Measuring environmental damage: A new perspective from the production side
By Kevin Fox and Erwin Diewert
When economists try to put a number on pollution’s impact on society, they face a surprisingly tricky problem. Existing approaches have made important conceptual advances, but they also face significant practical and empirical challenges.
In a new ESCoE discussion paper, we offer a complementary, production-based perspective that helps address some of these challenges and strengthens the link between measurement and policy.
Since the 1970s, economists have relied heavily on a framework developed by Martin Weitzman to measure how environmental problems affect our well-being. This welfare-based approach provides a powerful conceptual lens, linking environmental change to consumer welfare and overall economic performance.
In practice, however, applying this framework raises significant measurement challenges. Much of the empirical implementation focuses on consumer behaviour, asking how pollution affects people’s ability to enjoy goods and services. But consumers do not choose the level of pollution they experience. You can decide whether to buy a new car or take a vacation, but you cannot decide how much smog is in your city’s air or how contaminated your local river is.
This creates a practical identification problem. When economists try to estimate how much people value environmental quality using standard consumer demand models, they run into difficulties. The mathematical framework, as typically implemented, struggles to capture the full welfare effects of pollution because consumers are making choices conditional on a pollution level they don’t control, rather than actively choosing between different pollution scenarios.
To put it in technical terms: if pollution affects your utility but you can’t choose its level, then we cannot observe how you would trade off market goods for cleaner air or water. The data needed to measure these preferences simply does not exist in consumer purchasing patterns.
The paper explores a complementary producer-side perspective on environmental damage. This focuses on how businesses generate pollution as a by-product of producing goods and services.
This shift in perspective opens up new measurement possibilities. Unlike consumers, producers do have some control over pollution levels through their choice of inputs, technologies, and production methods. More importantly, we can observe and measure these choices.
We outline three different frameworks for incorporating environmental damage into production accounting, with varying levels of sophistication. The most promising approach links environmental damage to pollution taxes, using them as a transparent and policy-relevant way to value emissions from the producer perspective.
Even small taxes give producers an incentive to reduce emissions. This might lead them to adopt cleaner technologies, switch to less polluting inputs, or shift their product mix toward goods that generate less pollution.
If we know how different inputs generate pollution (say, burning coal produces certain amounts of carbon dioxide and particulates), we can adjust input prices to reflect their full social cost. A ton of coal isn’t just worth its market price – it’s worth that price plus the pollution taxes on the emissions it creates.
With this adjustment, standard economic tools for measuring productivity and efficiency can be applied, and we can track how the economy is performing while accounting for environmental damage.
Governments should impose at least small pollution taxes even if they are modest, because this forces firms to start tracking their emissions systematically. Better information leads to better policy, and even minimal taxes create incentives to reduce pollution.
Our preferred framework treats environmental bads as negative outputs rather than as inputs or as things that affect consumer welfare directly. When a factory produces both steel and air pollution, the pollution, valued at its tax rate, should be subtracted from the total value of production. This gives us a “net product” that accounts for environmental damage.
This approach has several advantages. It recognises that pollution is actively produced by the economy, not just passively consumed. It works with the data we can actually observe and measure. And it fits naturally with how we think about “net” national product (the total value of what an economy produces, minus what it uses up or degrades).
The implications extend beyond academic economics. If environmental damage is measured incompletely, it can impact policymaking. The consumer-focused approach that has dominated environmental accounting for decades faces challenges in capturing the full welfare effects of pollution and environmental degradation in practice.
Production-based frameworks (especially ones that incorporate pollution taxes) offer the potential to develop better metrics for tracking whether environmental quality is improving, quantify the true costs of different industries, and design more effective environmental policies.
Of course, this framework does not solve every problem. We acknowledge that a complete welfare analysis would need to account for how tax revenues from pollution charges are used, how different taxes create economic distortions, and how consumers ultimately bear the costs of environmental cleanup.
Our aim is to offer a practical alternative grounded in production theory. By focusing on observable production decisions and policy instruments that are already in use, this approach provides economists and policymakers with an additional tool for measuring and managing environmental damage. Our approach is particularly well suited to empirical implementation and policy analysis.
A natural next step is the introduction of pollution taxes, even at modest levels, to generate both incentives and information. These taxes encourage firms to track emissions systematically and to respond by adopting cleaner technologies and production methods. Over time, we can build measurement frameworks that better reflect the true costs of environmental bads, drawing not on unobservable consumer preferences, but on the observable ways in which production generates pollution and how producers respond to policy incentives to reduce it.
This blog introduces a complementary, production-based approach to measuring environmental damage that focuses on how pollution is generated through economic activity. By treating pollution as a negative output and valuing it using pollution taxes, the approach offers a practical way to link environmental measurement to observable data and policy instruments. Used alongside welfare-based frameworks, production-side measures can improve how we track environmental damage, compare industries, and design effective environmental policy.
ESCoE blogs are published to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of ESCoE, its partner institutions, the Office for National Statistics, or the UK Endorsement Board.