Skip to content

The Whac-a-Mole Economics of Border Carbon Adjustments

What can we learn from California’s pioneering efforts to price carbon at the border?

Everybody’s talking about border carbon adjustments. The EU has announced its plan to tax the emissions in its carbon-intensive imports. Senate Democrats have proposed to tax the carbon in U.S. imports from countries with less stringent GHG regulations. Canada has plans to incorporate border carbon adjustments into its carbon pricing regime.

These proposals are generating lots of excitement. And apprehension because implementation promises to be wickedly complicated and legally fraught.

The good news is that we’re not flying blind. California has been border-pricing the GHG emissions in its electricity imports since 2013. So today’s blog asks an important question: What can we learn from California’s pioneering border carbon adjustment?

Border carbon whac-a-mole

If you’re a state/country/region that’s aspiring to regulate your GHG emissions- when your trading partners are not– the border carbon tax idea looks pretty appealing. 

Taxing the GHGs contained in your imports can help level the carbon playing field for your regulated producers (rebating the GHGs embodied in exports is another part of this story– but not the focus of this blog). It can limit “leakage” of GHG emissions into areas with weaker regulations. And it could help nudge other jurisdictions to join the climate fight- which has to be the end-game since GHGs are global pollutants.

Sounds great in theory! But I want to focus on a real-world implementation complication: how to count the GHG emissions in imported products?  California’s experience helps illustrate why this can feel like an infuriating game of whac-a-mole.  

When California was designing its GHG cap-and-trade program, it was clear that the regulation needed to include GHG emissions from electricity imports. If out-of-state electricity imports were exempt from the regulation, studies projected that emissions leakage would significantly offset emissions reductions achieved under the cap.

emissions leakage

To tackle this GHG leakage problem, the California program requires electricity importers to cover their GHGs with program allowances. Implementing this pioneering policy requires measuring the carbon intensity of electricity imports. A simple approach would be to pick a single value for carbon intensity and apply this to all electricity imports. But carbon intensities vary a lot across power producers. So what’s the right number? 

If you pick a low number, carbon intensive sources of imports (think coal) will be under-taxed and emissions from imports will be under-counted (a.k.a. “carbon laundering). But if you pick a higher number, you unfairly discriminate against low-carbon imports. For example, a wind generator in California pays no cost to comply with the GHG cap-and-trade program. But a wind generator in a neighboring state would have to pay to play in the California market. Unfair protectionism! 

laundering and protectionism

To get around these problems, California has adopted a “source-based” border carbon adjustment. A relatively-high default GHG emissions rate is applied to “unspecified” imports. But if an importer can demonstrate that it’s getting its kWhs from a zero (or low) carbon source, it can zero out (or reduce) its GHG compliance obligation. Problem solved?

Not quite. This source-based approach creates an incentive to preferentially sell (at least on paper) low-carbon, out-of-state sources to California. Meanwhile, carbon-intensive suppliers will be redirected to places where GHGs are not restricted. This resource shuffling (also known as “reshuffling”) will shrink California’s carbon footprint on paper. But it will understate the true aggregate GHG impact of importing electricity into California. 

resource shuffling

California’s climate policy architects could see that resource shuffling had the potential to be a big problem. Economists had documented this depressing fact in great detail (see, for example, work by Jim Bushnell and others). So, at first, California tried to exterminate the problem by fiat. Electricity importers were required to attest they were not resource shuffling, under penalty of perjury. 

Turns out it’s hard to ban something you can’t precisely define or measure. Ultimately, California walked back efforts to ban reshuffling and introduced a suite of nuanced rules and accounting practices that are designed to mitigate the problem.

How’s this border carbon pricing experiment working?

To get a definitive answer to this question, we’d ideally compare GHG emissions in a world with the border carbon adjustment against one without. This would give us a clear read on whether this border adjustment is working to mitigate emissions leakage.

We don’t have parallel worlds at our disposal. But we do have economic models and data. In this paper, we use a detailed model of the western US electricity market to simulate GHG emissions with and without California’s border carbon adjustment (BCA). We simulate least-cost dispatch of the electricity suppliers we observe in the 2019 western wholesale market across three scenarios:

  1. Source-based BCA: A stylized representation of the policy design we observe. California GHGs are directly priced at $17/ton (the 2019 GHG price in California). Out-of-state import suppliers can either accept the default or specify their true GHG intensity. Importantly, we don’t attempt to capture the nuanced rules in place to limit reshuffling in our model.
  2. No BCA: California covers GHG emissions from in-state electricity suppliers, but does not attempt to account for GHG emissions from out-of-state producers. This is an unlimited leakage scenario.
  3. Complete GHG regulation: This is an aspirational fantasy world in which all states join California in pricing GHG emissions across the entire western electricity market. 

Like any modeling exercise, we need to make a bunch of simplifying assumptions to get this model off the ground. These are unpacked in the paper and appendix. The graph below summarizes some punchline results.  

The blue bars show simulated hourly GHG emissions across the western grid for each scenario. The orange bars show California’s GHG accounting for the electricity sector, including deemed emissions from imports. The horizontal lines show the average hourly GHG emissions  we observed in 2019. 

 Three results to highlight:

  1. The potential for reshuffling in the electricity-sector is high: If you’re squinting to see the difference in simulated GHG emissions with and without the border carbon adjustment, there isn’t one. There are lots of zero-carbon sources in neighboring states that can be dispatched to California. The source-based BCA has zero impact on emissions leakage. In other words, California could reshuffle its way to major reductions in its GHGs on paper without making a dent in aggregate, system-wide emissions.
  1. Observed emissions look better than our market model predicts: The GHG emissions we actually observe in the western US electricity market fall below what our  simulations project under unlimited leakage. And California accounted for more GHGs in 2019 than our stylized model predicts under full reshuffling. One reason could be that the rules and limits we’re using to discourage reshuffling are having a mitigating effect.
  1.  Observed emissions exceed our aspirational benchmark: To put these numbers in perspective, the graph shows that observed aggregate GHG emissions (blue line) were 30 percent above simulated emissions in the scenario where all suppliers in the western market pay the carbon price. We can’t expect to hit this complete regulation mark when California is acting alone, no matter how well-designed the BCA. But this comparison underscores the importance of bringing other states (welcome Washington!) into the GHG regulation fold.

A work in progress

This simple comparison between modeling simulation and real-world observation scratches the surface of what we can learn from California’s pioneering efforts. That said, here are two take-aways:

First, pricing carbon at the border is going to be challenging in markets where sources are highly substitutable and highly variable in terms of carbon intensity. In this respect, electricity markets are a worst-case scenario. In other carbon-intensive sectors (like cement or steel or automobiles), it will presumably be more costly/onerous to rearrange supply relationships in response to reshuffling incentives.

Second, California has shown that it’s possible for a jurisdiction regulating its own GHG emissions to price carbon at its borders. This has provided a valuable proof of concept that has endured legal challenges and influenced broader policy efforts. A great first step! But the BCA policy we have in place today has not solved  California’s GHG emissions leakage problem. Our efforts to refine this approach should continue. Lessons learned can help California reduce its carbon footprint. And, more importantly, guide global efforts towards more effective climate policies.

Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas

Suggested citation: Fowlie, Meredith. “The Whac-a-Mole Economics of Border Carbon Adjustments” Energy Institute Blog, UC Berkeley, September 7, 2021, https://energyathaas.wordpress.com/2021/09/07/the-whac-a-mole-economics-of-border-carbon-adjustments/ 

 

  

 

One thought on “The Whac-a-Mole Economics of Border Carbon Adjustments Leave a comment

  1. For California, at least a partial answer would seem to be to constantly update the “unspecified” emission factor to net out the “zero emission” imports in near real time. CARB has not updated the those values since their inception and a recent paper showed that they underestimate those emissions.

%d bloggers like this: