There’s this bit in a new MIT study where they calculate that a Boston company’s flights caused more PM2.5 damage in India and China than in Massachusetts, and it sort of reminded us of those arguments in the early 2000s about whether buying a Prius actually mattered if the battery manufacturing process created more emissions than you’d save over the car’s lifetime.

The underlying question was interesting because it was really asking: are we measuring the right thing here? And honestly, I think we’re still not measuring the right thing, we’ve just gotten much better at measuring the wrong thing.

The MIT researchers published this in Environmental Research Letters and what they found is that organisations have built these incredibly sophisticated carbon accounting systems, whole departments mapping Scope 3 emissions down to the catered lunch at the quarterly all-hands, and yet the thing that’s actually making people sick – the particulate matter, the ground-level ozone – that’s essentially invisible in corporate reporting.

Which feels like it should be a bigger deal than it currently is.

Not all emission reduction is the same

So here’s what the MIT team did. They looked at three organisations in the Boston area, two universities and one company, and they tried to figure out whether different activities that remove the same amount of CO2 would have equivalent benefits for air quality. And the answer, pretty definitively, is no.

They even monetised the damage at $265 per ton of CO2 for aviation versus $88 for electricity purchases. Same carbon reduction on paper but different health outcomes in practice.

And before this starts sounding too theoretical, the European Environment Agency just released data showing how wildly unequal the air quality burden already is, even within supposedly similar regulatory environments.

• Italy saw 43,000 deaths from PM2.5 in 2023.
• Finland saw 34.

That’s three orders of magnitude difference within the EU. And they’re both playing by the same rules, theoretically subject to the same Ambient Air Quality Directive that just got revised. Which suggests pretty strongly that when you’re adding or removing emissions, it matters profoundly whether those emissions land over Milan or over the Baltic Sea. The air quality burden is stubbornly, frustratingly local.

Air quality tackled

So if you’re a construction company trying to hit net zero by 2040, and you’re choosing between electrifying your fleet versus cutting business travel versus switching to alternative fuels, the carbon math might point one direction but the health impact math could point somewhere completely different depending on where your sites are, what the local air quality baseline looks like, and whose lungs are actually processing the particulates you’re either adding or removing.

But most organisations aren’t set up to think this way. They’ve got dashboards showing Scope 1, 2, 3 emissions trending down toward their science-based targets, and that’s what gets reported to the board and published in the sustainability report and used to answer customer RFPs. The air quality dimension, the near-term health impact of their decarbonisation choices, is largely invisible.

Which creates this odd situation where you could hit your net zero target right on schedule while making the air measurably worse for the people living next to your operations. But you’d never know which you’re doing without actually measuring the local response.

The measurement technology to help understand all of this exist. Continuous PM2.5 monitoring is standard practice on construction sites in London.

Businesses are spending serious money on carbon accounting platforms, hiring consultants to map their value chain emissions, setting science-based targets. But if the goal is actually to reduce harm – to make fewer people sick, to improve air quality in the communities where you operate – then tracking carbon alone is like trying to navigate using only longitude. You’ve got half the information you need, and you’re just hoping the other half works out.

tracking carbon alone is like trying to navigate using only longitude. You’ve got half the information you need, and you’re just hoping the other half works out.

It’s not that carbon tracking is wrong. Climate change is real and urgent and reducing atmospheric CO2 matters enormously. But it’s incomplete if what you actually care about is the health impact piece, and I think most organisations do care about that piece even if it’s not showing up in their reporting frameworks yet.

But you can’t optimise for something you’re not measuring.

And most organisations aren’t measuring the local air quality dimension of their decarbonisation strategies, which means they’re potentially missing opportunities to do more good sooner, even as they’re making real progress on the long-term climate piece.

Which feels like something worth fixing, honestly. If you want to take the first step, let’s talk!