At one point Morgan finds himself imprisoned by the suspicious denizens of Arthurian Britain. To escape he tells them that he is a powerful wizard, and that if they don’t release him he will use his terrible magic to black out the sun.

What Morgan knows, but his captors do not, is that his imprisonment coincides with the date of an historic solar eclipse. Sure enough, when the sun goes dark, Morgan takes the credit and is duly released.

Hank Morgan’s Eclipse.

Some years ago, someone who worked at a major national charity once told me that their head of policy had become aware that the government was about to announce a significant change in housing policy.

The organisation had done work on the topic in question some years previously (under different leadership). So in anticipation of this announcement they issued a press release based on this then five years old work, and called on the government to do the thing they were, in fact, already planning to do.

When the government announced the policy change, the charity’s leadership had successfully positioned themselves to take the credit. Much self-congratulation ensued, followed by a collective amnesia around the wizardry that had delivered this impact.

Hank Morgan’s Eclipse.

The above is just a particularly egregious and extreme example of a policy influencing trap that we should all avoid falling into. How content are we to take credit for things which are quite likely to happen anyway?

In William MacAskill’s book on long termism ‘What We Owe The Future’ he talks about ‘contingency’. The book is about how people seeking to improve the world can decide what to prioritise to maximise their impact — the ‘effective altruism’ movement.

Alongside ‘significance’ and ‘persistence’ (basically how much impact and how long it lasts), ‘contingency’ invites us to consider how things might have turned out without our actions. How much is down to us? MacAskill argues this third aspect of value is often neglected. (more here)

What contingency will mean to people working on net zero policy will very much depend on your answer to the question ‘if we did nothing new, would we get to net zero?’.

To me the answer to this question is clearly ‘no’. That isn’t to say we are definitely going to fail on net zero, rather that current trends all point to failure. Global carbon emissions are going up, long past the point they should have started to go down. This is ultimately the only metric that matters.

Most ‘carbon budget’ type pathways which chart pathways to net zero include significant contributions from things which are not yet happening, things which do not exist, and a few things which seem unlikely to ever exist. [we’ll return to the foundational intellectual errors of carbon budgeting another time]

Across the political spectrum, from the hardest green lefty to sincere net zero advocates on the neoliberal right, what I see is a consistent complacency on the contingency of net zero. Success is almost taken as a given — we will definitely get to net zero, so the question is how.

On the left this tends to present as concerns around who profits: We should have locally owned community energy which puts money back into communities.

On the right the framing tends to be around cost: We should find the most efficient and low cost route to net zero.

Ultimately as someone with no actual opinions, I have a lot of sympathy with both views. However both approaches essentially risk moving too slowly. We will not know if and when we have gone past the tipping point where we can retain a stable climate which supports human society. It’s not impossible this has already happened.

In the world of ‘policy wins’, the proxy outcomes of constructing a local wind farm, tweaking price signals, or setting new carbon targets are all highly tempting goals to set. They are all, it’s worth saying, good and real achievements which make a contribution to the macro goal. But they are also proxy outcomes — how they play out is what really matters.

They all also benefit from high attribution (getting credit) — it’s easier to know you achieved something if it’s either a fairly modest win, or an absolutely massive transformation. If you do something medium sized it’s likely you’ll need to share credit with lots of other people working in the same space, or with the natural movement of the markets etc.

If we try and show this on a graph we get what we might call ‘attribution spaghetti’.

As per the previous post, none of this is intended to cynically imply pure self-interest as the primary motivation of policy professionals. Not everyone is behaving as egregiously as in the example above, it’s just easy to be distracted by pats on the head.

There’s a fine line between a highly contingent outcome and a highly attributable one. If we care about genuinely having an impact, these two things should ideally be perfectly aligned. We no more want to assume we’ve succeeded when we haven’t, than we want to take credit we don’t deserve.

The more modest and incremental a tweak, the less likely it is to deliver the game changing impact we really need to achieve net zero. At present there is no default success scenario. Things will not be fine if we do nothing.

So actions which contribute to making net zero align better with our other values and objectives (cost, fairness etc), or which fall into the window of political acceptability risk distracting us from the absolute core mission of reducing the amount of carbon we emit to a safe level.

The sun will continue to rise long after we’re there to see it.

Hank Morgan’s Eclipse was originally published in Doing Things Is Hard on Medium, where people are continuing the conversation by highlighting and responding to this story.

It’s likely to take some fairly oblique routes to that topic. I’m almost inclined to call it an anti-blog.

There are two things that are going to guide how I approach writing this.

1. No single post can be longer than 1,000 words. Ever*.
2. Each post must be accompanied by a crappy drawing to undermine whatever point is being made.

Point 1 will inevitably mean that some of my longer chains of reasoning aren’t presented in full. On balance this is fine — for as long as I can be bothered to do so I will try to gradually go back and add links between posts to fill in the gaps. This may or may not work.

The very first post on this page is / was called ‘You’re confident because you’re wrong’. Unlike most of my opinions, this is a view I hold very tightly. However, the desire for brevity means I’m going to avoid lengthy caveats. So it’s likely I’ll generally come across as being totally confident in what I’m saying.

I’m not.

Comments and disagreements are welcomed. This is me thinking aloud — not a manifesto.

This blog is intended to land in a positive place. But many of the posts will constitute a taxonomy of all the ways people trying to make positive change can go wrong. I think we need to stare these errors in the face in order to address them.

This might mean a lot of the blog comes across as negative. That’s not the intention. I believe there is a universe of solutions and positive ways forward.

The runner up for the title of this blog was “A Slow Mess”. Slow mess is the essence of how we can make progress.

Another big theme will be humility — personal and intellectual. I’ll probably talk about incentives a lot, in a way that could easily read as being ‘you are only doing X for your own benefit’. This is almost never what I think about any specific person. I simply believe our incentives can lead us astray, and we should try and notice this happening.

I’ll get into all this sooner or later.

But why am I writing this?

Reason one is to get these things out of my head. I have a fairly complex relationship with my own mental health, and whilst I manage this fairly well most of the time — sometimes I don’t. A little too much of what troubles me stays inside my own head.

So this blog is at least in part an attempt to get what’s inside out — have a dialogue with others, maybe change some minds (including my own), and move more into positivity.

Another is that because when I talk about ‘doing policy better’ I am almost invariably talking about net zero — a policy area where failure is not an option.

You may spend your time — for example — campaigning to keep music venues open. This is something which is important to me, and I wish you every success. I would be sad if more venues closed, but it wouldn’t be the end of the world.

Failure on net zero would — in certain respects — be ‘the end of the world’ [nuancing this will be another blog], and that’s why I care about it. In most other policy domains you get a do-over every generation. You can always put right what goes wrong. Not with net zero.

As it stands right now, I wouldn’t predict that humankind will get to net zero. However I absolutely believe we could, if we chose to. We need to try harder, think differently, put the future first, and embrace the slow mess.

To do this we need to be clear eyed about how what we’re currently attempting is falling short. We don’t get to mark our own homework on climate change.

Why the title?

One reason is that I don’t think we’ll make progress by oversimplifying, pretending hard things are easy, or taking easy ways forward. In most spaces, achieving something is better than nothing, the perfect can be the enemy of the good.

There’s a prevailing view that you can always build on the successes of what went before and make slow, incremental progress towards very big goals. This may or may not be true in net zero, and we should weigh risk accordingly.

The main reason for the title is that I say ‘Doing Things Is Hard’ a lot, and people seem to like it. I say it to let people know it’s okay to struggle. To (hopefully) signal that their struggle is about the problem — not them.

Doing things is hard. But we can do hard things.

So this blog is an attempt at dialogue, and a medium for a little self care. And at my most optimistic (read: egotistical) — if I’m even 25% correct about how we can do better — then maybe someone reading will find they can do a few more hard things.

*ever ever ever

Why even start writing a blog? was originally published in Doing Things Is Hard on Medium, where people are continuing the conversation by highlighting and responding to this story.

Think of something you know lots about. Ideally something you’ve spent several years learning about, which you have strong opinions on, and where other people have equally strong opinions that you disagree with.

How have your opinions changed since before you really began to learn about this topic? Why?

Presumably the answer is ‘because I now know more about it’. Your Day One opinion will be based on hunches, priors, values, and whatever unconnected nibbles of information you’ve absorbed indirectly. Understandable.

Now, years later when you look back, you should not only have a better sense of what is ‘true’ about this topic, but a sense of how much you used not to know and how much more you still don’t know.

Reflecting on this leap forward in your knowledge, how should you respond? The answer isn’t quite ‘less confidence’ but ‘greater humility’. You now know lots about X. You know how much more you know about X than people who know nothing about X. [acknowledgement that this is basically the Dunning-Kruger effect]

But is X all that matters?

How much do you not know about the other things that matter? What might you be overlooking?

The only sensible conclusion is that the more you know, generally, the more humble you should be about everything you don’t.

I’ve met plenty of people who find this conclusion intolerable. Some people would rather have false certainty than incomplete, messy truth. This is understandable, we’d all like to be omniscient.

However where I lose sympathy is when someone suggests that embracing (or as some might put it ‘wallowing in’) this uncertainty is an enemy of progress. This simply isn’t true.

There is an instinct amongst certainty seekers to boil things down into numbers. To create models. [We’ll come back to ‘all models are wrong but some are useful’ another time — though it’s hardly a novel or controversial sentiment].

Psychologically, what purpose does a model serve? It helps people feel that they understand something, and therefore bring confidence.

But the only way to create a model is to omit most information, and to make tradeoffs between reality and the desire to have delivered a model. Everyone who makes models knows this, and they know it far better than I personally do.

There is no level on which I am saying ‘don’t ever make a model’ or ‘models aren’t useful’. What I am saying is that when using a model we should give due weight to the absent information, and remember that the model isn’t the reality.

Models exist for the people who need to understand something from which they are remote. Maps exist first for the mapmaker, second for the visitor, and last for the locals.

Such a map is fine if you want to find the zoo. But sometimes as ‘policy people’ we try to use a ‘tourist’ map to try and make plans for places we’ve never been.

Say we’ve been tasked with planting some trees in British towns to capture some carbon emissions. Where should we put them? We can’t visit every town, review every site, conduct focus groups with citizens up and down the land. It would be inefficient — we need to deliver the tree planting strategy.

So we consult a map. A model. Seeing a patch of land rendered uniformly in green we might say “This looks like a nice, open space. We’ll put the trees here.”

How confident should we be that this patch of green is in fact a uniform land mass where planting trees won’t be difficult, disruptive or have unintended consequences?

If we did go and do a site visit, and left with a greater knowledge of the topology of that place — would our confidence of how easy it would be to plant trees there be likely to increase or decrease?

Remember that the analysis here isn’t about whether a given decision is good or bad, it’s about where confidence comes from. We might come away from a site visit thinking “okay, this is going to be harder than I thought — but we should still do it”. Less confident perhaps, but better informed.

We would then know our prior confidence that the uniform green patch on the map was a fair reflection of the landscape was misplaced. We were confident because we were wrong.

The most confident people I’ve encountered in my seventeen years in and around policy, tend to be people who embrace the most simplistic models. The most pernicious being market fundamentalism, and the belief in the price point. [We’ll come back to Hayek another time too.]

The desire for certainty often goes hand in hand with a desire for pace, for impact, for change, for wanting to make a difference. I want these things too — most people in ‘policy land’ do. However we underweight for the sheer volume of what we don’t know.

We underweight for the fact that our desire to assess a whole landscape and render it comprehensible from our remote standpoint requires a decision to be wrong about lots of things. It requires us to overlook most information.

When the abstractions of policy and models meet the lumpiness of reality, that’s where things that are good ideas on paper don’t survive contact with the actually existing world. This is why ‘delivery’ so often fails.

When I see people speaking with our confidence I assume it’s because they either don’t understand this, don’t care about it, or have — maybe just for a moment — forgotten to hold it front of mind.

At worst, there are some who act like they believe the missing ingredient that will make the difference between success and failure is them — the confident person.

In such moments, whatever fine personal qualities such individuals invariably have, they are confident because they are wrong.

You’re confident because you’re wrong was originally published in Doing Things Is Hard on Medium, where people are continuing the conversation by highlighting and responding to this story.