About six months ago, a climate scientist from Arizona State University, Stephanie Arcusa, emailed me a provocative new paper she had published that warned against our growing reliance on life cycle analysis. This practice of measuring all of the emissions related to a given product or service throughout every phase of its life — from the time raw materials are extracted to eventual disposal — was going to hinder our ability to achieve net-zero emissions, she wrote. It was a busy time, and I let the message drift to the bottom of my inbox. But I couldn’t stop thinking about it.
Life cycle analysis permeates the climate economy. Businesses rely on it to understand their emissions so they can work toward reducing them. The Securities and Exchange Commission’s climate risk disclosure rule, which requires companies to report their emissions to investors, hinges on it. The clean hydrogen tax credit requires hydrogen producers to do a version of life cycle analysis to prove their eligibility. It is central to carbon markets, and carbon removal companies are now developing standards based on life cycle analysis to “certify” their services as carbon offset developers did before them.
At the same time, many of the fiercest debates in climate change are really debates about life cycle analysis. Should companies be held responsible for the emissions that are indirectly related to their businesses, and if so then which ones? Are carbon offsets a sham? Does using corn ethanol as a gasoline substitute reduce emissions or increase them? Scientists have repeatedly reached opposite conclusions on that one depending on how they accounted for the land required to grow corn and what it might have been used for had ethanol not been an option. Though the debate plays out in calculations, it’s really a philosophical brawl.
Everybody, for the most part, knows that life cycle analysis is difficult and thorny and imprecise. But over and over, experts and critics alike assert that it can be improved. Arcusa disagrees. Life cycle analysis, she says, is fundamentally broken. “It’s a problematic and uncomfortable conclusion to arrive at,” Arcusa wrote in her email. “On the one hand, it has been the only tool we have had to make any progress on climate. On the other, carbon accounting is captured by academia and vested interests and will jeopardize global climate goals.”
When I recently revisited the paper, I learned that Arcusa and her co-authors didn’t just critique life cycle analysis, they proposed a bold alternative. Their idea is not economically or politically easy, but it also doesn’t suffer from the problems of trying to track carbon throughout the supply chain. I recently called her up to talk through it. Our conversation has been edited for clarity.
Can you walk me through what the biggest issues with life cycle analysis are?
So, life cycle analysis is a qualitative tool —
It seems kind of counterintuitive or even controversial to call it a qualitative tool because it’s specifically trying to quantify something.
I think the best analogy for LCA is that it’s a back-of-the-envelope tool. If you really could measure everything, then sure, LCA is this wonderful idea. The problem is in the practicality of being able to collect all of that data. We can’t, and that leads us to use emissions factors and average numbers, and we model this and we model that, and we get so far away from reality that we actually can’t tell if something is positive or negative in the end.
The other problem is that it’s almost entirely subjective, which makes one LCA incomparable to another LCA depending on the context, depending on the technology. And yes, there are some standardization efforts that have been going on for decades. But if you have a ruler, no matter how much you try, it’s not going to become a screwdriver. We’re trying to use this tool to quantify things and make them the same for comparison, and we can’t because of that subjectivity.
In this space where there is a lot of money to be made, it’s very easy to manipulate things one way or another to make it look a little bit better because the method is not robust. That’s really the gist of the problems here.
One of the things you talk about in the paper is the way life cycle analysis is subject to different worldviews. Can you explain that?
It’s mostly seen in what to include or exclude in the LCA — it can have enormous impacts on the results. I think corn ethanol is the perfect example of how tedious this can be because we still don’t have an answer, precisely for that reason. The uncertainty range of the results has shrunk and gotten bigger and shrunk and gotten bigger, and it’s like, well, we still don’t know. And now, this exact same worldview debate is playing into what should be included and not included in certification for things [like carbon removal] that are going to be sold under the guise of climate action, and that just can’t be. We’ll be forever debating whether something is true.
Is this one of those things that scientists have been debating for ever, or is this argument that we should stop using life cycle analysis more of a fringe idea?
I guess I would call it a fringe idea today. There’s been plenty of criticism throughout the years, even from the very beginning when it was first created. What I have seen is that there is criticism, and then there is, “But here’s how we can solve it and continue using LCA!” I’ve only come across one other publication that specifically said, “This is not working. This is not the right tool,” and that’s from Michael Gillenwater. He’s at the Greenhouse Gas Management Institute. He was like, “What are we doing?” There might be other folks, I just haven’t come across them.
Okay, so what is the alternative to LCA that you’ve proposed in this paper?
LCA targets the middle of the supply chain, and tries to attribute responsibility there. But if you think about where on the supply chain the carbon is the most well-known, it is actually at the source, at the point of origin, before it becomes an emission. At the point where it is created out of the ground is where we know how much carbon there is. If we focus on that source through a policy that requires mandatory sequestration — for every ton of carbon that is now produced, there is a ton of carbon that’s been put away through carbon removal, and the accounting happens there, before it is sold to anybody — anybody who’s now downstream of that supply chain is already carbon neutral. There is no need to track carbon all the way down to the consumer.
We know this is accurate because that is where governments already collect royalties and taxes — they want to know exactly how much is being sold. So we already do this. The big difference is that the policy would be required there instead of taxing everybody downstream.
You’re saying that fossil fuel producers should be required to remove a ton of carbon from the atmosphere for every ton of carbon in the fuels they sell?
Yeah, and maybe I should be more specific. They should pay for an equal amount of carbon to be removed from the atmosphere. In no way are we implying that a fossil carbon producer needs to also be doing the sequestration themselves.
What would be the biggest challenges of implementing something like this?
The ultimate challenge is convincing people that we need to be managing carbon and that this is a waste management type of system. Nobody really wants to pay for waste management, and so it needs to be regulated and demanded by some authority.
What about the fact that we don’t really have the ability to remove carbon or store carbon at scale today, and may not for some time?
Yes, we need to build capacity so that eventually we can match the carbon production to the carbon removal, which is why we also proposed that the liability needs to start today, not in the future. That liability is as good as a credit card debt — you actually have to pay it. It can be paid little by little every year, but the liability is here now, and not in the future.
The risk in the system that I’m describing, or even the system that is currently being deployed, is that you have counterproductive technologies that are being developed. And by counterproductive, I mean [carbon removal] technologies that are producing more emissions than they are storing, and so they’re net-positive. You can create a technology that has no intention of removing more carbon than its sequesters. The intention is just to earn money.
Do you mean, like, the things that are supposed to be removing carbon from the atmosphere and sequestering it, they are using fossil fuels to do that, and end up releasing more carbon in the process?
Yeah, so basically, what we show in the paper is that when we get to full carbon neutrality, the market forces alone will eliminate those kinds of technologies that are counterproductive. The problem is during the transition, these technologies can be economically viable because they are cheaper than they would be if 100% of the fossil fuel they used was carbon neutral through carbon removal. And so in order to prevent those technologies from gaming the system, we need a way to artificially make the price of fossil carbon as expensive as it would be if 100% of that fossil carbon was covered by carbon removal.
That’s where the idea of permits comes in. For every amount that I produce, I now have an instant liability, which is a permit. Each of those permits has to be matched by carbon removal. And since we don’t have enough carbon removal, we have futures and these futures represent the promise of actually doing carbon removal.
What if we burn through the remaining carbon budget and we still don’t have the capacity to sequester enough carbon?
Well, then we’re going into very unchartered territory. Right now we’re just mindlessly going through this thinking that if we just reduce emissions it will be good. It won’t be good.
In the paper, you also argue against mitigating greenhouse gases other than carbon, and that seems pretty controversial to me. Why is that?
We’re not arguing against mitigating, per se. We’re arguing against lumping everything under the same carbon accounting framework because lumping hides the difficulty in actually doing something about it. It’s not that we shouldn’t mitigate other greenhouse gases — we must. It’s just that if we separate the problem of carbon away from the problem of methane, away from the problem of nitrous oxide, or CFCs, we can tackle them more effectively. Because right now, we’re trying to do everything under the same umbrella, and that doesn’t work. We don’t tackle drinking and driving by sponsoring better tires. That’s just silly, right? We wouldn’t do that. We would tackle drinking and driving on its own, and then we would tackle better tires in a different policy.
So the argument is: Most of climate change is caused by carbon; let’s tackle that separately from the others and leave tackling methane and nitrous oxide to purposefully created programs to tackle those things. Let’s not lump the calculations altogether, hiding all the differences and hiding meaningful action.
Is there still a role for life cycle analysis?
You don’t want to be regulating carbon using life cycle analysis. So you can use the life cycle analysis for qualitative purposes, but we’re pretending that it is a tool that can deliver accurate results, and it just doesn’t.
What has the response been like to this paper? What kind of feedback have you gotten?
Stunned silence!
Nobody has said anything?
In private, they have. Not in public. In private, it’s been a little bit like, “I’ve always thought this, but it seemed like there was no other way.” But then in public, think about it. Everything is built on LCA. It’s now in every single climate bill out there. Every single standard. Every single consulting company is doing LCA and doing carbon footprinting for companies. It’s a huge industry, so I guess I shouldn’t have been surprised to hear nothing publicly.
Yeah, I was gonna ask — I’ve been writing about the SEC rules and this idea that companies should start reporting their emissions to their investors, and that would all be based on LCA. There’s a lot of buy-in for that idea across the climate movement.
Yeah, but there’s definitely a fine line with make-believe. I think in many instances, we kid ourselves thinking that we’re going to have numbers that we can hang our hats on. In many instances we will not, and they will be challenged. And so at that point, what’s the point?
One thing I hear when I talk to people about this is, well, having an estimate is better than not having anything, or, don’t let the perfect be the enemy of the good, or, we can just keep working to make them better and better. Why not?
I mean, I wouldn’t say don’t try. But when it comes to actually enforcing anything, it’s going to be extremely hard to prove a number. You could just be stuck in litigation for a long time and still not have an answer.
I don’t know, to me it just seems like an endless debate while time is ticking and we will just feel good because we’ll have thought we measured everything. But we’re still not doing anything.
