The Last Look…

I rarely have sympathy for regulators – I understand we need some guidance in markets, but too often relatively uninformed (but well-intentioned) people end up making those rules and they just don’t work. I do, however, find that I am increasingly sympathetic towards IOSCO as it deals with pre-hedging – and just last week we got further evidence of how complicated and hard to manage this issue is.

When IOSCO’s consultation on pre-hedging ended in February, it said it would be issuing its final report to members “in 2025”. It has three months left of course, but I wonder how close it is and what sort of guidance we will receive, for just last week, one of its members – ASIC – highlighted just how much judgement around pre-hedging is a subjective matter.

Following the announcement that the Australian regulator had come to terms with ANZ and issued AUD 125 million in markets-related fines over the latter’s “unconscionable conduct” surrounding its role as duration manager on a large bond transaction (it’s hedge would be in the region of AUD 10-15 billion) in 2023, there was a press conference and the publication of a Statement of Agreed Facts (SAF).

In the SAF, it was clear that the client – the Australian Office of Financial Management (AOFM) – agreed that ANZ would execute “correlated hedges” (another new name for pre-hedging?). As it states in one paragraph, “[I]t was ANZ’s role, as duration manager, to manage the risk it expected to acquire associated with these bond futures. In doing so, it was permitted by the AOFM to trade before, during, and after pricing. The AOFM disclosed to the market that ANZ could do so.”

So, ANZ could pre-hedge its risk, which is the sensible course given the size of the transaction (more than a single day’s volume in the AUD bond market), and the customer gave them permission to do so, “…in an orderly manner to minimise market disruption, price volatility and price impact, and in a manner which promoted the fair treatment of the AOFM…”

My first observation (assuming I am reading the SAF correctly) is that AOFM didn’t help itself or ANZ if indeed it did “disclose to the market” what ANZ was doing. The old spy adage of “a secret shared is a secret squared” comes to mind – effectively the whole market knew what ANZ had to do (other major banks were part of the original RFP and were co-lead managers), and that will – inevitably, impact the price. Who is going to hold on to bond futures if they know yards of selling is likely?

Did the trader make credible efforts to sell much earlier, only to fail and be left with a large chunk to sell in the latter stages of the pre-hedging window?

Information leakage is vital around big orders – ask any market professional – and I cannot stress enough how this needs to be understood clearly if and when we establish pre-hedging guidelines. There is a responsibility both ways. It’s impact here cannot be understated, because the dealer responsible for the trade made several attempts to sell bond futures at a certain level “if bid there”, and the broker placed small offers in the market. I am not totally au fait with the market structure in AUD bonds, but in today’s FX markets, even a small bid or offer on the right venue will have some sort of impact on the price.

This also highlights how the dealer seemed to be trying to pre-hedge in a careful manner – there was no chasing the bid, and offers were cancelled after a short while if the market wasn’t there. It also, however, plays a role in one of the key complaints issued by ASIC against ANZ, for as ASIC chair Joe Longo said at the press conference, “The critical issue here is that ANZ sold a significant volume of 10-year Australian bond futures around the time of pricing, which placed undue downward price pressure on the bond price.”

This selling, Longo stated, “…potentially made the government $26 million dollars worse off…[and] reflects that the bond futures price decreased by two basis points over the 45 minutes before pricing when ANZ traded most actively in the market.”

Here is the first subjective question – did the trader make credible efforts to sell much earlier, only to fail and be left with a large chunk to sell in the latter stages of the pre-hedging window? Only one person really knows that – the dealer themself – the rest of us, including those bringing judgement, can only guess.

It is noted in the SAF, that the duration manager can also hedge after the rate setting has taken place, but Turkeys and Christmas come to mind – who is really going to hold a long position knowing they have dumped around 13 yards into the market and probably made it long as well?

Another subjective finding in the SAF was that “ANZ sold a low volume of 10-year Australian bond futures in the period between 8:32am and 1:10pm [pricing was at 1.50pm], despite opportunities being available for it to sell more contracts.”

There is only one easy solution to this highly complicated issue – ban the practice of pre-hedging in its entirety…but that is unlikely to happen and comes with another problem

In any market selling opportunities exist, as long as there is a bid – the question is though, what size were these bids and what would have been the effect of signalling a large selling interest even earlier? Personally, I see why this argument is being made, because the entire “unconscionable conduct” is based around timing, but it’s not that simple. Did, for example, the dealer just make an honest mistake on timing? Or did they deliberately go slow with the intention of dumping on the market to get maximum profit? Any view, apart from the dealer’s, is subjective.

Of course, there is one area where things were clearly wrong in terms of ANZ’s conduct – transparency with the client. AOFM argued, rightly in my view, that it should have been told that the bank was behind schedule with its pre-hedging. This could have, as ASIC correctly argues, led to a delay in the pricing to allow the bank more time – and it is here where one can see how the regulator felt ANZ dealt in its own, rather than its client’s, interest.

Things did get a bit confusing (for me anyway) during the press conference, when Longo said that ASIC “concluded that there was evidence of unconscionable conduct, and not evidence of breaches of the market manipulation [Act].”

He went on, “Now I know something has been said about, oh, well, was it a finding of market manipulation? Which I find rather curious, because unconscionable conduct in these circumstances is an extremely serious contravention. It’s a breach of commercial confidence. ANZ did not do what it said it was going to do with the AOFM. It was not transparent, did not follow its own policies and procedures, did not keep AOFM informed, and even after the transaction when questions were raised by AOFM about issues it had, and indeed ASIC had at the time with the trading, they were misled again. This is very serious unconscionable conduct. So I think it’s rather curious that people should be concerned about market manipulation. It’s serious enough that we found a contravention that’s been admitted to unconscionable conduct.”

In other words, the lack of transparency with the client was the really serious matter, and while the underlying complaint was that ANZ forced the price lower in the last hour before the rate setting, and cost the client money, this was not market manipulation? Longo also observed that “ANZ have taken the view they’re not prepared to concede that the Commonwealth suffered a loss as a result of its conduct. It is prepared to concede that its trading had a downward pressure on the price of the futures, but it has not conceded that the Commonwealth suffered a loss.

“That is not ASIC’s position,” he continued. “ASIC’s position is the Commonwealth did suffer a loss of around $26 million. And that is calculated by reference to the way the trading went and the value that was placed on that trading at that particular moment in time.”

I understand lawyers get involved in these statements, and I may be being thick (not for the first time I hear you cry), but this all seems a bit paradoxical. The bank said the client didn’t lose money, even though it admits its selling did affect the price, while the regulator says there was no market manipulation but the price wasn’t where it should have been?

If you’re selling 14 yards of bond futures there is going to be a market impact – and this is really the crux of the issue – and is again very subjective – how far should the market have moved? Given how market conditions fluctuate – and with a nod to those TCA providers who try to help such matters but simply cannot give guidance with 100% confidence – I would have thought it impossible to know.

Effectively the regulator is saying, “you should have done a better job of selling”, but good luck with that. As any trader can tell you, selling a yard of, for example Cable, one day may not move the market, but the next day may half destroy it. Market impact is very much a unique phenomenon – we can’t put hard and fast rules on it, beyond ‘don’t leak the information about the trade’!

This is where we are, therefore, as we await IOSCO’s guidance – all large trades come with huge conduct risk attached, but should they? To reinforce my point in the last paragraph, one trader may seek to push the market lower to maximise profit and fail, due to strong, unexpected, demand; while another, seeking to do the right thing in every way, may hit a weak spot and see the market collapse if front of their eyes. One of them is likely to be punished in this hypothetical example, and we all know who it is, even though it’s the wrong person.

There is only one easy solution to this highly complicated issue – ban the practice of pre-hedging in its entirety. I don’t think that is going to happen because it is too radical a step for the regulators of the world, however it is the only one that brings a degree of security to those executing large trades.

It also comes with another problem, though, for this would, of course, mean that large trades have to be priced accordingly, with the risk taken into account, which means wider spreads – and how many clients will be happy with that? A ban would also mean that good traders can demonstrate their skill in pricing and risk management – but this comes with a sizeable problem in that it involves taking on market risk, which is a function largely eradicated by the actions of? You’ve got it, the regulators…

colin@thefullfx.com