The reference-rate conundrum

Source: KangaNews 2 February 2019

The RBA has said we may see some change in BBSW usage in contracts if and when that happens – specifically, but not exclusively, in cross-currency swaps. That’s a logical observation, but we don’t yet know how big the impact will be.

The fact is that these rates, globally, are based on a type of activity that is drying up and I don’t see any obvious way to get around this without fundamentally re-examining the nature of the the use of rates markets.

Ultimately, what we are trying to do is add alpha from relative-value positions. We need an interest-rate derivative to do this and the reality is that LIBOR-based contracts are the benchmark for this type of transaction.

We use things like swaptions, and the last time we asked if it was possible to do a swaption based on OIS in the US market the answer was: “forget about it”. We had to go back to a LIBOR-based contract. What I’m saying is that there is no progress being made – or at least none that we can benefit from or encourage by adopting it ourselves.

Our attitude is that if IBORs are going to end we would expect to close out legacy positions and start replacing them with the replacements. But it doesn’t seem that this is going to happen any time soon.

The concern we have is that pricing credit instruments off a cash rate with no credit premium opens up basis risk. If I buy a credit instrument and there is a funding crisis tomorrow there is a decent chance that the risk-free rate won’t move – so what happens to that instrument?

We’re also not sure what happens in the event that an Australian issuer prints in Australia with BBSW as reference and also prints the same note offshore with a risk-free benchmark. Does the offshore bond trade at an initial premium to compensate for the lack of credit premium in the base rate?

To be honest, we quite like the status quo in which three-month BBSW is a standard and robust benchmark.

The reality is that you can base a transaction on whatever rate you want – I’ve never had any problem on that side. The difficulty is getting out of it or hedging away from it.

Markets tend to gravitate towards the frameworks with the greatest liquidity, and the longstanding liquidity of the BBSW environment is what makes it hard for me to see it disappearing with any great speed.

Pieter Bierkens has mentioned the challenge of different types of bases for cross-currency or other swap activity, for instance SOFR on one leg and an IBOR in Australia. I also think this is possible – it would just have to factor in basis risk.

History tells us that if a fundamental type of basis risk emerges it is likely that a new market will spring up to trade it. If a market provides sufficient liquidity to operate an IBOR in a fluid environment I don’t see any reason why it shouldn’t flourish. This is why I think BBSW, as currently constructed, has a sound future.

I’m not saying other benchmarks can’t eventuate – they clearly can. Indeed, there may be times when it is appropriate for securities – such as those that don’t carry any element of bank-credit risk – to price off them. But we know that markets gravitate towards liquidity and BBSW is the most liquid option we have.

Real-money investors are typically coming at it from the perspective of understanding the risks involved in buying a bond marked against a specific benchmark and how they will be able to exit positions. I think they tend to be reasonably relaxed about BBSW and believe it will be sustained.

It’s the issuers that are most engaged, in our market at least. This is especially notable for any issuers using US markets, where they can see regime change is clearly happening.

As a first step, we realigned our client borrowings. For the past four years we have referenced a borrowing rate as a margin over the RBA cash rate for any of our clients that borrow on a floating-rate or short-term basis.

The second step is to align our floating-rate issuance to reference the RBA cash rate, in order to remove the credit-risk element from both sides of the balance sheet.

This work has been ongoing for SAFA and it included keeping a close eye on regulatory developments. As momentum has built over the last 18 months, the evolution of benchmark reform in global markets has picked up pace in expectation of the imminent removal of IBORs.

In Australia we have a credit-based benchmark that is judged as suitable at this point in time. But our goal is to seek opportunities for an alternative benchmark that is appropriate for SAFA’s business.

From there, we break it down into the domestic and global markets. Locally, we also feel much more comfortable with BBSW under its new methodology. For instance, we issued a five-year floating-rate note (FRN) with BBSW pricing domestically in January 2019 and we had no qualms about doing so.

Globally, it’s worth noting that the majority of our issuance is in fixed-rate format. This means our legacy-issuance challenge is relatively small, although even though the volume is not material we will have to deal with it at some point in time.

Our approach has been to get ourselves ready to issue come what may. This means having fallback language in our programmes such that we have a mechanism for deals we have issued to move to a new risk-free rate subsequent to the demise of LIBOR.

We have issued two sterling transactions linked to SONIA in the past three months. These were relatively short in maturity but were intended to test mechanics such as back-office systems. Tenor and volume were less important than testing the processes we have in place.

We elected to issue first off the SONIA risk-free rate because the UK market is most developed in this regard and, at the time we issued, pricing was flat to where we could have issued in other markets. Not paying a premium to issue is clearly an important additional consideration.

The SONIA market is by now well established to the extent that I don’t envisage issuing linked to LIBOR again. As far as I understand, it may not be 100 per cent of sterling investors that are able and willing to buy SONIA-based paper but the majority certainly are.

TCorp has explored the impact on bondholders if BBSW ceased to be the benchmark market rate. TCorp’s documentation relating to existing FRNs does not include a fallback clause that allows an automatic amendment to the reference rate. To get bondholders to agree to a change would require 75 per cent engagement, which would be challenging.

TCorp has revised its documentation to include relevant fallback clauses which allow for flexibility of other indices. Clearly it gives us confidence that the RBA has indicated BBSW will remain as a benchmark alongside a risk-free rate.

The other reason we have worked on the impact of the cessation of IBORs is that we provide interest-rate and FX hedges for our clients. Most of the work we’ve carried out has been focused predominantly on this area. We have also sent a submission to the International Swaps and Derivatives Association (ISDA) with our preferred fallback clause.

We are very much in favour of the development of an AONIA product in Australia and there would be advantages to issuing in AONIA-linked format. This is notwithstanding the fact that the majority of the bank’s assets are residential mortgages that are paid on a monthly basis, and at times volatility in the bills-OIS spread has resulted in a change in pricing for customers.

I am not looking to challenge the robustness of three-month BBSW. But if there was an alternative – like the ability to issue linked to the cash rate, particularly for one month – and if the benchmarks could exist side by side, this would suit CommBank as an issuer and an institution.

Source: KangaNews 2 February 2019

Most asset managers, superannuation funds, pension funds and insurers benchmark their cash cohort against the 90-day average, so there is a tendency to gravitate towards this to reduce the basis risk of the construct.

However, we are in an environment of no change in monetary policy for a considerable period of time and, as a result, many investors would be happy to take on some exposures – subject to the attractiveness of the specific offering – with an underlying cash benchmark instead of 90-day BBSW.

If one-month BBSW is phased out it will, as others have mentioned, have considerable impact on mortgages. As a mortgage investor, I’d prefer to get amortising payments every month rather than every three.

Mortgages are less liquid compared with corporate or bank bonds, so the fact that they amortise monthly is helpful from a liquidity perspective. The cash rate is an overnight rate so there’s a term mismatch as well.

My feeling is that BBSW is still quite viable based on underlying trading activity. The problem with IBORs globally is that they are still survey-driven, which is likely to be discontinued. BBSW is still traded and the VWAP system is a good method of converting the activity to a price. It would be great if volume was higher but what we have is still a better option that what they are dealing with offshore.

If I’m right and BBSW continues, the next question is what happens to the basis swap when there is BBSW on one side and a risk-free rate on the other. I suspect the market will be able to develop to manage this, though.

It would be better if the banks were issuing. But banks are also interested in buying back one-month paper, which prints a rate in itself.

The construct of the time period around the mid-point will continue to be a work in progress to get a less volatile rate construct. But the rate itself – I don’t see it going anywhere.

We would be happy to issue RMBS off a three-month benchmark – it’s not impossible from an issuer perspective. But we want to issue what investors want to buy. If it’s clear that investors don’t want to buy a product, it’s not in anyone’s interest to offer it.

What I’d say is that there is definitely appetite on our side and encouragement from regulators to look at something different. If there was demand from the market we would certainly be happy to issue into that demand.

It’s really a question of whether the wider market considers something to be price-appropriate and sufficiently liquid. This is why I welcome SAFA’s step of saying it is going to issue an AONIA-linked deal – because any transition has to start somewhere. There are lots of good arguments as to why we should be issuing one-month-based securities linked to the cash rate. But we want to be confident that investors will buy it.

One associated change to mention is the introduction of regulatory standard APS 221. This requires banks to implement measures and set limits to monitor and control their large exposures and risk concentrations. The implication is that banks may hold less of each other’s assets and potentially further reduce liquidity in negotiable certificates of deposit (NCDs).

This may have an additional negative impact on the ability to set a rate for BBSW across the curve, especially in light of the impact that the NSFR has had on the one-month rate set.

I’m not overly concerned about this because the reality is that change is happening, globally and in Australia. We are already seeing dislocation in short-term markets based around regulatory changes such as the NSFR and APS 221, and these changes may further alter the shape of the NCD market.

We know the new product is unlikely to appeal to everyone. However, for a certain sector of the investment community it makes perfect sense. For instance, if you are an authorised deposit-taking institution with holdings including semi-government floating-rate commercial paper (CP), an AONIA FRN makes complete sense.

Instead of buying three-month SAFA CP you can buy a one-year SAFA AONIA FRN with monthly resets and a yield-equivalent pickup which, at the same time, decreases duration risk – because the product uses a daily compounded look-back rate – and extends the maturity profile.

There is no urgency to transition in Australia because BBSW will continue to exist for the time being. In the mindset of the investor community, until there is a desire to start changing the matrix for how the benchmarking for money-market funds is managed there will be no urgency for change in systems or technology.

Having said this, products have been developed that reference the cash rate – particularly deposit arrangements with extendable or callable features. But most superannuation funds, and therefore fund managers, continue to use the 90-day bank-bill rate as the primary benchmark.

The benefit from abandoning something that works is arguably zero. The cost is quite substantial. We have established that FRNs based on BBSW are the most well recognised, so when an issuer – taking SAFA as an example – says it wants to issue against AONIA it will be creating an instrument that is less liquid and less well recognised.

The bottom line is that an investor in an instrument like this will require an illiquidity premium for it – and the good people of South Australia will have to pay this premium for the benefit of the market.

There needs to be a good reason to make the change. The ultimate question – in Australia, putting international developments aside – is whether such a reason exists. The only one I could see is if there are grounds to believe that BBSW can’t exist forever.

From my perspective, as long as the four major banks are issuing bills every day – which they obviously are, at least in three- and six-month tenor – I personally don’t see why there should be an issue with BBSW.

I look at the situation in Australia somewhat differently. The move to a new calculation methodology is effectively a transition to a new index – so we have already taken the step that’s being discussed globally. I think this has been a great initiative.

The hangover we are dealing with is around the manipulation of rates. The banks are justifiably nervous about being accused of rate manipulation, with the consequence that nobody seems to know whether it is ok to transact in the market. The rules of conduct for what is and is not a legitimate transaction are still not clear. No-one wants to end up in court defending the legitimacy of a transaction.

Our goal is to use the AONIA-linked FRN to replace our CP programmes. We want to issue A$500 million in each quarter for a year – to have A$2 billion outstanding – and we hope to roll A$500 million every subsequent quarter. This will improve our control around liquidity management and profiling.

We have clients that take monthly-reset floating or overnight-rolling loans. A quarterly FRN benchmarked off BBSW re-embeds credit risk and basis risk into our book.

This doesn’t mean BBSW will disappear. It just means there is a clear economic rationale to have a risk-free rate that is more actively traded and broadly adopted in the market. It is not optimal for a riskless transaction, like a cleared derivative, to have its net-present value be dependent on the funding cost of an individual bank. The two-rate solution makes a lot of sense for Australia.

It has to be in conjunction with everyone else, though. We can only trade what counterparties will make markets in.

Given asymmetry in derivatives exposures, the impact of benchmark transition on contracts and markets more generally may be substantial. But the consequences will be commensurately less in Australia.

It is very positive to think we could establish a basis-swap market between an offshore risk-free rate and BBSW domestically. Everything we’ve seen so far suggests this could develop without too much trouble.

Things you should know: This information has been prepared solely for information purposes and is not to be construed as a solicitation, an offer or a recommendation by the Commonwealth Bank of Australia. The information may be incomplete or not up to date and may contain errors and omissions. Any projections and forecasts are based on a number of assumptions and estimates, including future events and contingencies, which may be inaccurate. It must not be relied upon as financial product advice and is not Investment Research. As this information has been prepared without considering your objectives, financial situation or needs, you should, before acting on this information, consider its appropriateness to your circumstances and if necessary, seek the appropriate professional, including taxation advice. We believe that this information is correct and any opinions, conclusions or recommendations are reasonably held based on the information available at the time of its compilation but no representation or warranty, either expressed or implied, is made or provided as to accuracy, reliability or completeness of any statement made. Commonwealth Bank of Australia ABN 48 123 123 124 AFSL and Australian credit licence 234945.