The Last Look…
Posted by Colin Lambert. Last updated: October 4, 2021
It’s time to get all “Oliver Stone” and probably see something that isn’t there, but…I wonder if we are building evidence around the activities of speculators at the month-end fixes?
The Full FX has published the sixth analysis provided for it by Raidne looking at the two month-end fixes, WM and Siren (owned by Raidne). Last week was the quietest month-end so far, but there should be some context on this, there were still significant savings to be made – even at 60% correlation to the flow there is $72 per million on offer, which should not be sneezed at by asset managers/owners.
What has intrigued me, and it is still very early days I accept, but looking at my notes of each month-end, there seems to be a pattern emerging between the level of surprise at the direction of the Fix, and the actual savings garnered from a longer window. So far as I can tell with my notes, the market consensus has been wrong on month-end flows in June, July and September – coincidentally (and it may be just that), the three months with the lowest saving.
Those three months have had a portfolio saving of $429, 430 and 370 per million at 100% correlation – still sizeable, especially when talking billion-dollar transactions, but below the $1,054, 790 and 966 savings in those months where expected dollar buying/selling actually materialised.
I can see two potential factors behind this, the least likely is (pre)hedging not seeing the benefit for the executing party as expected (although there were still benefits). The problem with this is, (pre)hedgers trade ahead of the Fix according to what their analytics tell them – it is very much about the size of the order rather than direction. If this is the case then the (pre)hedging window would probably also have seen lower market impact.
More likely in my mind is that speculators are positioning ahead of the Fix based upon the aforementioned consensus. If that is the case, and they get it wrong, they are likely to buy/sell back during the window, creating more offset flows for the fixing flow and commensurately lower savings for the longer window.
At the September month-end there were a few eyebrows raised when Cable rose steadily after the window closed, but again, this is likely to be speculators. Price action in the window was choppy and well-balanced (for once), which meant those speculators were “kept interested”. Because no real flow emerged, they didn’t have an exit signal. Once the window closed of course, it was there for all to see and Cable rose 20-30 points.
Cleverer people than me can probably run some proper analysis on this, but there are my thoughts on it. I am not sure there is much to be done about it, unless proponents of the WM Fix want to start a regular disinformation campaign!
I think it is important to note that we are still seeing evidence that the longer fixing window works better even when the consensus is wrong. Yes, the savings during these windows are something like one-third to a half what they are when the consensus is correct, but we are still talking savings in the hundreds of dollars per million.
And to paraphrase what the man said, “a million here, a million there – soon you’re talking real money”.