Escaping Dirty Growth

Last weekend, I was talking with a friend about Spanish politics. Typically, and more so with economists, it's difficult to talk about politics and not end up talking about economics. He correctly pointed out that the great GDP growth numbers coming out of Spain are central to the ruling party’s talking points; alongside some progressives, he finds GDP growth an unhealthy obsession.

At the core of this sentiment, is the interpretation that the growth we are experiencing is dirty. And well, this is true from an environmental perspective: the more we produce with existing technologies, the more emissions we put into the atmosphere, the closer we get to a warmer world, and the more frequent extreme weather events become. Moreover, these phenomena hit the poor the hardest, the jugular vein of progressive concerns.

However, I argue that the key here is not quantitative, rather qualitative. It is not about growing more or less, rather making growth clean instead. If we are to decouple growth from all of these negative environmental impacts, we need to rely on an entirely different set of technologies. Some of these clean technologies already exist: solar PV, wind turbines, or energy storage. But, many need to be invented or significantly improved: carbon capture and storage or low-carbon cement. One might wonder, how can we foster this technological revolution?

The Problems

As most economists do, I am going to use a model to discipline my reasoning, and highlight two potential solutions.^[1] I separate the discussion of each solution for simplicity, but keep in mind that these are complementary policies.

I am going to assume that you fundamentally care about your consumption and that you are logically consistent.^[2] Furthermore, I am going to let you buy whatever combination you want, given your budget, of a clean and a dirty product. Now, here is my prediction: you will buy relatively more of the clean product if it is cheaper than the dirty product, as long as the clean product can sufficiently replace the purpose of the dirty product.^[3]

If you find such predictions reasonable, we can agree that two problems arise when greening people’s consumption basket: clean products are too expensive compared to dirty products, and/or dirty products have qualities that cannot be replaced by clean products. The choice between internal combustion engine (ICE) cars and electric vehicles (EVs) is a good example: a) EVs are more expensive, b) EVs currently fail to fully replace the functionalities of an ICE car due to their limited maximum range and scarce charging infrastructure.

I am going to add more complexity into the model to illustrate other mechanisms at play. Instead of just imposing that the clean product is more expensive than the dirty product, I am going to let profit maximizing firms choose prices and investment. These firms produce both dirty and clean products. Moreover, I am going to assume the firms have some market power, allowing them to accumulate profits that can be reinvested towards either clean or dirty research, which lowers the cost of clean and dirty products, respectively.

If this setup seems reasonable, notice the emergence of a feedback loop: given the dirty product’s market is larger to begin with (remember the ICE vs EV example), the return on investment of dirty R&D will be higher.^[4] This makes things worse for clean innovation in the future, because you will buy more of the dirty product, incentivizing firms to invest in more dirty R&D, further lowering the dirty product’s cost over time relative to the clean product: a dangerous death spiral.^[5]

The Solutions

Alright, so which policy levers can get us out of this mess? Economists have traditionally proposed taxing the dirty product for its carbon content, to the extent carbon emissions damage society.^[6] It should be noted, however, that the political unpopularity of carbon taxes has led policymakers to pursue alternative, second-best instruments, which, ultimately, also stimulate (discourage) demand for clean (dirty) products, such as tax credits or mandates in the transportation sector’s case. At the same time, this relative increase in the demand for the clean product increases the returns of clean R&D. In addition, note that the societal damage produced by carbon emissions increases as we emit more: the hotter the climate, the more harmful further temperature increases are. Therefore, the tax or second-best measure would become more stringent over time up until the point where emitting more is just not allowed, or too expensive, at which point you will probably choose to spend all your money on the clean product.

A complementary solution is a direct clean research subsidy. This would be set high enough so that firms find it more profitable to invest in clean R&D, and would be deployed until firms find it more profitable to invest in clean even in the absence of the subsidy. As firms find ways to produce the clean product more efficiently, the clean product will eventually be cheaper than the dirty product. Consequently, you will buy more and more of the clean product, which in turn gives firms a bigger incentive to keep investing in clean research: replacing the dangerous death spiral with a convenient virtuous circle.

The Takeaways

Ultimately, this model illustrates two complementary policy dimensions which make clean growth possible: taxing dirty products and directly encouraging innovation in clean products. In a world divided into hundreds of national jurisdictions, each with distinct climates and levels of economic development, the way we should approach each policy lever is not trivial. In this article series, we will focus on and navigate the innovation policy dimension. Stay tuned.

Footnotes:

1. The model mimics Acemoglu, D., Aghion, P., Bursztyn, L., & Hemous, D. (2012). The environment and directed technical change. American Economic Review, 102(1), 131–166. https://doi.org/10.1257/aer.102.1.131 ↑

2. Logical consistency understood as compliance to three conditions: completeness (you are able to compare any alternatives), transitivity (if you prefer A to B and B to C, you must prefer A to C),and continuity (if you prefer A to B and C is really similar to A, then you prefer C to B). ↑

3. Economists formally define this as elasticity of substitution. ↑

4. Innovations that improve ICE vehicles generate higher expected profits than comparable innovations in EVs, largely because the ICE market serves a substantially larger pool of buyers today. Although the EV market may grow over time, profits realized in the future are discounted, lowering the present value of innovation in EV technologies. ↑

5. Acemoglu (2024 Nobel Prize Winner), Aghion (2025 Nobel Prize Winner), Bursztyn, and Hemous prove an analogous phenomenon in their paper The Environment and Directed Technical Change (2012), which they label as “path dependence”. ↑

6. The optimal tax per ton of CO₂ is equal to the economic cost of emitting an additional ton of CO₂. ↑