Whenever some important sports event looms, the popular press inevitably comes up with articles about the loss of productivity during said event. The quoted numbers have always been a mystery to me, as I have yet to see any serious study about this. Well, now there is one, sort of.
Michael Ehrmann and David-Jan Jansen analyze stock market activity during the 2010 FIFA World Cup in South Africa. Using minute by minute activity from 15 stock markets, they find that there are markedly fewer trades when games are on, especially when the home side plays, roughly half of normal trades. That may be because traders are less attentive to markets, but also because domestic investor are less on the ball, so to say. The effect is amplified by goals.
There are other times when attention is decreasing, such as lunchtime. The decrease in trades may have some welfare implications, as some arbitrage opportunities are not taken. However, it appears to be more serious that there is a change in patterns for prices during those football matches, unlike lunch time. Indeed, the cross-section of returns across firms is lower, and the correlation of domestic prices with world prices drops by 20% when a game is on. I cannot quite quantify how much of a welfare consequence this is, I would need a model for that, but I guess this is not negligible. NB: these effects are also present on US stock markets, despite the marginal interest in soccer.