Skiing, EVs, and the Kaya Identity

Last week I went skiing with some friends in Italy. The snow was good, the weather great, and the apres even better. Whether winters like this will still exist in a few decades is an open question. I am also acutely aware that flying to the alps is not helping that fact!

After a big day on the slopes, a debate I had with a friend stuck with me. We were discussing climate change and green technologies. His position was simple: he was happy to adopt green technologies once they were sufficiently better and cheaper than dirty alternatives.

Electric vehicles were his example. In his view, EVs are already competitive with petrol cars. For many drivers they are arguably better: cheaper to run, quieter, and increasingly practical. His optimised model was two family cars, one EV for town and one petrol car for longer trips. Makes carbon sense.

But beyond EVs, he was sceptical. In other parts of the economy, areas like aviation and heavy industry, he felt the green alternatives were not yet good enough. Until they improve relative to their dirty alternatives, we should wait.

That conversation led me to think about a classic framework economists use to understand emissions: the Kaya Identity.

The Kaya Identity and the Economics of Decarbonisation

The Kaya identity decomposes global carbon emissions into four factors:

CO₂ emissions = Population x Standard of living x Energy intensity x Carbon intensity

In other words, we have four levers to reduce emissions:

1. Population
2. Standard of living (measured in GDP per capita)
3. Energy intensity of the economy (energy used per unit of GDP)
4. Carbon intensity of energy (emissions per unit of energy)

Historically, climate policy, particularly in economics, has focused on the last two terms, which is logical.

Essentially the idea is to focus on:

• Improving energy efficiency, reducing energy per unit of economic output.
• Decarbonising energy, replacing fossil fuels with renewables and low-carbon sources.

If those two ratios fall fast enough, then emissions can decline even as population and incomes continue to grow.

This approach has been politically attractive because it avoids difficult questions about living standards or population growth. The implicit promise has been that we can maintain or even increase our prosperity while decarbonising the economy.

There are other views such as doughnut economics and degrowth but I would say these (whilst becoming more popular) still operate on the periphery of mainstream economic debate.

But that chat with a mate got me thinking, what if the bulk of green technologies simply are too slow to reach parity with their fossil counterparts?

A Fossil-Subsidised World

We agreed, a major problem is that for over a century, the global economy has been built around fossil fuels.

Cheap, dense, and reliable energy has powered extraordinary improvements in living standards. Transport, heating, industrial production, and agriculture have all benefited from a world where energy was abundant and relatively inexpensive.

You could call this a fossil-subsidised world. Pick any number you want, but a widely quoted one is from the IEA estimates that subsidies for fossil fuel consumption exceed $1 trillion annually.

So we’re shooting ourselves in the foot - no debate there.

Green technologies are now trying to replace that system. Sometimes they succeed spectacularly. Solar and wind power have seen dramatic cost declines. Electric vehicles are increasingly competitive.

But in many sectors the gap remains large. Aviation, shipping, steel production, cement, and heavy transport still depend heavily on fossil fuels. The green alternatives are often more expensive, less efficient, or technologically immature.

There is some good content on this blog from the distinguished economist Lucien Vargas about dirty vs green growth - there are sensible ideas on how to fix this. Notwithstanding, if these technologies fail to catch up, the Kaya identity forces us to confront an uncomfortable possibility.

If we cannot reduce energy intensity or carbon intensity enough, then emissions can only fall if GDP per capita falls.

The Elephant in the Room

That raises a difficult question: should we reduce our standard of living or population to lower emissions?

My instinct is we should try much harder in reducing the energy intensity and carbon intensity dimensions as we rightly have. But answering the question properly requires confronting a deeper debate in climate economics.

It’s also essentially a version of the famous Stern–Nordhaus debate.

William Nordhaus (American, Yale, Nobel Prize, Big Dog) argues for a relatively gradual approach to climate policy, balancing the costs of mitigation against the benefits of avoided damages. Nicholas Stern (British, Oxford, Big Dog) argues for much more aggressive action, partly because of the possibility of catastrophic outcomes, low-probability but extremely high-cost climate tipping points.

This introduces a key issue in decision-making under uncertainty.

It’s also where my mind went whilst having this debate and raises the question of tipping points.

Tipping points are major, non-linear events that could occur with climate change. Think the Gulf Stream stopping sort of thing.

I’m an engineer, so let’s make a statistical thought experiment:

Suppose reducing my standard of living by 10% lowered the probability of triggering a collapse of the Gulf Stream 30%.

Would I do it?

The honest answer is, maybe, probably. In my friend’s opinion, definitely not.

However, for me I need more information because it depends on the baseline probability.

If the probability of Gulf Stream collapse falls from 10% to 7% (a 30% reduction as per the thought experiment), that might matter a great deal. If it falls from 0.00001% to 0.000007% (another 30% reduction), the effect is essentially negligible.

Without understanding the underlying probabilities and consequences, the question cannot be answered.

Also let’s be real, it’s a thought experiment, nobody knows the precise probabilities of the Gulf Stream collapsing. Unless Laplace’s demon exists, then economists also have no chance of accurately defining this.

And this is inherently the problem! Much of the climate debate implicitly assumes we already know the answer.

No one can tell me if I stop eating red meat, or stop flying, what precise impact this will have on Gulf Stream collapse or weather in Mexico in 2049. It stops becoming a logical conversation and has to be an emotional one.

When Economics Isn’t Enough

So at this point the problem stops being purely economic.

We estimate the Social Cost of Carbonl, we model emissions pathways, we estimate technology costs, we simulate energy systems. But deciding how much present consumption we should sacrifice to reduce uncertain future risks is not merely a technical problem, it is metaphysical.

If you ask a physicist the boiling point of water, they can give you a precise answer. It is an empirical question. But I don’t know the answer to how much consumption we should sacrifice - if we should at all - today to reduce risks tomorrow, because this is not something we can measure.

To me, we need to knock on the doors of some very smart philosophers and ask them.

How should we value the future relative to the present? What obligations do we have to future generations?

These questions sit at the intersection of ethics, philosophy, and political theory. Not economics.

Ironically, in an age increasingly dominated by artificial intelligence, quantitative modelling, and engineering solutions, the climate problem may ultimately force us back to the humanities.

Not because economics and engineering are unimportant, they are essential (I have a degree in both so I am biased), but because they cannot answer the deepest questions.

What level of risk to the planet are we morally willing to accept in order to maintain our current standard of living?

The Ministry of Philosophy

Governments should recognise that the deepest questions in climate policy are not economic or technological but ethical.

Before economists construct models or engineers design decarbonisation pathways, societies must first decide how much risk to future generations they are willing to tolerate and how present costs should be weighed against uncertain future harms.

Governments should convene diverse panels of eminent philosophers and ethicists to help articulate these principles, clarifying society’s stance on intergenerational responsibility, catastrophic risk, and acceptable trade-offs in living standards.