We're Making Bad Decisions Because We're Using Bad Metrics

Nearsighted Metrics Reward Nearsighted Choices

Imagine you’re fresh out of high school, deciding what to do with your life. Financial security is important to you. So, obviously, you go work at McDonald’s.

Wait, what? Well, McDonald’s is hiring, and you don’t need any special qualifications. You can start earning money right away. If you need to earn more, you can take a second job stocking shelves at the Walmart down the road.

Sure, you could go to college, or trade school, or at least send out resumes to look for something that pays more than minimum wage. But none of those actions would earn you more money today.

Too much activity around climate suffers from a similarly shortsighted metric: “emissions reduced today”. If a company makes a net-zero pledge, then for each ton of CO₂ they emit, they are expected to find some compensating action that removes (or, sigh, “avoids”) one ton of CO₂. If you go to a site like TerraPass and pay to offset the carbon emissions of your next flight, then again, they’ll be promising to find an equivalent specific reduction in near-term emissions. When a charitable foundation evaluates grant proposals, they’ll often be evaluating them using the same yardstick. Tax credits and cap-and-trade programs, same deal.

For example, a coal plant operator in the US can receive a $50M 45Q tax credit for installing equipment to capture and sequester one million tons of the plant’s carbon emissions. That same $50M might go a lot farther in the hands of an organization like Spark Climate, laying the groundwork to address neglected sectors such as methane emissions. But the IRS won’t give you a 45Q credit for supporting that work, and TeraPass won’t go there either. It doesn’t measure on their scale.

Tracking progress on climate change by the change in CO₂ emissions is like tracking progress on your job hunt by the change in salary. It’s the right long-term goal, but a terrible short-term yardstick.

Nearsighted Metrics Channel Money Where It’s Least Needed

In Optimizing for 2050 and Mind the Gaps, I talk about how we need to be making early-stage investments to address every sector of the world economy. It’s not enough to scale up the solutions that are already working – solar and wind power, electric cars, and so forth. We need to lay the groundwork to scalably, economically address every source of greenhouse emissions. We need to clear away all the barriers that stand in the way of a zero-carbon economy.

Unfortunately, the focus on immediate emission reductions steers most climate spending into the handful of solutions that are already scaling, and have the least need for financial support. Meanwhile, other sectors and solutions receive a fraction of the funding they merit¹. For instance, from a Founders Pledge report:

Philanthropic funding skews heavily towards clean electricity. Philanthropists continue to pay little attention to hard-to-decarbonize sectors. While there is some uptick in philanthropic funding for those sectors such as industry and heavy-duty transport through Bezos Earth Fund commitments, these are dwarfed by other increases. Insofar as philanthropy sets the agenda for civil society and policy, this is worrisome as it continues the overemphasis on relatively mature and popular technologies rather than focusing on those parts of the economy that are not yet trending clean.

It’s not just philanthropy. One of the largest sources of climate spending is corporations seeking to “offset” their emissions. Much of this goes to projects of dubious merit, such as questionable forestry or land-preservation projects. But even the highest-quality programs are generally evaluated based on how much they reduce emissions today, rather than how well they speed our path to net zero.

Frontier, a $925M commitment from a group of (mostly) tech companies to pay for permanent carbon removal, stands out specifically because they are not seeking out the cheapest offsets they can find; instead, they are willing to pay higher prices in order to foster new solutions and incubate the portfolio of permanent carbon removal technologies the world urgently needs. But mostly, dollars are being spent on low-leverage projects, such as subsidies for yet another wind farm – or, worse, cheap low-quality offsets that don’t accomplish much of anything, but do register on someone’s insufficiently rigorous definition of “avoided emissions”.

Finding a Better Metric Is Difficult, But Necessary

As the cliche goes, “what gets measured gets done”. If we want to encourage more activity and funding around measures that accelerate the path to net zero, we need a way of measuring that acceleration.

This won’t be easy. How do we rigorously define “accelerating the path to net zero”? How do we account for the inherent uncertainty of projects that are still at the research stage? How do we account for the need to address methane (and other short-lived gases) especially quickly?

Should our metric allow for political actions? A 2021 paper proposes the idea of “political carbon offsets” – measuring the carbon impact of donating to climate-friendly politicians. This captures an important idea – influencing policy can be a very efficient way of improving climate outcomes. And we need the corporate world to lobby for climate action. But this is inherently controversial, and including political action in our metric might make it harder for some parties to adopt.

Then there’s Goodhart's Law, which basically says that as soon as you start measuring people by a metric, they start finding ways of gaming it. (See: low-quality carbon offsets.) The more complicated our metric, the easier it will be to find loopholes.

But the hard work of defining a metric is basically the hard work of deciding where resources are needed. If we can’t do that, we can’t expect to make good decisions.

What a Wonderful World It Would Be

Imagine that we manage to define a metric that truly measures progress toward net zero, even if approximately. How wonderful would that be?

Philanthropists, government agencies, corporate sustainability teams, and other decision makers would have clear guidance as to how to make best use of their resources.

Those same decision makers would have a powerful tool for defending decisions that optimize for the long term.

Purveyors of low-quality offsets would have a harder time justifying their work.

And we’d have a handy benchmark for progress that wouldn’t suffer from the same near-term focus as the metrics everyone talks about today.

I don’t know whether we can feasibly define such a metric, or get people to pay attention to it. But it sure seems worth trying.

1. The Inflation Reduction Act, which just cleared the House!!!, will do a lot to help here. It includes very substantial funding for a wide variety of sectors, some of which have been relatively neglected. Of course there is still plenty of further work to be done, but this will be a big step in the right direction. And it’s worth noting that discussion of this bill centers on the impact it will have on emissions in 2030 and beyond, not what it will do this year.