Month-end flows make prices behave in annoying ways
Posted: Sat Oct 31, 2026 6:37 am
Pension funds, asset managers, and corporates rebalance currency exposures at the end of each month. When equities have moved a lot, hedging flows can be large and one-sided, and the 4 pm London fix on the last business day often absorbs a large share of them.
That produces unusual behaviour. Pairs can drift against the prevailing intraday trend for no obvious reason, then reverse once the flow is done. Technical levels that would normally hold get overrun by orders that care about price much less than usual.
Quarter-end is worse. The same mechanics apply with larger size, and the influence can start a few days before the date. A strategy that works well mid-month may suddenly lose in the final week, and the cause has nothing to do with the strategy itself.
Marking month-end days on the calendar and comparing results with other days is an easy test. If performance consistently dips, reducing size or sitting out those afternoons is a cheap fix.
Does performance change in the final days of the month, and was it ever tested properly?
That produces unusual behaviour. Pairs can drift against the prevailing intraday trend for no obvious reason, then reverse once the flow is done. Technical levels that would normally hold get overrun by orders that care about price much less than usual.
Quarter-end is worse. The same mechanics apply with larger size, and the influence can start a few days before the date. A strategy that works well mid-month may suddenly lose in the final week, and the cause has nothing to do with the strategy itself.
Marking month-end days on the calendar and comparing results with other days is an easy test. If performance consistently dips, reducing size or sitting out those afternoons is a cheap fix.
Does performance change in the final days of the month, and was it ever tested properly?