Limits to growth

The limits to growth are wrong. And AI proves it. 

I grew up reading books on systems. The Limits to Growth (the seminal Club of Rome report) was published in the early 70s just before I was born and I first read about it as a teenager. 

That there cannot be infinite growth on a finite planet would appear to be obvious - almost a tautology. Without terraforming Mars and interplanetary mining and trading corporations, growth based on resource extraction and consumption was going to reach physical limits. The Limits to Growth stood as an inconvenient mirror to the dominant strategy of almost all ruling political parties - that growth would ‘deliver’ benefits, hence it should be central to strategy.  

The recent wave of “Growth no matter the Cost” pronouncements from politicians (where GDP “trumps other concerns”) are a doubling-down on that thinking. The ultimate casino red-black strategy, doubling stakes every round in an attempt to win big; yet recognising at a deeper level that each spin of the wheel is sending the world deeper into the red.

 Later, I learned about ‘decoupling’ - that growth might become separated from the environmental damage of extraction, converting most of the economy into services. Of course there can be some near-term environmental and economic win-wins, but not everywhere. And as we move beyond environmental tipping points, our economic past can actively misdirect our understanding of the future. As brilliantly laid out by Duncan Austen in his backward-bending Kuznets Curve, we get to a point where economic growth turns to decline; where we are running faster to (not even) stay still financially. Our methods to stimulate growth cause more environmental (and then economic) harm, transgressing planetary limits as we slide into economic quicksand, falling further and faster the more we try to wriggle out of it.

AI: executioner and heir 

But now we have the ultimate accelerator: AI. Not only can AI play the role of dehumanised executioner as we sacrifice everything at the altar of growth, but it can be included in what counts as growth itself, before colonising the concept completely.

As Yuval Noah Harari has observed,  when you have incorporated AI (legal persons) who have their own wealth and can exert their own influence, it will be difficult to compete with them. Financial traders have already declared surrender to the AI trading bots. The proportion of prosperity that lies in human hands will dwindle fast. Want to predict how quickly AI will obliterate social capital? Look at how quickly humans have destroyed natural capital – two-thirds of wildlife has been lost since 1970 - and multiply that by the exponential rate of machine learning. 

As AI places greater value on the AI-to-AI services it delivers in the non-human economy, we might wonder why measuring GDP growth would be useful at all. But this has not stopped us in the past. Bobby Kennedy highlighted the flaws in his 1968 speech, noting that GDP counts: “air pollution and cigarette advertising, and ambulances to clear our highways of carnage. It counts special locks for our doors and the jails for the people who break them. It counts the destruction of the redwood and the loss of our natural wonder in chaotic sprawl.” 

We live in a world where GDP per capita has increased by 10x since 1970. And yet there are more that 25 million children and adults with no access to healthcare in the US; 60% of humanity is vulnerable to extreme weather events with no insurance globally; 1 in 4 children are at risk of poverty in Europe. So it’s clear something pretty fundamental about the type of growth we’ve had is not working. We’ve seen some decoupling, but mostly between GDP growth and wellbeing. If we squeeze the agriculture sector (which feeds people) to an almost negligible 4% of global GDP, what will happen when the growth-agents of the future don’t need food? 

Growth in a post-human economy

Ultimately, this leads us to a world which is enslaved to GDP growth as a means of paying past debts (incurred to fuel growth), forcing us to boost growth ‘at any cost’ which surrenders prosperity to AI (and their shrinking population of human owners). AI might well grow themselves to their ‘algorithmic-heart’s content’. In which case most humans would not matter any more and neither would the limits for human survival. Growth would continue. Most of us might not. I’ve had a few social media conversations recently about whether humans can pull the plug on AI. It’s a comforting mental image, bringing back the 1970s appliance, when we actually knew where the plug was. What we’ve seen accelerating over the past 12 months is the plug being pulled on human endeavour. It’s certainly highly likely that there will be new jobs created by AI. But it’s even more likely that these new jobs could be taken up more easily by AI. All that ‘free time’ being ‘opened up’? It’s more likely to lead to redundancy than leisure and space for creativity in a growth-at-all-cost society. So what would an economic trigger-point decision to ‘pull the plug on AI’ even look like? 

The actual urgent economic question of our time

In conclusion, the limits to growth are wrong. Infinite growth is possible on this finite earth; just not without humanity to experience it. If we choose a different goal, that means more than ditching growth. It means defining what post-growth looks like – something laid out by leading experts (including Kate Raworth, Tim Jackson and Jason Hickel recently). It means putting wellbeing (e.g. factors set out by the Wellbeing Economy Alliance) and the ‘qualities of growth’ contained in WEF’s recent Future of Growth report (Innovativeness; Inclusiveness; Sustainability; and Resilience) as primary drivers for economic policy, rather than skillfully curated afterthoughts.  

Mathematical solutions, as David Pitt Watson observed recently, don’t translate easily into economic thinking. The question The Limits to Growth posed might have a surprising solution: that infinite growth is possible in the future. Growth might lose its human meaning but that wasn’t the question being asked. 

For all of what AI might represent, it is not actually a mankind versus the machines problem. It’s a battle between those propelling AI-for-Growth-at-all-cost and those harnessing technology (or not) for positive transformation. The former appears to have a head start. Team humans, and all those designing for wellbeing and for the planet, need to achieve far more than some nice innovations. They need to change the goals of the game itself.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.