Coal, the Manski bounds, and the correct choice of Milibands

Is there any point trying to make consumers save the world? One of the most important debates of our time is whether we should approach climate change as a primarily individualistic or collective problem. Individualistic approaches include things like cap-and-trade at the consumer level, carbon taxes imposed on the final consumer (like VAT), shouting at people to adopt this or that diet. It’s probably not escaped anyone’s notice that there’s a lot of this about. The collective approach is more about infrastructure, regulation, and either cap-and-trade or taxes imposed upstream, in the energy sector.

This distinction cuts across a wide range of other political divides – the collective approach is supported both by technocrats and by Green New Deal socialists, while the individualistic one, ironically, is supported by some really radical voices and also by libertarian economists and lame centrists. This cross-cutting suggests it’s probably interesting, although guess which one I prefer?

The issue I have always wondered about here is essentially how much information individuals actually get from prices. Energy is not a huge percentage of the price of most goods other than energy itself, so even a really swingeing carbon tax, or a comprehensive consumer cap-and-trade system, or universal sainthood, might not have all that much impact. For all this to work, it has to be true that the price of more carbonic goods increases relative to less carbonic ones, by enough to meaningfully influence demand, and that the change is perceived by their manufacturers as being something to do with their choices about the production process. It wouldn’t help, for example, if the manufacturer thought the drop in demand was something to do with their branding or flavour.

And it turns out there is actually a useful way to think about this in the framework of standard, rational-choice, Chicago economics! Charles Manski came up with a way of answering a whole class of similar questions as long ago as 1990, in the context of treatment/control-group studies where not all the experimental subjects bother to respond to the follow-up.

Imagine you think your Dutch East India Company shares are worth at most 49 guilders. (I am going to borrow, and bastardise further, Dan Davies’s popularization here.) You see someone on the Dam offering to buy at 50 and sell at 51, and not surprisingly take the deal. Then someone else turns up who thinks they are worth 99 – this coffee stuff is really popular, to say nothing of tobacco and exploitation – and not surprisingly, she also takes the deal. The market has cleared, everyone is satisfied by the transactions, but although one of us thinks the VOC is trading close to its fundamental fair value and the other thinks it is wildly cheap, the price hasn’t budged at all. In fact, the market-clearing price could be anywhere between 26 and 74. Although there are many other possible states of the world in which we would agree on a price, they are lost to us, like the experimental subjects who didn’t come to the follow-up, and all we can say about them is that they must fall somewhere within a statistical distribution defined by the limits of the possible. There is information in the price, but there is also a great deal of noise.

The lesson from this, I think, is that trying to tweak consumer behaviour is always going to be a hiding to nothing because there are so many other contributors to the final bargain, and the additional information introduced by the tweak will tend to vanish into the noise. This would even be true of something like the national consumer cap-and-trade scheme David Miliband flirted with as a minister back in 2007, even with its panopticon awfulness. I am therefore convinced that the collective approach is right, and I would even point out that David’s younger, better brother really achieved something in that line. The real Ed stone should have this visualization of coal-free days carved into it.

3 Comments on "Coal, the Manski bounds, and the correct choice of Milibands"

  1. So the tweak is a signal that ideally comes through the filter of the market, and survives duration and noise.
    It might indeed by too small. It may also fail to integrate coherently enough to be noticed. This can happen with policy U-turns such as a cultural change to “get rid of the green crap”. You will tend to discount stuff that your experts tell you is going to go away (for instance if your lobby community is going to make it go away).
    This is perhaps invariant to the scale or level of intervention.
    One thing is for certain, effective collective intervention is going to attract the attention of the Koch family promptly. Magnitude trades against duration.


  2. I won’t re-engage my rant about how post-Hayekian treatments of price collapse too many axes of information into a single metric to be able to do the things they think it can do – but this is another example.

    Some kind of carbon tax can probably be useful for the most energy intensive activities, but we know we have to go so much further than that – and quickly, too.


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.