The Last Look…
Posted by Colin Lambert. Last updated: June 14, 2021
A couple of times over the last week I have found myself wondering why I missed a news event only to find out it was during my three months’ enforced absence around the turn of the year. Apparently the world carried on and news still happened, which is, naturally, crushing for me!
I mention this because someone drew my attention to the case of John Gorman, the former Nomura interest rate trader, who was charged at the start of February with our old favourite, market manipulation.
The CFTC alleges that Gorman identified a particular SEF broker whose prices were displayed on page 19901 on Eikon – a page that is, apparently, seen as an “informal benchmark” for interest rate swaps. I have a problem with “informal” anything when it comes to trading, but the fact of the matter is page 19901 is an information page. What were people thinking in using it as a benchmark? There’s no firm liquidity behind it – even if inter-dealer brokers do post prices to the screen. It’s like creating a Fix on prices that are last looked – they are meaningless.
It surprises me sometimes, how people interpret my comments about the London 4pm Fix as a criticism of Refinitiv, the company responsible for it. It is nothing of the sort, although I will admit I believe the firm should be more proactive and lengthen the window. As for the actual mechanism itself, it’s fine, if, as I have often observed, measured over too short a time frame.
The fact is we need a benchmark of some sort, one that is regulated and calculated according to a fair and transparent process. My problem with the Fix is very much that it has not evolved as people have used it differently. As a traded benchmark it needs to better reflect the amount of flow.
To give you an example of how naïve using a page such as 19901 is, consider the days before electronic trading (I know!). For more than a decade Reuters page FXFX was on every dealer’s screen (and because clients didn’t have access to Matching or EBS they had another 10 years of using the page). This page gave indicative rates, and sometimes they bore little or no resemblance to the real market.
There were banks out there, trying to get their name out (why I do not know) who would immediately copy the latest update from another bank to get their name on the screen (the olden days equivalent of pipping!) There was a small eastern European bank that few had ever heard of that was always on the screen with a Cable price – every time the price was updated by another player, this bank’s price would immediately replace it.
Clearly the bank concerned rarely did Cable, it certainly wasn’t a market maker (this was confirmed when my desk once called said bank for a price and was told they don’t quote it!), but this didn’t stop them being “top of book” when people checked the screen for the level.
This bank was the talk of the London market for a little while because they would just blindly follow another quote, and on Friday afternoons if it was quiet, some banks in London used to enter prices 50 points away from the market just to see the eastern European bank update at the same level.
All good fun for a boring Friday afternoon (there was no TV to turn to the sport), but why relate this story? Well, I happen to know that some firms used to use FXFX as their benchmark for FX executions, just as they were apparently using 19901. This meant that they would buy or sell whatever they had to do, and then hit ‘print’ on FXFX – as long as the timestamp was correct, the rate was what it said. This meant, if said bank was updating and someone was playing around, some firms could get a very nice print to show how well they had done.
There was, according to the unimpeachable source of everything true in the world – the rumour mill – a nasty incident at a European corporate in the early 90s, however, because the firm executed a Cable trade and then watched as FXFX published nonsense rates for the entire minute of the “fixing”. The treasury then had to explain to the board why it had executed at a rate that was 20 points worse than the displayed market, which was, apparently, not easy!
When word of this got around, people stopped messing around with the page and, presumably, the customers stopped using it. We all knew it was an unreliable source of price information and the world moved on, but clearly the interest rate world didn’t get that memo.
I will be interested to see how the CFTC’s case goes against Gorman, predicting the US court system is, I have discovered over the past five years, anything but easy, however for me, the real lesson from yet another court case is how the buy side needs to recognise its own responsibilities. I understand as an industry we have moved on, but have we moved far enough.
In 2014 the Financial Stability Board’s report into FX benchmarks noted how users should ensure the benchmark is fit for purpose – this message has been reinforced more than once by the Global Foreign Exchange Committee. I understand that the Gorman case is in the interest rate world, but it does beg the question, “exactly how many buy side firms have actually studied their use of reference rates?” The Gorman case makes clear that just months after the FSB report, the answer would probably be “not many”.