If FX Benchmarks were Manipulated, How Often did it Occur?

FX Week (subscription) reports that the counsel for one of the plaintiffs in the FX benchmark class action suit says that their analysis of London Close benchmark trades for the six most liquid pairs on the last day of the month for the past 10 years shows manipulation occurred between 26% and 34% of the time.  The headline of the article, however, is a bit misleading: FX benchmark manipulated more than 25% of the time, plaintiffs say.

Even if the interpretation of these results can be said to be true, this still would not mean that there was that much manipulation on all trading days.  The last day of the month was most likely reviewed because most asset managers that hedge the FX in their portfolios adjust the hedges on that day, leading to particularly high volumes.  These high volumes, and the tendency for these hedge adjustments to be in the same direction, would make those the days most likely for manipulation if there was any.

So even if you did accept the results, it does not answer the question of how extensive was any FX benchmark manipulation.