The changing issuance and debt management landscape – what matters most next? | Australian Business Economists luncheon, Sydney
29 May 2018
Presentation delivered by Rob Nicholl – CEO, AOFM
Thank you once again to the ABE for the opportunity to provide an AOFM update. For the year ahead we are as usual looking to find an appropriate balance between detail to satisfy investor desire for transparency and predictability; while retaining sensible flexibility to deal with changing circumstances as they emerge.
In some respects this has become harder over the past few years as changes to the cash portfolio have been more nuanced and the cash position can add a buffer to changes in the financing task. We have also broadened our operational tools with the regular buybacks for example having created some confusion about what gross issuance means. In addition, increasing opportunities to either engage in or unwind pre‑funding have become more significant. It is a while since changes to issuance have directly aligned with announced changes to the financing task and this has required more thought into how we communicate with the market after official Budget updates.
As the title of this presentation suggests, the operating environment is continually changing – but change since the GFC has come at a modest pace.
From 2011 until mid-2017 the AOFM focused on how to achieve growing issuance tasks and planning for commensurate growth in the debt portfolio. Improving our understanding of the investor base was central to this task. Our thinking on these issues has been revealed in detail over past years but in summary we have: (1) complemented the regular tender program with a refined use of syndications; (2) extended the yield curve; (3) raised the average term-to-maturity of the debt portfolio; (4) introduced regular buybacks; and (5) maintained intensive investor engagement. We also thought it prudent to run high cash balances most of the time while reducing our reliance on Treasury Notes.
From next year the focus will be on: how to support interest from a wider variety of investors but with declining issuance and reviewing our approach to managing the cash portfolio.
But it is not just declining issuance that will influence the outlook. We must also bear in mind that global financial markets are in an unusual place, with reduced synchronicity amongst major central bank monetary policy settings. In this regard we can’t ignore the potential for a marked recalibration of asset prices in both absolute and relative terms as global QE programs are slowed, brought to a conclusion, or unwound. There are also recent signs of a resurgence of risks to fiscal and financial stability in parts of the world that notably influence global financial markets.
What we as an issuer can specifically do in anticipation of these changes is difficult to know because we are unsure as to how this might all play out. I also think we have less influence over investor decisions than we would like to believe. But changes to the financial market landscape remain central to our assessments and viewing the world through a risk prism seems to have worked well. This is something we plan to continue.
Our best defence against sudden, large and unanticipated changes to market conditions can best be described in one word – ‘options’. Having benefitted from a diversified investor base, built an ability to weather unanticipated restrictions to raising finance, developed a range of issuance points, and expanded our operational tools, it makes little sense to unnecessarily give up room to manoeuvre.
The difference in global monetary policy settings I mentioned has in most part arisen from a marked divergence in economic growth rates. This is creating new challenges for investors that in turn will impact patterns of demand for fixed income assets.
Despite the Australian sovereign bond market having risen in status compared to its role prior to 2010 (and in particular it is seen as a liquid safe haven with relatively attractive returns), it is naïve to think that demand for AGS will be unaffected by changes in yields across markets and the potential for increased volatility in the $AUD.
The fact that about half of the AGS curve is trading through US Treasuries seems already to have had some impact. With a market consensus for the Australian cash rate to remain steady, and the likelihood of the Fed Funds rate continuing to rise, it is easy to imagine negative spreads persisting.
As to the level of demand overall, we expect this to be broadly stable, largely because of diversity to the investor base. The fact that our paper trades freely across the world and many investor types suggests that a market clearing level of yield should support continued aggregate demand. This will facilitate the smooth absorption of future issuance, although periods of turbulence cannot be ruled out. What we can’t know, however, is the extent to which yields or cross-currency rates will change, or the precise timing for heightened levels of disruption along the way. But given the strong prospect for both, we feel justified in having reduced the Government’s exposure to higher and more volatile borrowing costs through increasing the average term-to-maturity of the debt portfolio.
Interpreting the outlook has its challenges though due to the increasingly complexity of the channels and mechanisms through which changes to investor behaviour are revealed. This means that simply tracking movements in yields will not reflect the full extent of changes in investor behaviour. For example, we cannot ignore the importance of currency and basis levels for offshore investors.
To this end we have put considerable work into developing a suite of indicators that will help us to more fully understand changing patterns of investor behaviour. Together with the information we get via feedback from the banks – and in many cases from investors directly – we are satisfied that our knowledge of the AGS market is as good as it has been. But while we can’t act specifically on all the information to hand, it is important context to our considerations.
Domestically, we have noted increased real money buying over the past few years, together with strong buying from the bank balance sheets. As a result, the domestic investor base is now in volume terms about as important as the offshore investor base.
However, the heavy balance sheet buying seems to have abated recently, as portfolio adjustments required for the Basel III liquidity rules have been bedded down. Indeed, we note that bank ownership of AGS levelled out last year with ABS National Accounts data showing a modest reduction in the last quarter, although holdings of other HQLA assets remained steady. We can’t be sure whether this adjustment was coincident with, or a result of, an announcement last year to increase the Committed Liquidity Facility, but we will watch with interest for further changes.
I touched on this in my opening comments but a challenge ahead for the AOFM will be our ability to maintain credibility with the market while stepping back from the ‘exactness’ we have become expected to give. We are aware that investors will judge us on the basis of transparency and consistency, which means that the Australian taxpayer will ultimately benefit from the AOFM behaving in a manner that is widely understood and conducive to a liquid and stable market. Often said but worth repeating – we do not underestimate the potential for an issuer to be punished if it does not acknowledge the task of borrowing in the context of investment alternatives. Sovereign issuance is a long game but should be approached with a close eye on the risks to government and the cost to taxpayers.
The most recent example of the challenge of messaging to the market was the Government’s purchase of the Snowy Hydro shares.
At around $6 billion, this is a financing change sufficient in size to warrant market interest in terms of changes to planned issuance, and so not something we could have remained silent about. On the other hand, the change was announced in the lead up to the Budget when we were being updated on financing forecasts, both for the remainder of this year and for next year. In addition, we were carrying a stronger than expected cash position from a better than forecast Budget outcome for last year, pre-funding from some unexpectedly large syndications, and strong revenue receipts since MYEFO. Announcing that the purchase would be funded over the remainder of the calendar year, rather than being specific about fiscal year changes reflected a balance between wanting to maintain regular issuance at recent levels, taking the financing change into account, and trying to inform the market with as much accuracy as possible without creating the need for multiple advisories, or revealing information ahead of official updates.
Another contributing factor is that operationally the difference between 30 June and 1 July is an arbitrary reporting point and the closer we get to the former, the more we are taking into account what will happen after the latter. As I have said many times before to investors – no one has as much information about the Government’s issuance needs as we do – including the possibility for these needs to change. Therefore, it will often be difficult for others to know what judgements lie behind our advice to the market on issuance.
On this note, I can already see a period of confused debate ahead as the Budget returns to surplus but the wind back in issuance may not match broader expectations of a diminishing need for government borrowing. While to us business as usual, the need to fund large maturities and planned buy backs may not be widely appreciated. Taking into account a range of investment related outlays, such as infrastructure commitments, could add to the confusion.
As I also mentioned, the AOFM has run up precautionary asset balances in recent years when it felt that a temporary reduction in access to the market was possible, while also providing the capacity to pare back the issuance rate when markets were under stress. Similarly, the AOFM can reduce precautionary asset balances. In fact with an outlook of declining issuance and the benefit of having observed resilience in the AGS market over periods of global volatility, this is something we are currently considering.
Once the Budget has returned to continued surpluses there will be a policy question as to how the surpluses should be applied. While not something the AOFM can give specific guidance on to investors, it will eventually raise a related question as to what is an appropriate long term size for the AGS market.
Fifteen years ago the decision was taken to maintain the Australian sovereign bond market at around $50 billion, despite a run of budget surpluses that would have eventually enabled gross debt to be fully repaid. The cash arising from these cumulative surpluses ultimately formed part of the Future Fund seed capital – in itself a major debt management policy decision. This was a period of reducing net debt (which soon thereafter became negative) while maintaining a discretionary level of gross debt.
Today of course asking the threshold question of whether to maintain a sovereign bond market for Australia seems a long way off. At the earliest, we could expect a decade or more to pass before the topic is even relevant. However, issues relating to how the sovereign bond market facilitates the development and efficiency of the Australian financial market generally are already relevant. In fact recently there has been commentary on opportunities for the corporate sector to extend its access to domestic debt capital markets and it is difficult to imagine this not being facilitated at least in part by a well-functioning sovereign bond market.
Australia will not be the first country to have this debate, and in fact as I have just noted we have been through this before. The experience of that debate, together with our deeper understanding of how Australia interacts with global financial markets, and looking back to the task of financing a rapidly emerging deficit in the midst of the GFC, should prove useful when informed judgement is required.
Let me now turn to some specific comments on the program for the year ahead.
Total gross issuance will again be lower next year highlighting 2016-17 as the peak year for issuance.
For the nominal bond program of around $70 billion, which includes issuance to fund buybacks, the focus will be on issuing to maintain the average term-to-maturity of the portfolio and to meet demand. The three and 10-year futures baskets are where we tend to see most demand, but we will also be looking to support the 20-year futures basket.
The nominal program will involve establishing a new 5-year maturity, two new 12-year maturities; and a new 23-year maturity. The timing of each remains open except for the first of the 2030 maturities, which has been announced for a tender next week. The tender volume has not been settled because we are seeking feedback on this but in any case we are mindful of needing to give space between building this line over the coming months to at least $8-10 billion and then establishing the next one. The second 2030 will probably be established by syndication but not until the second half of the fiscal year.
The need for two maturities per calendar year for the three and 10-year futures contracts has been carefully considered and will undergo even closer scrutiny before the issuance program is settled for 2019-20. The reasons for introducing these additional bond lines some years ago are falling away as issuance is forecast to decline. Furthermore, given we have made much of our aim to support liquidity by restricting the number of lines and building them to sizeable liquid volumes, I don’t see that we can abandon this principle just because issuance has reduced. This means we are not going to be fixated on particular patterns of bond lines but will continue to balance debt portfolio considerations with what will be functional for the market.
Apart from the new maturities, non-basket issuance could be relatively light during the year, depending on demand.
As to the new 2041, this has now (unfortunately) been announced several times and I am aware that this may have eroded rather than reinforced our stated reputation for doing what we say. I am hesitant to indicate timing for it other than it is planned for syndication some time later this calendar year.
While syndicated taps in this part of the curve have been successful, we are unlikely to use these again for 20-year futures basket stocks until performance of the new 2041 bond has been firmly established. The taps undertaken this year, together with tenders, have given us confidence to call the ultra-long part of the curve as established. Therefore, how best to achieve ongoing maintenance of the long-end will now become the focus.
While we will continue to look for opportunities to use syndicated taps, they can be planned at short notice so long as print volumes remain modest, and so there is no need to treat these announcements differently to the way regular tenders are announced. As an aside, the relatively flat yield curve and current estimates of a near zero (or even negative) term premium are indicators from a cost perspective that reinforce the value to us of looking for ultra-long issuance opportunities.
This brings me to a quick word on 30-year bonds. Feedback clearly suggests that investors prefer liquidity to the number of available lines. To this end, our current thinking is that four-year maturity gaps will be appropriate, in turn suggesting that a 2051 will be the next 30-year benchmark bond. We are committed to establishing a new 30-year benchmark bond but are not planning to do it prior to July 2019.
The buyback program has now been in place for two years and has successfully achieved its aims through: (1) assisting cash management; (2) increasing portfolio duration; (3) supporting bond trading efficiency; and (4) assisting the RBA in its liquidity management task.
Repurchasing short bonds commences as they exit the three-year futures basket. As I said prior to us launching this program, we will need to monitor buybacks to avoid creating undue shortages of stock. This means that there are no set volume targets to meet but we will continue to give the best guidance possible as to our forward expectations.
With the benefit of experience we have settled on a rule of buying back short bonds until they are around 9-10 months out from maturity. We will explicitly notify the market when we are to stop buying a particular line. The RBA will continue to buy the bonds from this point if that suits its liquidity operations.
Buybacks may play a different role in future years as the funding task continues to fall, but a continuation of our current approach will be maintained for the foreseeable future. Any decision to change this will be announced well in advance.
Buybacks of linkers are a different matter and we are not planning a regular program for these. They will continue to be advertised just ahead of appropriate opportunities.
In terms of linker issuance for the coming year it will again be around $5 billion. This volume may seem relatively static but when gross issuance for this year was revised down, all of the reduction was to nominal issuance. This year the announced gross program is lower again, with the reduction also coming from nominal bond issuance.
A new long dated linker has been announced and this will be a 2050 maturity. The aim is to satisfy domestic investor appetite for ultra-long linkers, although we are not ruling out the prospect for meaningful offshore interest in this new bond. The appropriate timing for this is harder to determine but we are aware that active discussion of a linker curve extension has been out there for some time. Furthermore, now that it has been confirmed we are also aware that speculation as to when it might be established will tend to hang over the market. While there are competing considerations around timing we have decided that market conditions permitting it will go ahead by syndication some time during September-November.
Before concluding I would like to offer some additional comments on the investor base – something we are often asked about.
As I mentioned before, there has been a gradual rebalancing of the investor base from offshore to domestic ownership and this has been happening for some years. For those who follow our comments it will be clear that this has not involved persistent net selling by offshore investors, nor can it be attributed to a consistently slower rate of buying. Accumulation by offshore investors has continued at a steady pace, but the rate of this has not matched the rate of issuance, particularly since mid-2014.
It was around the end of 2014 that the rate of central bank entry to the market began to slow and up to that point central bank accumulation had played a highly influential role in absorbing post-GFC issuance. The role of central banks remains important but interestingly after a few years of little change in the number of participating central banks, we have in the last year become aware of the potential for some new ones to join the market. While central banks differ in terms of their style of engagement, we view them as a relatively stable subgroup. They tend to hold AGS to cover their currency allocations and therefore, liquidity and capital preservation dominate their investment decisions, while outright yield and spread are less relevant.
Then there are the passive offshore and domestic fund managers, which are an increasingly large group of AGS investors. They need to match benchmarks and AGS constitute about 2 per cent of global indices. They are less focused on trading relativities than active fund managers and hedge funds. Local passive fund managers also grow their holdings as the proportion of AGS within the local indices continues to grow.
Of the remaining offshore investors, there are many that focus on currency-hedged returns. The outright yield spread is less a focus for them too although hedging costs are highly relevant; many Japanese institutional investors for example are highly sensitive not only to yields, but also to currency volatility and hedging costs.
Then there are investors that rely on funding their positions through access to repo. The recent changes to the cost and availability of repo is therefore something that could have a more lasting impact on these investors depending on whether the recent market changes turn out to be structural.
Extending the yield curve to include a 30-year benchmark has had a positive influence on investor diversity – particularly from the offshore sector. How this further develops will be interesting and it is something we continue to monitor with close interest. We have recently learned of the potential for new industry sectors from offshore jurisdictions to increase their interest in ultra-long AGS. Should these developments gather momentum this will provide a useful level of new support for long end issuance.
The domestic investor story I have touched on earlier, particularly regarding the bank balance sheets, which now hold just under 20 per cent of AGS. Furthermore, we cannot overlook domestic investor interest in the linker market and we have spent considerable time looking to understand the aims and investment mandates of that investor cohort.
There are continual changes to the pattern of investor behaviour in AGS and it has often been difficult to summarise constant shifts across the domestic/offshore divide, across broad geographic offshore coverage, and between different investor types. This also makes it difficult to anticipate changes to buying and selling patterns in response to changing market conditions.
Let me conclude by saying that the challenges ahead are going to be different to the ones we leave behind. The fiscal outlook is positive, which means the pressure that was building on issuance has passed. The AGS market has been resilient through periods of global market volatility. What the year ahead holds in this regard is anyone’s guess but I am confident we are well-placed to deal with a wide range of possibilities. Managing market expectations and investor demand while consolidating our portfolio management goals, during a period of marked issuance decline, is a good challenge to have.
Last updated: 29 May 2018