Climate policy: where economics went wrong
Regular readers of this blog will be aware of a twelve month hiatus since I last posted anything. There is a good reason. Corinne Sawers, a sustainability expert, and I have co-authored Supercharge Me, a discussion on designing policies to collapse carbon emissions. The book is short, and should be readable in half a day. It should also be accessible to a very broad readership. This required some constraint on our nerdier tendencies. To make up for this, we will be posting a series of articles looking at many of the themes in more detail. Here is the first. It explains how economists, by focusing on the theory of externalities, stopped at the wrong chapter of the textbook. One result has been incoherence and ineffectiveness of policy-making. As we argue in Supercharge Me, the better chapters relate to regulating utilities (the subject of a forthcoming paper), and the price elasticity of demand – the focus here.
Economists have dominated thinking on climate change policy. As is often the case in economics, the most damage is done right at the beginning, almost without thinking.
Nobel prize winner, William Nordhaus perfectly frames ‘how economists think about carbon emissions’, drawing on the concept of an externality – an idea which dates back to the great Scottish political economist, David Hume. The idea of an externality is relatively simple: a cost or benefit of an activity which is not captured by the market price. Pollution is the textbook example. The price of clothes fails to capture the cost or damage caused by chemicals, dyes and carbon emissions pollution that results from their production. In The Climate Casino (pp18-19), Nordhaus spells it out:
“The economics of climate change is straightforward. Virtually everything we do involves, directly or indirectly, the combustion of fossil fuels, which has resulted in emissions of carbon dioxide in the atmosphere. […]
The problem is that those who produce the emissions do not pay for that privilege, and those who are harmed are not compensated. […]
Economists call such costs externalities because they are external to (i.e., not reflected in), market transactions or prices. An externality is a by-product of economic activity that causes damage to innocent bystanders. These are also called public goods in the economics literature …”
The power of this framing should not be under-estimated. It is the preeminent focus of climate policy makers and is typically assumed to be the basis of policy by non-economist experts. For example, in his clearly-written book, How to Avoid a Climate Disaster (p186), Bill Gates, argues:
“Today, when businesses make products or consumers buy things, they don’t bear any extra cost for the carbon involved, even though that carbon imposes a very real cost on society. This is what economists call an externality … There are various ways, including a carbon tax and cap-and-trade program, to ensure that at least some of these external costs are paid for by whoever is responsible for them.”
In Doughnut Economics, the thoughtful and insightful re-thinking of macroeconomics, Kate Raworth rejects the idea of an externality, arguing that a systems-based framing of the problem, wherein externalities are internalised (p143). But when discussing policies, and specifically, taxes, even she seems more sympathetic:
“Economic theory recognises the potentially damaging effects – the ‘negative externalities’ – of such industry, and has favoured market-based tools for addressing them: quotas and taxes. […]
Taxes, quotas and tiered pricing can clearly help to ease humanity’s pressure on Earth’s sources and sinks, but here’s the trouble with believing they will do the whole job. In practice they fall short because they are rarely set to the level required …”
Raworth’s scepticism is warranted. But we would go further. The problem is not that the level of carbon taxes is too low, but starting with taxes in the analytical context of externalities leads down the wrong path.
Broadly, this summarises the primary influence of economics on climate policy. Emissions are an externality, and ‘pricing’ them accurately will decarbonise our economy. The main beef economists are perceived to have is that carbon taxes haven’t worked because they are not high enough.
So why do we believe this is the wrong chapter of the texbook to drive rapid decarbonisation? Part of the force of this framing – arguably its most valid dimension – is the ethical subtext: those responsible for damage should pay for it. This is relevant to who should bear the costs of the green transition – and something we explore in Supercharge Me when discussing asset values, but in the meantime, we need to collapse emissions – very quickly. Given this goal, which are the right chapters of the economics textbook? In Supercharge Me, we argue that the microeconomic theory we should be focused on is not the treatment of externalities, but two very different areas: the chapter on substitutes and the price elasticity of demand, and the chapter on regulating natural monopolies. Here’s the punchline: the policy implications could not be more different. What is crystal clear is that you do not start with a carbon tax. It can play a role, but late in the game.
We will discuss the relevance of the chapter on regulating utilities in another post. The elevator pitch is that sustainable electricity is a central policy goal in any viable approach to collapsing emissions. Electricity generation is a challenge of utility regulation, and we require a rapid acceleration in capital investment – the key to which is a de-risking of electricity prices, a reduction in the cost of capital, vastly more ambitious investment timelines and smart regulation. Carbon taxes are not the central lever. To put the significance of sustainable electricity in context, around 30% of global emissions are directly created by electricity generation. Electrifying buildings, transport and heavy industry can get this share of emissions to ~70%.
The price elasticity of demand may sound like a mouthful – a marketing failure compared to ‘externality’. Unfortunately, it is the crux of the matter, and again points to a profoundly different approach to policy. The history of carbon taxes should really have provided a heads-up. Fuel duties were introduced in America to pay for federal road infrastructure maintenance. We don’t need to go into the theory of optimal taxes, but one of the main attractions is precisely because petrol (or gas) is price inelastic.
In other words, when you raise the price of gas, demand does not change very rapidly. Fuel duties have historically been successful precisely because they don’t significantly change behaviour and reduce emissions. By contrast, they raise lots of revenue. This is the contradiction at the heart of carbon taxes. We want a rapid change in behaviour. We want everyone to stop consuming carbon. So a successful carbon tax which delivered these two objectives would raise little revenue.
The policy challenge is not to raise the price of gas so high that workers are unable to afford to travel to work, and that businesses go bankrupt. The problem is not that the carbon price should have been far higher. The key issue is price elasticity. In a climate policy aimed at collapsing emissions we in fact want to dramatically alter the price elasticity of demand for carbon-intensive goods and services, and then we need to target relative prices. What does this chapter of the textbook tell us to do? And what are the implications for policy and for sequencing?
If we focus on the price elasticity, the key lesson is to create substitutes. Demand inelasticity is high among goods and services which are essential, and for which there are few close substitutes. Ideally, we want near-perfect substitutes. This is of course relatively obvious. But being explicit about the analytical framework points to a very different set of policies. What is the most effective way to tackle emissions in road transport? We can simply halt road transport – largely an exercise in state force, whether simply through banning forms of transport or a tax so high that it makes its unaffordable for large swathes of the population – or we can accelerate the adoption of substitutes. But fuel taxes do not affect the price of a substitute. Targeting the relative price of electric vehicles does. Petrol is price inelastic, but petrol-fuelled vehicles are not.
Fuel demand is partly price inelastic precisely because we have collectively sunk a major capital cost in internal combustion engine vehicles (ICEs). So what has been the most successful policy to date? Targeting the relative price of a close substitute. As we illustrate in Supercharge Me, the nations and cities which have reached 50% or 90% EV share of total car sales are those which are pricing them at significantly lower list prices than ICE vehicles. In order to make electric vehicles proximate to a perfect substitute, there must be an extensive charging infrastructure. And to change behaviour rapidly at scale there needs to be a major price differential in favour of the green substitute. Targeting our political and financial resources at fuel taxes will not achieve this. Charging infrastructure and the relative price of EVs is the key. The third, and most costly leg of this journey, is to incentivise the accelerated depreciation of the existing stock.
Bill Gates tries to shoehorn the economics of externalities in this direction (p186). He talks about reducing the ‘Green premium’ that we currently pay for less carbon-intensive alternatives:
“By progressively increasing the price of carbon to reflect its true cost, governments can nudge producers and consumers toward more efficient decisions and encourage innovation that reduces Green Premiums. You’re a lot more likely to try and invent a new kind of electrofuel if you know it won’t be undercut by artificially cheap gasoline.”
Sometimes being half-right in theory, results in being 100% wrong in practice. And the economics of price elasticity suggests that gradualist, modest reductions in ‘Green Premiums’ don’t make rapid differences at scale. Primarily due to price inelasticity, carbon taxes have typically come at considerable political cost and little gain, a point well made by the Canadian climate economist, Mark Jaccard. The economics of price elasticity says that either you need perfect substitutes and a small price advantage, or near-substitutes and a large price advantage, to dramatically affect behaviour. In Supercharge Me, we introduce the concept of EPICs. Overwhelmingly, this policy is shown to work in the success stories. If we are serious about transforming road transportation, we need to invest rapidly in a charging infrastructure and ensure that the combination of taxes, exemptions and subsidies, resulting in the list price of electric vehicles being substantially below ICEs.
The analytical clarity of focusing on price elasticity of demand is mission-critical – particularly if sequencing in reverse is self-defeating, which it may be both economically and politically. This strategy needs to be repeated across all major sectors, including manufacturing, steel, cement, and really difficult areas of behavioural change such as meat and dairy consumption. We need to mobilise investment in substitutes, ensure they are ‘close’ enough for the user and that they are priced more favourably than the emissions intensive option. In this relative pricing, taxes can and will play a role, but they will only contribute to reducing emissions once the near-substitute exists. In Supercharge Me, we argue that contingent carbon taxes in various forms can be more powerful. The risk of future taxes affects the incentives of investments pending today. Legacy oil companies, such as BP and Shell, are now effectively operating with dual costs of capital now – high double-digits for fossil fuel development, single digits for offshore wind. This is precisely what we need to occur. Businesses which have access to cheap capital redirecting capital investment at scale. Now, it is likely that the threat of carbon tax on future oil revenues, combined with a host of other factors, helps in skewing this cost of capital assessment. But contingent taxes embed sequencing – they incentivise the creation of substitutes without cutting off the funding. Taxes are useful – indeed essential – but not in the context of correctly pricing externalities. And framing the problem as Nordhaus and most of economics has done since (Raworth is a notable exception), or attempting to devise highly complex global carbon taxes, is distracting and likely ineffective.
Why has this framing of externalities and the pricing of carbon had such a pre-eminent hold on climate policy? A point we make in Supercharge Me, is that the power of this idea is significantly attributable to its ethical force, rather than its practical relevance: we want those who are responsible to pay. But the really important conceptual frameworks for rapidly collapsing emissions relate to accelerating investment in regulated utilities, and natural monopolies, which are the market elements of electricity transmission and distribution, and how to alter the price elasticity of demand.
Within this context, there is a role for smart taxes, but the purpose is very different. They are not aimed at correctly pricing an externality. Taxes, exemptions and credit guarantees, should serve one purpose – accelerating the depreciation of carbon-intensive assets, financing the creation of carbon free alternatives, creating substitutes at scale, and targeting the relative price of substitutes. This will rapidly decrease emissions, whilst ensuring the overwhelming majority are better off.
This article is co-authored with Corinne Sawers.Supercharge Me will be published by Agenda in February.
Raworth goes on to argue, “These policies fall short in theory too: from a systems-thinking perspective, quotas and taxes to limit the stock and reduce the flow of pollution are indeed leverage points for changing a system’s behaviour – but they are low points of leverage.”
In fairness to Stiglitz and Stern, who do go into the theory of optimal taxation, like Raworth, they also acknowledge that over-emphasis on the theory of externalities has hampered policy-making. They draw this conclusion not from the perspective of ‘systems theory’, but rather expanding the framework of ‘market failure’. Again when it comes to policy, the emphasis unfortunately falls back on trying to ‘price’ carbon accurately. They make little reference to sequencing, price elasticity of demand, or the success stories of rapid capex in regulated utilities.
The long and short run price elasticity of demand is discussed in the context of carbon taxes in the excellent paper by Richard Rosen in INET, “Carbon Taxes: a good idea, but can they be effective”.
It is no coincidence that Chris Goodall in his excellent book, “What we need to do now”, which focuses precisely on the challenge of electricity generation, makes no mention of externalities. His discussion of carbon taxes is also nuanced.