Nick Bostrom and Toby Ord, while discussing existential risks, advocate using a  "dynamic" notion of sustainability. This suggestion plays a role in a new open access paper that I just published with Jonathan Symons in Global Policy, where also Bostrom's 2013 paper on existential risks appeared. We are somewhat sceptical of the utility of appeals to sustainability. The title of our paper is "Operationalising sustainability? Why sustainability fails as an investment criterion for safeguarding the future."

To give an idea of our argument, here is a Twitter thread in which I summarize some of the main points:

---

New paper by @symons_jon and me, open access in Global Policy

We argue that any definition of sustainability will fail as an investment criterion for safeguarding the future.

---

Initiatives to improve the longer-run future of humanity and the environment often come with the banner of "sustainability."

The most prominent example may be the European Union Taxonomy for Sustainable Activities.

---
Aiming for "sustainability" is prima facie attractive. 

Intuitively:

• Predicting the future is notoriously difficult.
• But sustainable practices are good for the future.
• Identifying and adopting sustainable practices is comparatively straightforward.
• So, making the future go well becomes comparatively tractable by aiming for sustainability. 

---

But does it really work like that?

Let's start with the standard "indefiniteness" account of sustainability:

A practice is sustainable if and only if it can be performed indefinitely.

---

Applying this is not straightforward. And it the EU Taxonomy hasn't defined sustainability this way

Problem: It's not "operational" because "practice" is ambiguous.

For instance, is it sustainable to keep the types of trees in your forest the same, or to change species for a changing climate?

---

Choosing the right "practices" that deserve to be maintained requires detailed considerations about ways the future might go.

Choices among possible futures also involve contests over values.

No simple definition of 'sustainability' will permanently resolve these questions!

---
But once we have to consider values and possible futures anyway, why focus only on practices that can be performed indefinitely?

To get out of a crisis, actions that are intuitively unsustainable are often better or even needed.

---

Bostrom, in the same journal in 2013, suggested to define sustainability dynamically, as what reduces existential risk.

And he argued that we may sometimes have to use up finite resources to do that.

---

We agree with Bostrom -- though I personally think that "dynamical sustainability" is a misnomer.

The dramatic expansion of mining to build more renewable energy is a good example:.

It is unsustainable, by the indefiniteness account, but it improves humanity's future prospects.

---
 With respect to sustainable finance taxonomies our conclusion is:

They can either be based on operational criteria – but then policy outcomes will be uncertain;
Or they can be more loosely based on visions and expectations of desirable futures – and then there is no reason to expect that they will conserve natural capital or be capable of indefinite repetition.

---
 We conclude it might be better if investment taxonomies didn't pretend to provide comprehensive measures of "sustainability".

Instead, narrower measures of environmental impact - like life-cycle carbon emissions - will be a better aid to public debate

---

So as other political communities developed 'green' investment taxonomies, we suggest they might:

- focus on quantifiable measures of environmental impact, not comprehensive claims

- promote ongoing deliberation on environmental futures

- decline to impose their own visions of sustainability on communities elsewhere.