There was a debate on Twitter this morning about how economists can better engage with the public.

Our efforts to do so have not been wholly futile. Granted, economists' influence on the debate about Brexit and austerity hasn't been as great as we would like. But on the other hand, Tim Harford and Steve Levitt have gotten a wide public audience - one perhaps more deservedly so than the other, And my attempts to bring proper economics to the albeit niche readership of the Investors Chronicle have not been met with complete derision.

Such episodes show it is possible for economists to gain some influence. What follows are some suggestions as to how or when this can be done.

First, economists cannot hope to challenge beliefs that are part of people's self-identity. If you tell someone "economics says you're a twat" you'll not get a sympathetic hearing even if you are bang right. This helps explain the success of Freakonomics and the failure over Brexit: nobody had strong prior beliefs about whether Sumo wrestlers cheat, but they did about the EU.

In this context, I have it easy at the IC. People might not want to hear they are fools but they do want to know ways of making money, or at least to stop losing it egregiously. I can therefore try to be Keynes' dentist, offering humble but competent advice without telling them they are fools.

Secondly, try to appeal to facts rather than the consensus. If you say "most economists believe x" you are at best only inviting the question: why might they believe that? And at worst, you're inviting a sneer at experts.

Instead, tell them hard facts. For example, Chris Giles' statement that the UK trades 3.3 times as much with Spain as with Australia was a lovely clear way of alerting people to gravity models. And some of my more successful efforts have consisted in pointing out that some types of stock, such as defensives and momentum, do tend to beat the market. Facts aren't just useful in themselves. They answer the question: what do you know? They get you a hearing.

Thirdly, be clear about what you know and what you don't. Claiming to be able to forecast short-term GDP moves when you can't, or proclaiming a "Great Moderation" when this was temporary luck due to a lack of shocks, have brought economists into disrepute. Equally, if R-squareds are low or correlations are variable, say so. You've got to know your circle of competence.

Fourthly, don't talk down to folk. In the day job, I try to imagine my reader as an intelligent and sceptical non-economist: a professional person or successful businessman. (I'll concede, though, that I don't always do this on the blog.) When I write about cognitive biases, for example, I try to not to say "this is why you're stupid" but rather: "these mistakes have been seen in other people: be on guard against making them yourself.".

Of course, even if economists could always follow this advice (and I don't claim to do so myself), they won't always succeed. It's difficult to talk to those who won't listen, and there are few channels through which economists can talk even to those who want to listen. Our media (including the BBC) selects for blowhards rather than sober minds - a problem not confined to economics: how much influence do philosophers or sociologists have on the public discourse?

Economists will always have a tough job in the public sphere. Perhaps, though, there are some ways in which we might make it less tough.