The evolution of the weekend effect in US markets

• The weekend effect initially disappeared soon after its “discovery” in 1973.
• Breakpoint tests confirm new regimes in 1975, and thereafter.
• Unit root and stationarity tests suggest mean reversion toward zero.
• Our analysis points toward the long-run disappearance of the weekend effect.

This paper applies cointegration and breakpoint analysis to test for the disappearance of the weekend effect in US stock markets. After its formal discovery in 1973, the weekend effect immediately declined, but subsequently went through periods of reappearance and even reversal. Unit root tests suggest stationarity in the Monday return differential from 1975 to the present. Our results support mean reversion toward the same return as for the other days of the week, rather than a wandering weekend effect. Breakpoint tests indicate a major change in return regimes within about two years of the discovery of the weekend effect, subsequent periods of adjustment that vary by index, and the long-run disappearance of the weekend effect.