Tim Harford says "the world needs more wagers between pundits." But he omits to say that, in fact, we see trillions of pounds of such wagers every day. I'm referring to financial markets. Many trades (apart from those undertaken for hedging or liquidity reasons) are bets on an economic theory - that there's predictability in a particular exchange rate, that a share is under-priced, and so on.

Now, most of those bets are gambles with other people's money - there are good reasons why so many traders work for big banks rather than on their own account - and traders can be selected for bumptious overconfidence.

Nevertheless, I'm tempted to agree with Tim that this form of betting can be a good thing in terms of its cultural effects.

The point here isn't simply that a bet is a tax on bullshit. It's that bets (or trades or investments - they amount to the same thing) teach us an important lesson. Everyone who has made a decent number of them learns that however good is your research, theory and evidence, you'll often go wrong. In this sense, trading teaches us that we have "severely limited knowledge of the opaque, complex, social world."

This is especially necessary because most of us go through life over-exposed to like-minded people; our friends tend to think like we do and we read websites which echo our own beliefs. There's an element of the smackhead in most people; we go through life surrounded by a comfortable cushion which numbs us against reality. Betting is a way of exposing us to that reality.

If more politicians had been traders, we might less less daft certainty about economic policy.

Of course, one could learn this lesson from reading the literature on bounded rationality and knowledge. But the best lessons - those that really change us - are learnt by practice not reading; I'm always surprised by the fact that so many Austrian economists who are supposed to have read Hayek often seem so damned certain about their opinions.

It's in this context that my readers at the Investors Chronicle are so different from others. If a Telegraph or Guardian reader has silly ideas about the economy, he hurts nobody other than those stupid enough to read the comments below columns. If an FT reader has a daft idea, he hurts hurt other people - employees, colleagues or tax-payers. But if an IC reader has a stupid idea, he hurts himself. This, plus his experience of the occasional misfortune, gives him a sense of humility and an awareness that thinking about economics is tricky.

What I'm saying here is a variation on a theme of Deirdre McCloskey. Markets - proper markets, not rigged ones - create a culture amenable to some virtues, those in this case being humility and scepticism.

Is it possible to create political institutions which foster such virtues?

Yes. One argument for demand-revealing referenda is that a system in which people have to vote a sum of money to back their favoured policy would drive out blowhards and favour evidence-based arguments. But despite the right's enthusiasm for introducing markets into so many other areas of life, such referenda have never been popular. I really can't think why.