Note to gilt traders: betting markets are not opinion polls

Are political betting markets really moving UK government debt prices? Maybe! Should they be? Absolutely not!

What happened was a tweet:

🚨 𝐁𝐈𝐆 𝐍𝐄𝐖𝐒 developing this lunchtime…

Boris Johnson is now the odds-on 𝐅𝐀𝐕𝐎𝐔𝐑𝐈𝐓𝐄 to become the next Prime Minster. pic.twitter.com/nCEZAdK9Kq

— Betfair Exchange (@BetfairExchange) October 21, 2022

Which was followed by lots of tweets along these lines:

NEW
UK gilt yields have jumped sharply in the past hour or so following news that Boris Johnson is now favourite, in betting markets, to be the next PM.

NB all bonds moving a lot today but UK more than most
Chart is the two year gilt yield pic.twitter.com/I48CwbXFWQ

— Ed Conway (@EdConwaySky) October 21, 2022

And yes? Maybe? Overlay the Betfair Exchange implied-chance graph for Boris Johnson becoming next PM (red line) onto the 2-year gilt yield (white) and there’s a sort-of correlation:

(Note: Betfair doesn’t include a timescale on its markets graphs so we’ve had to use some guesswork framed around how long the book has been open.)

The fundamental problem here is that peer-to-peer betting markets don’t test public opinion. Nor do they measure changing probabilities in any meaningful sense. Moment-by-moment pricing is much too thin and too susceptible to manipulation, with each price twitch amplified by trend chasing and circular reasoning.

The weight of money that briefly moved Johnson into pole position on Friday is unlikely to have exceeded the low four figures sterling — though the publicity value extracted by the ex-PM bookmaker is a multiple of that.

Anyone new to exchange betting should also know that much of the liquidity is bot driven (Betfair offers recommendations) and tied to imperfect arbitrages. In addition to a “Prime Minister after Liz Truss” market there are much thinner books for next Conservative leader, Johnson reaching the 100-plus nomination threshold and Johnson as Conservative leader at the next election. Scalping any potential disconnect (or selling training courses that teach the art) is a cottage industry.

Since we started this post, Rishi Sunak’s probability of winning is back at 60 per cent-ish and the 2-year gilt yield has slid more than 25bp back to its session low. Nothing of note has happened during that period, either in politics or fixed income. But the bookmakers will have made some money, which is nice for them.