Unearthing market inefficiencies: How bettors outpaced financial traders in predicting Brexit

Gamblers, instead of financial traders, were the first to accurately forecast the Brexit vote, a study led by Cambridge University researchers has discovered.

Gamblers, instead of financial traders, were the first to accurately forecast the Brexit vote, a study led by Cambridge University researchers has discovered. The study revealed how financial markets lagged behind betting markets in accurately predicting the results of the Brexit referendum by several hours. The gamblers’ prescience unlocked a potentially risk-free and lucrative arbitrage opportunity, according to the economists involved.

International financial markets were rather slow in assimilating the Brexit result on the night of the EU referendum, trailing behind gamblers who had correctly anticipated the Leave vote. This discrepancy offered a window of arbitrage where the disparity between betting and foreign exchange (FX) markets could yield a return of up to 7% on the pound.

Punters vs currency experts: The power of prediction

The research team from the University of Cambridge meticulously scrutinised the behaviours of the Betfair betting market and the pound-dollar exchange rate from the closure of the polls at 10pm, a time when the odds were 10 to 1 in favour of Brexit. The researchers discovered that both markets were “informationally inefficient”, responding sluggishly despite the influx of available data. This created a valuable opportunity for early trading on either market.

By 3am, the betting market had gravitated towards a Leave result, with the odds for Brexit now reversed to 1 to 10. In stark contrast, the foreign exchange market did not fully adjust to the reality of Brexit until around 4am. Not until 4:40am did the BBC announce a forecast for a Leave victory.

Capitalising on market inefficiencies

The variance in efficiency between the two markets gave rise to a one-hour period when selling £1 and hedging the referendum outcome on Betfair could have reaped up to nine cents of profit per pound sterling. This significant “unleveraged return” could, in theory, have allowed sharp traders to accumulate substantial gains.

The study’s results lend support to the notion that gambling or “prediction markets” may offer more accurate forecasts of election results than conventional expert analyses or polls. Dr Tom Auld, the lead author of the study, postulates that Betfair punters demonstrated a more acute intuition that the Leave side could emerge victorious, or at the very least, that the outcome could swing either way.

Challenging the ‘efficient market hypothesis’

Dr Auld and co-author Prof Oliver Linton used publicly available expected outcomes for each voting area to create a forecasting model. They claim that, had their model been used on the night, it would have accurately predicted the final result from around 1:30am.

Contrary to theories such as the ‘efficient market hypothesis’, which posits that markets discount all publicly available information, the study showed that the financial markets were markedly inefficient. According to Dr Auld, these markets should have predicted Brexit potentially over two hours before they did using the available data.

This study underscores the value of “prediction markets” such as betting exchanges, where participants put their money where their mouth is, rather than offering “cheap talk”. For future research, Dr Auld plans to focus on how these markets could be used to help value or price financial assets during significant events like major votes.

The advent of sport betting sites and other online platforms might soon lead to a paradigm shift in how financial markets respond to globally significant events. Understanding this shift could arm us with better prediction tools, fostering more efficient and informed markets.

Vey Law

Vey Law is a reporter at Breakthrough.

Latest from Blog