Compared to the mayhem of April, the June month-end WM/R 4pm Fix went off reasonably OK – there were no huge moves, just a steady grind lower by the dollar except for against the yen, where it was in demand for the second time that day (the first being the Tokyo rate set at 9.55am local). Indeed, as has been argued by some of you who have engaged with me on this, the fact that most market remained above (or below) the print highlights the Fix is just reflecting demand. All of which makes my point that it is no longer fit for purpose.

It is exactly because this Fix reflected the demand that tells me we need a longer window. Why would investors want their administrative FX hedges causing market impact and costing them slippage? To be clear (again!), as far as I can tell there is no misbehaviour going on – there was demand against the dollar and yen, which meant those currencies went lower. Had there been a bunch of spec accounts jumping in front of the Fix we would have seen some sort of reversal towards, or at, the end of the five-minute window. This is not to say that no spec accounts were involved, just that their impact was not that huge.

During the five minute window EUR/USD slipped around 10 pips and USD/JPY around 16. While the Australian and Canadian dollars were broadly flat, this neglects the fact that USD/CAD had actually moved by 20 points in the five minutes leading up to the window, the Aussie had also risen by some 12 pips in that period.

Cable was an interesting one, it was in demand in the pre-hedging window, it was up some 15 pips in the 10-15 minutes prior, but then actually reversed 11 pips in the window – so perhaps there was, after all, some spec activity there.

After the window, EUR/USD kept grinding higher, USD/JPY reversed slightly (by 3 or 4 pips), while Cable, AUD and CAD were all in steady demand.

It highlights just how serious the issue is when we are tempted to dismiss such slippages in a Fix window, until you consider the amounts involved. Someone with access to decent information told me recently they thought that Fix flows could be as much as 15-20% of average daily spot volume at month-end, so just the $300-400 yards then. Chuck in decent netting of, say, 60%, and we only have one hundred and something yards to execute – what could go wrong? Even allowing for that number being too high, the fact is the amount of flow now being seen at the Fix is too high for a five minute window. Yes that window reflects the supply and demand, but why should managers pay away such huge amounts on slippage all for the sake of tracking error? When did we get to the stage where being as good (or bad) as your peers was paramount – including over actual performance?

It seems very simple to me, and I will point readers in the direction of my interview with Shirley Barrow, global head of benchmarks at Refinitiv, the firm responsible for the Fix for details of how change can be effected, but Fix flow ishaving an impact on market functioning. The continued attempts to push too large tickets (and I accept they may be aggregated) does not, to my mind, signal market participants taking care when submitting orders that those deals do not interfere with normal market functioning. As such they are in breach of the FX Global Code (not that many of them have bothered to sign it yet of course) and while they are not being bad market citizens per se, they most certainly aren’t being good, responsible participants in the FX market.

Everything will be OK, however, because when this Fix is highlighted as unfit for purpose, the asset owners will simply find someone to blame the way they always seem to…as long as they conveniently ignore their lack of interest and neglect of the issue.

Colin_lambert@profit-loss.com

Twitter @lamboPnL

Colin Lambert

Share This

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

Related Posts in ,