade program designed by the proportion of allowances away for free. ARB will create allowances, with some being These issues have significant fiscal implications. auctioned for sale and others given away for free. If If, with a two-thirds vote, the Legislature chose, for more allowances were given away for free to covered example, to auction all allowances now planned entities, this would reduce the overall economic to be given away for free between now and 2020, burden the program is likely to have on the state’s the state would receive total revenues potentially economy. On the other hand, some observers, ranging from $27 billion to $140 billion (in including economists who provided advice to the 2012 dollars). The amounts received would probably ARB about the establishment of the cap-and-trade range from $2 billion to $11 billion annually in program, have recommended that greater use be the early years, and from $3 billion to $22 billion made of auctions in the program and that fewer annually in later years. allowances be given away for free. They argued that use of Revenues derived from Allowances greater use of auctions would make the program less reliant on inherently subjective decisions about Background on Fees and Taxes. Proposition 26 which entities should receive allowances. Also, in (November 2010) expanded the definition of what choosing to give a certain portion of allowances constitutes a tax and a tax increase so that more away for free, the ARB has in effect chosen to forego proposals would require approval by two-thirds tens of billions of dollars in state revenues that could of the Legislature (or, in some cases, by local otherwise be allocated by the Legislature over the life voters). For example, some regulatory charges of the program to address its priorities. (As will be that benefit the public broadly would if passed discussed below, however, there are legal constraints now be considered taxes instead of fees. While regarding the eligible use of auction revenues.) Proposition 26 was applied under its terms retro- Auctioning More or Fewer Allowances. While actively to January 1, 2010, it does not apply to any it must be careful of the legal implications of such revenue measures adopted prior to that date. Also, an action, the Legislature may wish to consider Proposition 26 did not change an existing provision the option of making a different choice than of the State Constitution, known as Proposition 98, 26 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt that generally requires that a minimum share of Legislature wished to consider the option of using General Fund tax revenues be provided to public such auction revenues for purposes unrelated schools and community colleges. to GHG emissions mitigation, it would need to Auction Revenues Would Constitute enact a new statute for this purpose that would Mitigation Fees. As discussed earlier, ARB intends supersede AB 32 as the authority for the collection to sell some allowances at auctions under the of these revenues. However, in taking such action, authority granted it in AB 32 to establish a market- the Legislature would now need to consider the based mechanism to reduce GHG emissions. Based potential application of both Proposition 26 and on an opinion that we received from Legislative Proposition 98. In some cases, what formerly were Counsel, such state auction revenues constitute considered mitigation fees if passed now would be “mitigation fee” revenues. Because AB 32 was deemed taxes under Proposition 26. Moreover, the enacted (by a majority vote of the Legislature) prior auction revenues would constitute proceeds of taxes to the voter approval of Proposition 26—and well that could affect the state’s Proposition 98 obliga- before its specified retroactive date of January 1, tions. For these reasons, we recommend that the 2010—we are advised that the provisions of Legislature seek the advice of Legislative Counsel Proposition 26 would not apply. As such, no in the future regarding the ARB’s proposed use of additional action by the Legislature is required for these funds as well as any such proposals it may collection of auction revenues under ARB’s plan. have of its own. Also, because the proceeds from the auctions are In our view, it is ultimately the Legislature’s fee revenues and not the proceeds of taxes, we are prerogative—in exercising the authority given to advised that the state’s receipt of these monies it in the State Constitution to appropriate state would not affect the state’s Proposition 98 funding monies—to decide the best use of the state revenues obligation for schools and community colleges. derived from the auctioning of allowances. As it This set of circumstances has other important contemplates its choices, we recommend that the implications. Because auction proceeds are deemed Legislature carefully consider a number of policy to be mitigation fee revenues, we are further concerns, such as how the particular use of these advised that they must be used only to mitigate revenues could impact—positively or negatively— the harms caused by GHG emissions. Therefore, the effectiveness and efficiency of the cap-and-trade the Legislature would be constrained in the types program in meeting AB 32’s goals. For example, of state programs for which it appropriated these if part of the revenues from the auction were used mitigation fee revenues. Appropriate uses of the to reduce electricity rates, this could run counter revenues for mitigation purposes could potentially to efforts to motivate energy users to change include expenditures on energy and water use behavior to reduce GHG emissions. (The ARB itself efficiency programs, alternative fuels programs, has not proposed such an approach.) New state and investments in renewable energy projects. revenues derived from the auction of allowances Use of Auction Revenues for Other Than could potentially be used to broadly ameliorate Mitigation. The constraints discussed above on or reduce the potential negative impacts of the the use of auction revenues apply, we are advised, cap-and-trade program on the California economy so long as the auction revenues are collected under and consumers. While the following options would the authority already granted to ARB to establish likely require a two-thirds vote, the Legislature a market-based mechanism under AB 32. If the could use auction revenues to: www.lao.ca.gov Legislative Analyst’s Office 27 An LAO RepORt • Pay Dividends to All Californians. from cap-and-trade may be highly variable from As noted earlier, ARB has suggested year to year. This means that they may be more that auction revenues be distributed appropriately used for one-time or short-term to Californians to offset the impacts purposes rather than for the support of ongoing of cap-and-trade regulations on fuel programs or tax reductions. costs. Our analysis indicates that such Changes to increase Certainty revenues could be returned directly to That the Emissions Target Will Be met Californians—such as in the form of a check—as a dividend that would be The ARB will rely on the declining overall intended to offset their increased expendi- cap on emissions, its standards for the quality of tures on goods and services that ultimately offset credits, and its policies to prevent leakage would become more expensive as a result to provide some level of assurance that the of the cap-and-trade program. Such a emissions reduction target for cap-and-trade will dividend program could be designed in be met. However, some aspects of its approach a way that preserves the incentive for to enforcement still leave some uncertainty as to recipients to change their use of goods and whether the emission reduction target will be met services that significantly contribute to on time. There are steps the Legislature could take GHG emissions. to increase confidence that emissions will actually be reduced as planned. • Address the State’s General Fund Increasing Penalties—or Lower Thresholds— Deficit. Revenues could be used as part of for Underreported Emissions. The Legislature a multiyear approach to reduce the state’s may wish to consider the option of directing ARB projected General Fund deficit. The avail- to better ensure that GHG emissions reductions ability of these revenues could allow the actually occur by changing policies regarding state to avoid other actions, such as cutting the underreporting of emissions. For example, governmental programs or increasing the Legislature could ensure that underreported state revenues, that could slow the state’s emissions are covered with compliance instruments. economy. The threshold of errors in reporting that would As it examines these options, as well as any result in corrective action or penalties could also be other recommendations that come forward set lower than the 5 percent level set by the ARB. from ARB, the Legislature should consider how Changes to Reduce Overall Compliance Costs they would interact with the Proposition 26 and Proposition 98 requirements discussed above. For Some features of ARB’s cap-and-trade example, if the Legislature chose to enact a new program—such as the limits on the use of offset statutory measure that allowed the use of auction credits—would likely have the unintended effect revenues to address the state’s General Fund of increasing the costs of compliance with the deficit, it would probably want to take into account program. However, our analysis indicates that the potential for the receipt of these monies to there are changes the Legislature could make to the affect the Proposition 98 funding guarantee for program to reduce future compliance costs. schools and community colleges. In addition, the Removing Limits on the Use of Offset Credits. Legislature should keep in mind that the revenues The ARB limits each covered entity’s use of 28 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt offset credits to covering at most 8 percent of its cap-and-trade program have not been justified compliance obligations per compliance period. analytically, were set before actual trading in As we have discussed, this limit is a somewhat allowances could be observed, and do not adjust arbitrary one, and potentially leads to higher automatically with changing market conditions. compliance costs. The Legislature could consider Therefore, ARB’s holding limits are unlikely to be the option of taking a different approach to optimal. As discussed above, there are significant regulating the use of offsets—one directing the consequences if the holding limits are set too ARB to eliminate the 8 percent limit and instead tightly or too loosely. If set too tightly, they might using stringent verification standards to limit the make trading unnecessarily costly and reduce flexi- use of offset credits. Provided verification standards bility regarding when emissions reductions take were sufficiently high, there would be only a place. If set too loosely, they would be ineffective at relatively low risk of emissions exceeding targeted deterring the manipulation that they are intended levels due to failed offset projects. to prevent. Attempting to change the holding limits Addressing Problematic Offset Projects by over time to be more optimal presents its own set Means Other Than Credit Invalidation. Under of problems. A potentially time-consuming public the cap-and-trade rules, ARB currently can seize process would be needed to modify the ARB’s offset credits if an audit revealed a problem with regulations on holding limits. By the time such the associated offset project. However, the risk regulatory changes were made, market conditions that offset credits could be invalidated and seized might have changed again. after they have been verified creates uncertainty Since it is possible that ARB’s holding limits about their value in the carbon market and may may be ineffective in deterring market manipu- deter offset credit use. If the Legislature wanted lation or may serve to unnecessarily increase to avoid these consequences, it could consider the compliance costs, the Legislature could consider option of directing ARB to use other means to the option of eliminating holding limits. In their address failed offset projects. For example, offset place, the Legislature could rely on better tools to producers could be required to carry insurance deter manipulation, such as increased penalties on against the potential costs of offset project failures. those found guilty of market manipulation. (While such an insurance market does not exist Setting a Lower Floor for Allowance Prices. As currently, it potentially could arise.) If a project we noted above, the ARB plan for cap-and-trade failed, the insurance proceeds could be used to buy would rely on minimum bidding requirements compliance instruments from the carbon market in auctions to set what amounts to a floor on to make up for the project failure. If that approach allowance prices. The minimum bid will be $10 per turns out not to be practical, another option would ton of CO e in 2012 but will grow at 5 percent per 2 be for ARB to establish a reserve of offset credits year in real terms. This would help to stabilize the from which it could draw to make up for failed prices of allowances but could increase cap-and- projects. That reserve could be established and trade compliance costs. maintained through contributions required from The Legislature may wish to consider the offset producers as a condition of selling offset option of directing the ARB to set a lower price credits. floor than $10, or have it grow more slowly than the Eliminating Holding Limits. The particular current 5 percent per year, if market prices below holding limits chosen by ARB in its design of the the targeted price floor were a realistic possibility. www.lao.ca.gov Legislative Analyst’s Office 29 An LAO RepORt We note that recent national proposals to create look at alternatives for achieving the state’s goals cap-and-trade programs have floors that grow far under AB 32. There are two main alternatives for slower than ARB’s annual growth rate of 5 percent achieving the GHG emissions reductions assumed plus inflation. under the ARB’s cap-and-trade program. The Begin Reducing Uncertainty About the Years first set of alternatives involves making changes After 2020. Assembly Bill 32 states the intent of or additions to direct command-and-control the Legislature that the statewide GHG emissions regulations that apply to GHG emitters. The second limit continue beyond 2020. However, the exact involves the imposition of some form of carbon structure of the cap-and-trade program and the tax. These two sets of alternatives are not mutually compliance obligations faced by regulated parties exclusive, and could be combined in various ways after 2020 is not specified in the legislation or to replace the emissions reductions expected from current regulations. The Legislature may wish to cap-and-trade. consider the option of laying out these policies well Command-and-Control before 2020, including such technical issues as what Regulation Alternatives value banked compliance instruments from the current cap-and-trade program would have. This Developing Direct Command-and-Control would make it easier for participants today to make Regulations for All Entities Covered by Cap-and- smarter long-term investments and comply with Trade. As part of the process of developing its cap-and-trade in less expensive ways. cap-and-trade regulations, the ARB was required to identify potential alternatives to the cap-and- Some Options Would improve Program trade program. In its regulatory documentation, And Have Little downside the ARB stated that one such alternative would be Our analysis indicates that some of the options to implement command-and-control regulations we discuss above for modifying ARB’s design of targeted at specific GHG emission sources that the cap-and-trade program would improve the would likely achieve comparable levels of GHG program and have relatively little downside from a emissions reductions. policy standpoint and would be consistent with the To meet AB 32 goals without a cap-and-trade overall goals set forth in AB 32. If the Legislature program, the Legislature would have to consider determines that it wishes to proceed with a expanding the types of direct regulations that cap-and-trade program, we would recommend already exist in the Scoping Plan. For example, the that the Legislature seriously consider the ARB could be directed to evaluate the potential for following modifications to ARB’s program design: expanding an existing program that audits firms (1) making producers of offset credits liable for for their energy efficiency and mandates upgrades offset project failures, (2) eliminating holding limits that would help them reduce GHG emissions. New to improve the way the carbon market functions, regulations focused on the industrial sector— and (3) reducing uncertainty about how and if the including power plants, refineries, and cement cap-and-trade program would operate after 2020. plants—which collectively constitutes 23 percent of the state’s total estimated GHG emissions would p a c - -T OTenTIal lTernaTIvesTO ap and rade also be possible. This is because under the Scoping If the Legislature decided not to proceed with Plan the vast majority of emissions reductions from the cap-and-trade program, it would need to the industrial sector are planned to come from the 30 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt cap-and-trade program rather than command-and- targeted levels at a lower cost than traditional direct control regulations. regulatory approaches. This is because a carbon Command-and-Control Regulations Could tax provides an economic incentive to all regulated Be Designed to Provide More Certain Results. A emissions sources to find the mixes of emissions major trade-off with traditional command-and- reductions and tax payments that minimize their control regulations is that one loses the certainty costs. about emissions levels that the cap component Carbon taxes are or have been in use in several of the cap-and-trade program would provide. places, including Finland; Sweden; Great Britain; However, it is possible to design command-and- Boulder, Colorado; and Quebec and British control regulations in ways that provide relative Columbia in Canada. The program in British certainty about emissions levels. This would entail Columbia, for example, taxes different fossil fuels major changes to the traditional design of these at different rates depending upon the intensity of regulations. For example, command-and-control their carbon emissions. Carbon tax revenues have regulations could be designed to require specific been used to reduce the rates of other taxes and to amounts of reductions in GHG emissions from allow the creation of a new income tax credit for regulated entities equivalent to the reductions low-income persons. they would have faced under the cap-and-trade Basic Design Choices for a Carbon Tax. program. Setting such performance standards In theory, a carbon tax could be established in would require ARB to obtain much more California for selected parties based on the GHG information about regulated entities’ characteristics emissions associated with their activities. The than it currently has, such as through the energy design of a carbon tax involves similar choices efficiency audits discussed above. and trade-offs as those involved in designing a In order to increase the likelihood that the cap-and-trade program. If the Legislature wished emissions target would be met, and to help contain to consider this option, the key choices in its design compliance costs, the Legislature may also wish to would include: consider the option of incorporating some market- • What Would Be the Basis of the Tax and based features of the cap-and-trade program into From Whom Would It Be Collected? command-and-control regulations. For example, A California carbon tax could only be the ARB could allow the use of offset credits to imposed on GHG emissions from specified meet a firm’s obligation to comply with a particular sources and activities within the state’s emission reduction requirement. In effect, such legal jurisdiction. The activities targeted an approach could provide at least a limited for cap-and-trade regulations would be backstop in the event that the new direct regulatory a logical starting point for the selection approach were unsuccessful. of a tax base. The emissions reports now provided to the ARB could also be used for Carbon Tax Alternatives tax assessment purposes. Approach Used in Other States and Countries. As referenced earlier in this report, a carbon tax • What Would the Tax Rates Be? Setting could be imposed in California as an alternative the optimal level for such a tax to meet to a cap-and-trade program. Like a cap-and-trade the state’s emissions reduction goals program, a carbon tax could reduce emissions to would be a challenge. Tax rates could start www.lao.ca.gov Legislative Analyst’s Office 31 An LAO RepORt low, however, and increase over time to • How Would Tax Revenues Be Used? gradually shift the economy toward lower How tax revenues were used would be a GHG emissions. legislative prerogative. The options could be the same as we discussed above for auction- • Any Adjustments? The tax base, tax related revenues under a cap-and-trade rates, and credits against the tax could be program—providing dividends directly to adjusted to achieve important policy goals Californians or addressing the state’s fiscal related to the regulation of GHG emissions. problems. Tax rates could be lower, for example, in sectors that would otherwise be at a high risk of competitive disadvantage as a result of the imposition of a carbon tax. COnCLuSiOn The design of the cap-and-trade program as Finally, as we have noted, AB 32 has many adopted by ARB involved a number of key policy goals—often competing with one another—that choices that have been explored in this report. guide the design of the state’s climate change plan These policy choices affect such fundamental and emissions reduction measures, including outcomes as the extent, and degree of certainty, the cap-and-trade program. These goals include of GHG emissions reductions, impacts on local requirements to minimize leakage, maximize air quality, cost impacts on regulated parties, and cost-effective emissions reductions, ensure that impacts on the overall state economy. For each compliance activities do not disproportionately policy choice, the ARB weighed the perceived impact low-income communities, and minimize benefits with the potential trade-offs—many of the administrative burden of implementing which are very significant—of pursuing its chosen and complying with AB 32 regulations. The course of action. requirement to meet and balance all of these goals Our analysis indicates that ARB has made a has made the design of the cap-and-trade program reasonable effort to balance these various policy necessarily complex, and has involved the making trade-offs in the particular design of the cap-and- of trade-offs by ARB that may not be in line with trade program it has adopted in its regulations. As legislative priorities. To help reduce the level of we have demonstrated, there is no one right way complexity of the cap-and-trade program, the to design such a complex program. Accordingly, Legislature might consider enacting legislation we recommend that the Legislature carefully that sets its priorities among the many goals and consider both potential changes to the design of the stated criteria for emissions reduction measures cap-and-trade program as well as such alternatives that are found in AB 32. Such action could provide to cap-and-trade as expanded direct regulatory useful direction to ARB on how these goals are efforts or some form of a carbon tax. appropriately balanced in a way that is consistent with legislative priorities. 32 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt www.lao.ca.gov Legislative Analyst’s Office 33 An LAO RepORt 34 Legislative Analyst’s Office www.lao.ca.gov An LAO RepORt www.lao.ca.gov Legislative Analyst’s Office 35 An LAO RepORt LAO Publications This report was coauthored by James nachbaur and Tiffany roberts, and reviewed by Mark C. newton. The legislative Analyst’s office (lAo) is a nonpartisan office that provides fiscal and policy information and advice to the legislature. To request publications call (916) 445-4656. This report and others, as well as an e-mail subscription service, are available on the lAo’s website at www.lao.ca.gov. The lAo is located at 925 l street, suite 1000, sacramento, CA 95814. 36 Legislative Analyst’s Office www.lao.ca.gov