It is possible to wonder whether some people have quite grasped the argument that they themselves are making. You know, the people who insist we must not listen to superstition because the tarot cards said not to. Or, given the focus upon economics of this column, those lovely people at the Centre for Policy Dialogue.
As reported in this newspaper they tell us that we should change the method of fuel pricing in Bangladesh. Seems fair enough and why not? Then that we should use artificial intelligence networks to do that for us. Also, that the Bangladesh Energy Regulatory Commission (BERC) should take full charge of fuel pricing. Finally, that everything should be fully market based.
But which? Markets, computers, or the bureaucracy? The use of any one of the three excludes the other two by definition.
There’s a lovely essay out there online, In Soviet Union Optimization Problem Solves You, which points out that these neural networks, artificial intelligence, or even computers themselves, cannot calculate their way through an economy. We simply do not have enough computing power, we won’t for at least a century, and no, changing the name from computer to neural just doesn’t do away with this problem.
But that’s a technical problem, we’ve also got a logical problem. Before we plan a market -- and the prices in it -- we have to know utility functions. That’s how society will react to a change in prices, a change in the market. Societal change is an aggregation of each individual utility function. On that idea of fuel prices. We all have different transport possibilities, trains, cars, rickshaws, bicycles, or by foot.
Let’s change the relative prices of each -- so, what is each individual going to do? That’s what will then change fuel consumption as a result of the change in prices after all. But we ourselves don’t know what we’re going to do. We’ll only actually make the decision when we’re faced with the change in prices. No, asking people first doesn’t work -- how much of our lives is made up of us not quite doing what we’d said we were going to do anyway?
No one, until faced with the change in prices, really quite knows what they’re going to do.
The only exact and detailed way of finding out the reaction to a change in prices is to change prices and see what happens. Which means that planning the effect just isn’t possible. Not just technically but even in theory. That we use computers to not know faster really doesn’t aid us in overcoming this problem. We don’t have the basic underlying knowledge -- of those utility functions -- to be able to plan and cannot because no one, until faced with the change in prices, really quite knows what they’re going to do.
This is why bureaucracy -- the BERC -- cannot plan fuel prices for us. It’s also why computing isn’t going to do it either. Which leaves just using actual markets to do it. Which is, in fact, where we should be anyway.
Yes, supply of this, demand for that, changes a bit. Prices then alter and everyone then reacts to the changes in prices by altering their supply and or demand. As Hayek got his Nobel for the only method we’ve got of calculating all the way through all of this is the market economy itself. No, not just because computing isn’t good enough, but because we’ve not got and cannot get that information -- the utility -- to enter into a program as the input to a calculation.
So, markets to decide prices it is.
At which point it’s possible to be a little unkind about the CPD. They’re part of the -- or a local affiliate of, that sort of thing -- the World Economic Forum, what the rest of us often know as “Davos.” Which is a collection of really bright people who are really certain they know how to run the world for us. Unfortunately, they’re not quite bright enough to grasp that they can never have the information about us to run the world the way we want it to be.
Which is why Davos -- and here the CPD -- never do quite come to the conclusion that it’s too tough a problem for them really bright people, so they’ll have to use markets instead. Despite it being obvious that we’d all be vastly better off if they did.
Tim Worstall is a senior fellow at the Adam Smith Institute in London.
