Price Containment in California’s Carbon Market: a Difficult ARBitrage

On June 25, 2013, while the world was watching President Obama give a historical allocution on climate change, California emitters and regulators were absorbed in a nerdy and fascinating discussion on whether and how California’s cap-and-trade price ceiling mechanism should be amended. The Air Resources Board (ARB) brought some of the smartest minds in the country to discuss the proposal, which made for a very stimulating discussion on this altogether important issue of what is an acceptable cost to meeting the state’s reduction target, and what is the best way to keep costs under this acceptable level. This post largely draws on the arguments experts and industry participants laid out – for more details you might want to refer to their presentations as well.

Note that the same workshop also discussed two important topics that will likely see changes in the round of upcoming amendments in September: compliance timeline and data transparency. I’ll write about the implications of the compliance timeline when ARB stops moving dates around.

Why change the APCR

The Allowance Price Containment Reserve (APCR) is a mechanism by which a certain percentage of allowances are set aside from the main allowance budget in the California cap-and-trade program, and offered for sale on a quarterly basis at a prix fixe, starting at $40/$45/$50 in 2012 and rising to $65/74/82 in my estimates by 2020. The APCR is divided in three tiers, hence the three prices, and APCR allowances may only be purchased by compliance entities and cannot be resold – they go straight to an entity’s Compliance Account, which has a one-way revolving door for all compliance instruments.

The first concern with the APCR is that the pool of allowances (122 million total) could be exhausted  – modeling from the Energy Institute at Haas shows non-negligible odds of seeing all the APCR allowances sold out, after which prices could shoot up unchecked, possibly to three-digit numbers. Nobody likes the thought of that.

UC Berkeley's model shows a 3-22% chance that prices would get higher than the APCR.
UC Berkeley’s model shows a 3-22% chance that prices would get higher than the APCR.

The second concern is that, because nobody likes this idea, there is a widely shared expectation that policymakers would step in and try to prevent the APCR from getting depleted, either at the last minute or during a sudden high-price event. That’s also not a good idea – it’s very disruptive, potentially costly, and it undermines the credibility of the program. The intervention could come from the Legislature, from ARB itself, or even from the Governor, as provided in AB 32 section 38599. (a): “In the event of extraordinary circumstances, catastrophic events, or threat of significant economic harm, the Governor may adjust the applicable deadlines for individual regulations, or for the state in the aggregate, to the earliest feasible date after that deadline.” What is the point of writing 275 pages of regulations if regulated entities believe Superman Brown or his successor will step in and change the rules at the last minute?

The Emissions Market Assessment Committee, (EMAC), composed of economists from UC Berkeley, Davis, and Stanford, highlighted this issue in a presentation back in September 2012. Heeding the advice from EMAC, the Board mandated staff to bring to the Board, “by mid-2013, a proposal for incorporation into the cap-and-trade program one or more mechanisms that will achieve the policy objective of ensuring that allowance prices will not exceed the highest price tier of the Allowance Price Containment Reserve while minimizing the impact on existing allowances and maintaining the environmental objectives of the program” (ARB Board Resolution 12-51). Last week’s workshop was an opportunity for the ARB staff to discuss some of the ideas they’re looking at and gather feedback from experts and stakeholders.

What is the purpose of the APCR?

Cost containment is generally a very consensual idea – but it can mean different things to different people. In this case, the panel’s economists – Dallas Burtraw from RFF, Jim Bushnell from UC Davis, and Brian Murray from Duke University all agreed that the APCR was better suited to prevent short term price spikes and market manipulation than long term upward pressure driven by a large, sustained imbalance between supply and demand. Industry and market observers at large instead also seem to expect the APRC to be the backstop that guarantees that prices will stay in check even if supply and demand clearly cannot balance overtime under the top tier price of the APCR.

If the goal is to prevent upside price risk and volatility, the APCR is adapted and arguably only needs to be bolstered a bit, with a deeper reserve of compliance instruments, to act as a credible deterrent. But if the goal is to ensure that prices will never, ever go above the top tier price, and yet ensure that California will meet its target in any case, then ARB is really trying to square the circle – possibly in vain since a long term imbalance would in any case warrant a review of the program. In addition, panelists argued we’ll get plenty of notice before prices creep up to the reserve – if the first tier or two of the APCR get depleted over time, this will act as a warning sign and trigger scrutiny from a lot of players in any case. In my estimate, the earliest we might need APCR allowances to balance the market is the third compliance period, 2018-2020. That’s not to say a nasty combination of nuclear outages, wildfire-driven transmission breakdown and heat wave couldn’t push prices high this summer – but this kind of temporary price spike is precisely what the APCR should contain effectively.

What is a good cost-containment system?

Jim Bushnell outlined three important qualities for a good cost-containment system:

  • It should be ex-ante transparent, since a deterrent only works if market knows it exists
  • It should be credible, both in terms of price level and mechanism
  • It should be timely – it needs to kick in immediately when prices rise, not be dependent upon regulatory or political action

Bushnell and Murray both questioned the term “environmental objectives” of the program cited in Res. 12-51. Since we’re talking about greenhouse gas (GHG) emissions, a global problem in essence, one might want to consider emission reductions from a global standpoint as well, or at least not exclusively focused on California. Murray insisted on two dimensions to consider: the timing and the spatial location of GHG emission reductions. California’s program already provides flexibility on both accounts – reductions may take place at any facility in state, or in Quebec, or in other states for some offset projects, and the banking and APCR already provide some flexibility on time. When it comes to GHG emissions, argued Murray, timing matters a little bit (at least within such a short time period), and space doesn’t. Hence California would still be meeting its environmental objectives if it relied on emission reductions taking place in other geographic location – say, the EU or Australia for example…

ARB’s proposal

ARB staff laid out a wide range of options that were received with a mixed response:

  1. Do nothing. Received a thumb down from the panelists – if there is up to 22% chance that the APCR will run out, it’s not effective and needs to be strengthened. Moving on.
  2. Cancel (temporarily) the compliance obligation – also a losing proposition. It adds disruption, and potentially creates equity issues amongst compliance entities, a lot of whom have emissions that vary a lot from one season to another (e.g. power generators and food processors)
  3. Delay compliance obligation – called “staggering” by Dallas Burtraw, this would effectively create an overlap between compliance period, allowing emitters to use allowances from the new compliance period to meet their obligation from the previous compliance period. Yikes! The compliance timeline is complicated enough as it is – shall we really make things more complicated by changing the rules at the last minute? Tell me about disruption… That option was also shunned by the other panelists.
  4. Allowing compliance via the payment of a fee rather than by turning in compliance instruments. To me that’s just a rhetorical difference. If the fee revenue is used to purchase credits 1:1 from outside the program, you might as well require emitters to surrender said credits – skip to proposal 5. If the fee revenue is used for other purposes, such as long term R&D in energy technology, it’s akin to a “safety valve” as proposed in some federal cap-and-trade bills back in 2008-2009, where the supply of additional allowances is unlimited in order to guarantee prices – and therefore would likely fail to meet the environmental objectives of the program. Go straight to jail and don’t pass go.
  5. Replenish the APCR with additional credits, to provide a larger safety cushion and signal to the market that regulators are serious about keeping prices low and discouraging attempts at market manipulation by making it much more costly to try to create artificial scarcity on the market. By far the most discussed option, and in my view the best option by far, but of course the question is – how?

How to replenish the APCR?

Once again, ARB staff laid out a wide array of options – I’ll shoot down a few right away to focus on the ones that matter most from my standpoint.

  • Require additional reductions pre-2020 from sources not covered by the program. Uh, good luck with that. There’s a reason they’re not covered after all – emissions are difficult to measure and potentially costly to curb in the agricultural and forestry sector. Waste is already regulated. And all of them together make up a mere 13.5% of covered emissions – a tough ratio if we expect any meaningful number to come from such a small cumulative sector.
  • Redistribute allowances in the 2013-2020 period. I’m not sure I fully get it. Set more allowances aside to put it the APCR? That would not change the overall supply and demand balance, only make it more likely that we’re at the APCR price. There’s very little wiggle room in any case.
  • Require additional post-2020 emission reductions. We don’t know what emission reductions will be required post-2020 – how could we commit to more? It’s going to be a while before ARB sets a 2030 and beyond allowance budget (that’s part of the Scoping Plan Update, which then needs to be turned into actual regulations), so I’m not sure that would meet the “credible” criterion, since it would have to be drafted in very vague terms in the upcoming round of regulations.
  • Borrow from post-2020. I don’t really like that option either – for the reasons as just mentioned, and also because it seems to me that it’s making a silent assumption that we’re going to magically find a technology that will reduce emissions so much that we’ll meet our post-2020 target easily, so we can even borrow some of those future reductions to meet today’s target. I’m always a little weary of this faith in technology – it might be so, and then the world will be a better place and I can’t wait to see it. But it might not be so, and then we’ll be stuck with an even harder to reach target, and we won’t have made the right investments at the right time. It is worth mentioning, though, that panelists thought it most important and useful that ARB publish the allowance budget for post-2020 well in advance, as a long-term cost containment measure – it might push prices up in the short term, but it would ensure the necessary investments are made early on to meet the later cap.
  • Procure additional credits or allowances – my favorite option, but of course, the devil is in the details.

Shall we buy Made in California credits only?

This is where the crucial question of spatial flexibility arises. Should compliance instruments come from California only? From the rest of the U.S.? From the rest of the world? Could California buy allowances from other compliance programs as a way to replenish the APCR?

When it comes to California-based offsets, as discussed earlier, the potential is extremely limited since most sources of emissions are already covered in some form or fashion by the cap-and-trade program, or a regulation, or economic incentives. There is a large pool of offset credits in the U.S. outside of California, which would require ARB to consider offset protocols that are not currently eligible for compliance, such as industrial processes or land-use projects, which could provide large volumes and incentivize reductions in other states. This option has the benefit of reluing on well-known, trusted organizations – the current registries. But if those offsets are good enough for the APCR, should they allowed for compliance in general?

Outside of the U.S., California could consider selected credits from the Clean Development Mechanisms, following the new qualitative restrictions imposed by the EU ETS, whereby projects have to be located in Least Developed Country only. No more industrial gas from South Korea or “efficient” coal plants from China. California could also build on its existing political partnerships with Shenzhen or Mexico. Other options include EU Allowances themselves, or Australian allowances, or New Market Mechanisms under the UNFCCC. The options are many, and the hurdles not as insurmountable as one might think.

And then of course there is the Regional Greenhouse Gas Initiative (RGGI), long-time political partner of California, but long plagued by its overallocation. RGGI would be a natural place to look for out-of-state additional allowances, but only once the program’s fixes are implemented. And RGGI’s very low price ceiling might be an issue as well.

The panelists made two very important points on this issue. First, “every ETS has their warts,” said Bushnell. No point in pretending California has a perfect system and other markets are not worthy of its attention (really, who would think that?!…). The allowances are solid emission reductions in all of these programs, it’s only a matter of supply and demand – low prices don’t stand for low quality. Second, Burtraw suggested allowances from other ETS that are significantly cheaper could be converted through some kind of exchange rate or trading ratios – building on a recent paper he published at RFF. For example, California could buy five RGGI allowances at $10 each, sell one APCR allowance for $50 to a California emitter and retire the four others. In addition, Bushnell also insisted that the State of California buying allowances from other programs under strict conditions was nothing like a full linkage where private entities could buy allowances directly from their counterparts in other trading systems – an important precision since California is in no rush to establish such broad linkage with other markets.

Policy expectations and recommendations

The APCR debates touches to the very heart of California’s climate policy. The underlying question is: how much will it cost to meet California’s target – not so much the 2020 target, but the 2030 and beyond target, and is California willing to bear that cost? Does it make sense for California to try to go it (and pay for it) alone or should California rely on more emission reduction imports, just like it takes responsibility for emission imports?

I think it’s important to circle back to the distinction between short term and long term price containment. The APCR is made to prevent extreme volatility, short term spikes, it won’t do as a long term price ceiling. To meet its goal of preventing sudden, temporary price increases, I would largely prefer to see the APCR being replenished with allowances from other reputable trading systems, like RGGI and EU ETS, using a trading ratio that would ensure eager Californians get their money’s worth. As a nice collateral benefit, it would send a positive political signal at a time where the rest of the country is just starting to get its act together on emission reductions.

Yet I expect ARB might shy away from this solution, and will prefer the vague but reassuring option of borrowing from post-2020 – which is more a way to focus regulated entities on the big picture and pressure them into early reductions than real new supply for the APCR.

As much as I support the idea of expanding California’s cap-and-trade to additional offset protocols from the U.S. and abroad, including REDD and some carefully selected CDM project types and locations, I don’t think this is the right time or the right way to introduce such offsets into the program. ARB would be all too easily accused of letting through the back door credits that have been excluded from the current regulations. In a context where the California Senate just recently passed SB 605, a bill that requires ARB to “limit the use of offsets, to the maximum extent feasible, to those offsets originating and achieved within the state,” it probably wouldn’t go over very well. (The bill is now being amended in the Assembly.)

When it comes to long-term cost containment, I think it will be absolutely crucial for California to re-open the debate on international offsets and linkages down the road, but it should do so through the front door, using the same process it currently uses for examining new offset protocols, and not via the APCR. It may come to a point where California will have to recognize that it may not be able to solve this problem alone, and that relying more on its may will support its long term strategy and the global, joint goal of reducing GHG emissions.