My Introduction To Systems Change

Incentives, Externalities, Multi-Polar Traps, and a Third Attractor

The following is the largely unedited transcript of a presentation I gave to the SF Climate Action Club on May 14^th^* 2025. Whilst my understanding of our predicament has evolved since then, this transcript serves as a valuable baseline to introduce a few important concepts that continue to shape my thinking, and so I decided to post this to Substack as-is. I plan to build on these concepts in future blog posts. This presentation was heavily inspired by this insightful podcast episode with Daniel Schmachtenberger and Nate Hagens, and this paper on Development in Progress.*

I want to touch upon a few different concepts that begin to explain the ways that we seem to be failing on the climate crisis. I don’t claim that these are the entire answer, but these are ways of thinking that may guide us in the right direction.

Concept 1: Jevons’ Paradox

History of the Climate Movement

I want to start with a little bit of the history of the climate and environmental movements.

130 years ago, in 1896, Svante Arrhenius, a Swedish scientist, made the first prediction of global warming.
About 60 years ago, in 1962, Rachel Carson published Silent Spring, a breakthrough book that kicked off the modern environmental movement.
In the 1960s and 70s, multiple scientists, including scientists at oil companies, published findings about climate change.
Nearly 20 years ago, in 2006, Al Gore, the ex-Vice President, published An Inconvenient Truth, a pathbreaking documentary about climate change.

This image shows a history of the COPs, which are the intergovernmental panels on climate change, where the world’s leaders come together to address how to tackle this problem. It started in 1992 in Rio.

Improved Efficiency

Given that we’ve known about this problem for such a long time, what have we done to address this?

For one, we’ve made amazing technological progress in renewable energy. Specifically in solar energy and in battery technology, especially over the last decade or so, we’ve seen an amazing increase in efficiency accompanied by a drastic decrease in price. That’s what you’re seeing in this graph.

It’s not just renewable energy that’s gotten better. Our vehicles have gotten drastically more efficient too.

So have our washing machines, our heaters, our air conditioners, and our dishwashers.

And this is just a small slice. Our planes have gotten more efficient, our ships have gotten more efficient, our data centers have gotten more efficient. Nearly anything you think of, it’s probably gotten more efficient.

Energy Consumption Keeps Going Up

You would hope that with all this amazing technology that’s pushing the boundaries of efficiency across every industry, we would use less energy, less fossil fuels, right?

You would be mistaken.

Here is a graph of global energy consumption. You can see it’s always going up (except for those two recession years in 2008 and 2020).

Here’s the same graph, but with only fossil fuels. You can see that it’s still going up.

You might think this is global, and that this is because of third world countries that burn dirty coal, and we’re the clean green ones over here in the US. But no. Here are some graphs from the US Energy Information Administration.

And just to highlight the point, here’s the same COP timeline from before, but overlaid on the PPM of CO2 in our atmosphere.

That’s a pretty depressing picture. And it’s hard to see it on this graph, but the PPM of CO2 is actually accelerating in the second half of this graph, up from 1.8 ppm increase per year to about 2.6 right now.

What gives? What’s going on here? We’re making all of our stuff more efficient, we’re making renewables more efficient, cars more efficient, fridges more efficient, but why aren’t we emitting less CO2?

The Paradox of Efficiency

What we’re seeing is the paradox of efficiency. This is Jevon’s paradox.

William Stanley Jevons was an English economist in the 1800s. He was alive around the time when steam engines were being developed. The steam engine ran on coal, and he was perplexed by the paradox he was seeing: every year, people were coming up with better, more efficient steam engines, yet the total demand for coal was going up, not down.

This happens because, as Jevons explained, the increased efficiency lowers the effective cost of using the resource, making it more attractive and affordable, which in turn stimulates higher demand (under an unbounded profit-driven economy) and ultimately greater overall resource consumption. While individual engines might use less coal, the total number of engines or their increased usage due to lower operating costs results in a net increase in coal consumption.

Let’s take a modern day example.

Let’s say I’m the owner of a trucking company and I have trucks that get me 5 miles per gallon. Since my fuel costs are so expensive, I have to be conservative about which routes I take, and I don’t do too many trips. But then I upgrade to trucks that give me 10 miles per gallon. Suddenly, I can actually drive much longer routes, because my gas costs aren’t as high. So I start driving 3 times the number of routes I used to, and now I actually end up using more fuel overall. I may even buy an extra truck with all the new profits I’m making, and drive around even more!

When we make more efficient ships and planes and trucks, we don’t save energy or fossil fuels with shipping. We just send more ships around the world, because global trade becomes cheaper and we get stuff like Amazon Prime’s one-day shipping and Temu.

Our more efficient refrigerator technology doesn’t make us use fewer fridges; it leads to more of them. It just means that now I can put an extra mini-fridge in my upstairs bedroom, where it wouldn’t have made sense earlier. Or it means that somebody in a cold country that didn’t really feel the need for a fridge before now buys one anyway because it’s so cheap.

This is actually happening for renewables too. It’s harder to see with renewables, but paradoxically, the increased efficiency of renewables (i.e. energy production)

means that there’s an increase in energy demand overall, which means that renewable energy can’t actually substitute fossil fuels entirely, it’s just feeding its own induced demand, which is why fossil fuel consumption continues to go up. (editors note: the exact extent to which renewables replace fossil fuels has been hotly debated, and this explanation is certainly an oversimplification, but the core concept of the paradoxical efficiency trap still holds)

We also see this lately with AI and data centers. People are working on better hardware and software models, and every individual data center is becoming more and more efficient for training these LLMs. But that just means it’s cheaper to train LLMs, so tech companies just go ahead and train even more and even bigger LLMs. None of the efficiency leads to less electricity consumption---it actually leads to more energy consumption.

This is the first concept. Jevon’s paradox is why you should be a bit more sceptical about pro-climate work that “improves efficiency”. Improved efficiency is good for a lot of different reasons: but addressing climate change is not one of them, at least not under our current system, unless we actually solve Jevons’ paradox too.

Let me be clear: I absolutely value all the work being done in renewable energy. I think we need a complete transition, I think we need better batteries, better solar energy, and better wind energy. But it’s also clear to me that those changes alone will not solve our predicament: in fact, they may even make our predicament worse, unless we solve Jevons’ paradox.

Concept 2: Perverse Incentives and Externalities

What is a perverse incentive?

A perverse incentive is an incentive structure that’s normally well-intentioned, or at least logical from a narrow perspective, but ends up creating deeply undesirable or unexpected outcomes. It’s when the rules of the game unintentionally encourage players to act in ways that harm the greater good---especially the commons.

Examples

Leaded Gasoline

In the 1920s, engine knocking was a major problem. Knocking occurs when there are unexpected combustions in the piston of an engine at the wrong time. These combustions cause damage and make engines less efficient. Scientists discovered that they could add a compound called tetraethyl-lead that prevented knocking and made cars run better---and companies realized that they could make a lot of money selling this chemical. So they did.

As a result of putting lead in the gasoline that all our cars burned on all our roads and cities, we blanketed the globe in a potent neurotoxin. Studies suggest this diffuse lead exposure has robbed billions of IQ points worldwide and has made some populations up to 4 times more prone to aggression before the lead was phased out. Here’s the thing though: all that lead is still in our environment, still potentially making us dumber and more violent today, and it takes a really, really long time to go away. Imagine the number of wars that have been fought, the number of police aggression incidents, the number of school bullies, that wouldn’t have happened if it wasn’t for the lead.

Here’s a well-produced Veritasium video about the man who added lead to Gasoline

DDT

We saw this with DDT. Hailed as a miracle pesticide after World War II, DDT promised disease control and bumper crops. The incentive was for chemical manufacturers to produce and sell this chemical that effectively and cheaply eliminate pests. And then in 1962, Rachel Carson wrote Silent Spring, which revealed that DDT was ravaging ecosystems, causing bird populations to plummet due to thinning eggshells, accumulating dangerously up the food chain, and potentially even causing cancer in humans. It took us decades to finally ban it, during which time we continued to spray it on many of the crops that end up on our dinner tables.

More examples

Cigarette companies had advertisements that said ‘More Doctors Smoke Camels’, when they knew internally about the disastrous health effects of their products. Oil companies knew about climate change well before it was in the public domain, and they chose to play misinformation games and lie to the public, as Naomi Oreskes and Erik Conway have documented in Merchants Of Doubt.

Forever Chemicals

This is the one that really gets me, because it’s so recent. It’s clear we haven’t learned our lesson. You have probably heard about PFAS in the news. Companies like DuPont developed chemicals like PFOA, used in Teflon, for stuff like waterproof jackets and non-stick pans. They made billions of dollars in profit.

Yet, these companies knew about the environmental persistence and potential health risks---links to cancers, thyroid disease---for decades, possibly since the 1960s or 70s, long before the public became aware or regulators stepped in. They still haven’t been jailed for it. Mark Ruffalo starred in Dark Waters, a great movie about a lawyer that tries to fight against these mega-corporations. Veritasium has a great video about the science here too.

And now you have big-tech and social media. These companies---Facebook and Twitter in particular---knew about the risks of their products. This was first revealed most prominently in the pathbreaking documentary The Social Dilemma. I was most stunned by reading The Chaos Machine, in which Max Fisher writes about how these products have been proven to be responsible for multiple violent uprisings across the world: the Rohingya genocide in Myanmar, the anti-Muslim violence in Sri Lanka, as well as the far-right shift across the world, including the election of Donald Trump.

Explanation

Little Eichmanns

Are all these people---who work at these companies that do these things---bad people?

Well, for some of them, yes. If you know you have a literal poison and you give it to people anyway, that’s a bad thing to do, and it’s certainly true that some of these executives are bad people with little remorse for their actions. But I think the system that we have predisposes good people to do bad things: and that’s what a perverse incentive is.

No matter how evil the head of an oil company is personally, he can’t do anything by himself. He needs a team, of managers, of engineers, of oil rig workers, of marketing folks. He needs thousands, even tens of thousands of people.

Are all of those people evil? No. They’re not.

They were just doing their jobs. They just needed to make money to put food on the table. They had kids, they went on vacations, they probably enjoyed baseball games with their neighbors, some of them probably even coached little leagues or volunteered at non-profits. But the sum total of their impacts has been disastrous for society.

I have friends that work at companies that aren’t necessarily doing the best for the world. I myself work for a tech company that has received a lot of criticism. And when I got this high-paying job at this tech company, or when my friends landed high—paying jobs at well-known companies, society actually celebrates us. (*editors note: I no longer work there, please see *my other post)

When I look at the people around me, the people on my team, none of them seem like “bad” people. There’s just a tendency to not see the bigger picture, because seeing the bigger picture doesn’t make you good at your job. Seeing the bigger picture and asking questions about the bigger picture isn’t what gets you a promotion---it may even be what gets you fired.

There’s this term called “Little Eichmanns”. It’s when people that are individually “good” come together to do bad things. And so we need to ask: what are the systems, and what are the forces that cause good people to do bad things? When is there a situation, where what I’m doing for myself, which is seemingly the obvious thing to do, is not good for society at large?

As long as we have perverse incentives in society, we are going to struggle with managing the commons.

Externalities

But it’s not always about bad people, or even good people in bad systems. Sometimes it’s even about good people asking the wrong questions: or rather, “incomplete” questions.

From the point of view of “was this a great solution to the problem of engine knocking”, lead was the right answer. From the point of view of “did this pesticide kill pests effectively”, DDT was the right answer. These aren’t the “wrong” solutions, they just happen to be the “right” solutions to the “wrong” questions. And this delves into our issue with problem solving.

The world is a complex place. There are multiple systems, they’re interrelated, and they all interact with each other. When we try to solve a problem, our general approach is to break that problem down into bite-size parts. We hold a bunch of variables constant, make assumptions about the world, build small models, and then extrapolate those small models out. And in many situations, this approach works! With this approach we put a man on the moon, we built amazing nuclear power, we came up with vaccines for pandemics, and we built Formula 1 cars that go really fast. Human engineering and ingenuity is something to be proud of.

But our problem solving approach has a blind-spot, and that’s externalities. All of the examples I mentioned in the previous section had externalities: the externality of diffuse lead in the environment, or the externality of CO2 emissions in the atmosphere, or the externality of the destruction of the world’s attention spans because we tried to maximize engagement.

Climate change is one of these complex problems. You may think climate change is about CO2 emissions, and you’re not wrong: you’re just incomplete.

We may come up with geo-engineering solutions that use particulate matter to block sun rays and cool the earth, but what about the externality of acid rain, or messed up monsoons? We may come up with better EVs and battery tech, but what about the externality of mining on some of the only remaining biodiverse ecosystems on the planet, and the human rights violations of those local communities, often kids, in those mines? We may come up with carbon capture solutions, but what about the energy required to keep those running, and the nuclear power plants that we build for those that then become geopolitical targets in war?

It’s never possible to predict and plan for all the externalities: but that doesn’t mean we don’t try. We can do a much better job than we’re doing now, both individually and collectively. And that’s something we have to do. Jevons’ paradox was an externality of building better renewable energy, and so if we actually wanted to reduce carbon emissions, that’s something we should have thought about.

Concept 3: Multi-Polar Traps

But we can even go one level deeper in this analysis and ask, why do these perverse incentives and ignored externalities persist? Why is it so hard to just… fix them?

This brings us to the concept of multi-polar traps.

Some of you have probably heard of the prisoner’s dilemma, a common concept in game theory. It’s this idea that what’s best for the individual is not best for the whole. Imagine you have two prisoners, A and B. They’ve both been charged with a crime, and they’re being interrogated in separate rooms. The interrogators are crafty, so they offer a deal, the summary of which is in this diagram.

If both remain silent (cooperate), they each serve 1 year on a lesser charge. If one betrays the other (defects), the betrayer goes free while the silent one serves 3 years. If both betray each other, they each serve 2 years. This setup creates a paradox: while mutual cooperation yields the best collective outcome (2 total years in prison), each individual has a strong incentive to defect. For each prisoner, defecting and ratting out their partner seems like the “rational” choice. But the socially optimal situation is for both of them to cooperate.

Veritasium’s video on game theory is great here as well.

The multi-polar trap is analogous to the prisoner’s dilemma, but on a larger scale, and with many different players. Imagine a situation where you have multiple independent actors all pursuing their own rational self-interest. The trap occurs when this individual pursuit of self-interest inevitably leads to a collective outcome that is detrimental to everyone involved, often including themselves in the long run.

The insidious part is that once you’re in the trap, it’s incredibly difficult, perhaps even impossible, for any single actor to change their behavior unilaterally.

If one company decides to stop polluting completely when its competitors don’t, it might go bankrupt. If one nation disarms while others build up weapons, it risks invasion. Defecting from the harmful pattern is punished by the system dynamics itself. What’s rational for one, becomes irrational or catastrophic if everyone does it, yet no single actor can afford not to do it if others continue.

These traps essentially generate the perverse incentives we talked about. They create situations where ignoring externalities isn’t just lazy, it becomes a competitive necessity.

Climate Change and Geopolitics

Let’s look at climate change through this lens. Historically, there’s been a very tight correlation between GDP growth, energy use, and carbon emissions. Nations are fundamentally incentivized to grow their economies — for stability, geopolitical power, and greater material comforts. In a globalized world, they are in constant competition. Because growth has been so tightly coupled with energy (and thus, emissions), any nation that unilaterally decided to drastically cut its emissions risked severe economic disadvantage compared to its competitors. And an economic disadvantage also means a military disadvantage, which also means a security and geopolitical disadvantage. Neither the US nor China wants climate change to happen, but if the US cuts emissions first, they will lose to China. If China cuts emissions first, they will lose to the US. So they’re both forced to keep emitting even though they don’t want to. It’s a collective action problem on a planetary scale.

This is a classic multi-polar trap, explaining why decades of climate conferences (those COPs we saw) struggle to produce binding actions that meet the scale of the crisis. Everyone is trapped by the perceived need to prioritize short-term economic competitiveness.

AI

We see a similar dynamic emerging right now with Artificial Intelligence, particularly concerning the race for Artificial General Intelligence or AGI. The potential economic and military power conferred by superior AI is immense. This creates an incredibly powerful incentive for companies and nations to develop AI as fast as possible---the fear of being left behind is enormous. There’s a potential first-mover advantage; the first entity to develop truly transformative AI might gain an insurmountable lead, potentially dominating the global landscape.

Here’s the trap: Even if developers, CEOs, and governments recognize the profound existential risks associated with rushing the development of potentially uncontrollable AI, can any single entity afford to pause? Can they implement costly, time-consuming safety protocols and alignment research if they believe their competitors (other companies or nations) are racing ahead unscrupulously? The rational incentive for each player is to push forward as fast as possible, hoping to win the race, even if that collective race leads us towards a potentially catastrophic outcome for humanity. The competition itself forces risk-taking.

Working A Job can be a Multi-Polar Trap

When we look at the ‘Little Eichmanns’ problem---good people participating in systems that cause harm---multi-polar traps provide a deeper explanation. The competitive pressures often leave individuals feeling they have no choice but to play the game according to the existing rules, even if they see the collective harm. Questioning the bigger picture, as we discussed, can be detrimental to one’s career or a company’s survival within these competitive dynamics.

Understanding multi-polar traps reveals the limitations of purely individual or even national solutions to problems like climate change or AI safety. These traps highlight that the structure of our global competition itself generates destructive outcomes. Escaping them likely requires unprecedented levels of coordination, trust-building, verification mechanisms, and potentially fundamental changes to the rules of the game---moving from zero-sum competition to collaborative, positive-sum frameworks. Recognizing these traps doesn’t make the problems easier, but it clarifies the profound depth of the challenge we face.

Concept 4: The Third Attractor

The Two Attractors We Have Today

So if we do nothing, we don’t solve these multi-polar traps, and we just keep competing with each other until we mess everything up---quite possibly leading to human-extinction level consequences.

On the flip side, people are very scared of the scary ideology words that cannot be said---but I’ll say them anyway---“communism” and “socialism”. A lot of this fear is irrational: there’s been decades of documented propaganda from the US government to make people scared of these concepts. But some of the fear is rational. It is true that some past experiments with centralization of power have led to that power being used corruptly and for personal gain, at the expense of people.

China, for example, has strict limits on how much kids can play video games, and they have a highly regulated and vetted TikTok. If that were to happen here, we would call it a massive violation of free speech, and we’d be up in arms about it. But that system does work: the kids there are much less addicted to video games and to scrolling on their phones. We can and should discuss the pros and cons of both approaches, but that’s not the point here: the point is to demonstrate that these are both sides of the same coin.

If we go too far with decentralization, we end up with more multi-polar traps that destroy the commons. If we go too far with centralization, we end up with corruption, bureaucracy, and a few people have a lot of control with police and surveillance. It’s easy to see that neither of these approaches is going to carry us through in a way that allows us to preserve our humanity.

A Third Attractor

This is why we need a “Third Attractor”. A third attractor is a fundamentally new way of looking at human relationships and activities. It’s a way of binding multi-polar traps without resorting to centralized power. It’s a way of recognizing that technology has the power to reshape society, but we need to steward that technology with wisdom. It’s recognizing that our current paradigm of infinite growth on a finite planet cannot work, so we need something else that can. It’s recognizing that we can’t solve these problems alone, and that we need to act in community.

Governing The Commons

Elinor Ostrom, the Nobel-winning economist, has done foundational work on Governing The Commons, and trying to solve multi-polar traps in fundamentally new ways. I’m just starting to delve into her work, but it’s reassuring to know that people have thought about this.

Digital Democracy

Audrey Tang, the digital minister of Taiwan, has done great work with g0v on using technology and AI to facilitate better discourse and encourage commonality and consensus instead of polarization. There’s a great podcast episode with her and Tristan Harris here.

New Economic Paradigms

Our obsession with growth has put us in situations where Jevons’ paradox prevents us from making any headway. It’s not just about carbon emissions---we’ve crossed 6 out of the 9 planetary boundaries that Johan Rockstrom and the team at the Potsdam Institute have defined. It’s clear that more growth cannot get us out of this mess, so what will?

There’s been multiple calls for Degrowth and Circular Economies, that are built on a set of principles of sustainability, circularity, local production, inclusion, sharing, and slowing down. I’ll call out some thinkers and concepts that have shaped my views on this most profoundly: Jason Hickel’s ideas on Degrowth, Naomi Klein’s analysis of our system, and Kate Raworth’s conception of Doughnut Economics.

An Entirely New Way Of Thinking

So much of the way we think about the issues of the world is deeply shaped by conditioning. There’s a deep fundamental shift in ourselves, in our relationship with each other, and in our relationship with nature, that has to happen if we’re to make it through this next stage. There’s philosophical, and even spiritual work, that has to be done individually and collectively. There are many philosophers that have thought about what this means, and I’m still going through my own journey here, but I will call out just a few authors and thinkers that have shaped my thinking: Daniel Schmachtenberger, Vanessa Andreotti, and Jiddu Krishnamurti.

Conclusion

So we’ve seen how efficiency can backfire, how incentives misdirect us, and how competition traps us. This clearly shows the limitations of our current approaches. There are no obvious, easy answers here. The way that human civilization has existed till this point is unsustainable, and it’s coming into its limits. But there’s a lot of smart and caring people around the world that have recognized that, and are working on finding a way out. My hope with this presentation is that you will join those fighting for our future, and help us find a Third Attractor. Thank you.