Does the 2014 Budget and fiscal policy regime set a new course for the CGS market? – CEO presentation to the annual Australian Business Economists luncheon
3 June 2014
Presentation delivered by Rob Nicholl – CEO, AOFM
Thank you once again to the ABE for allowing me the opportunity to address the forum. As I have done over the last couple of years I will use the opportunity to share with you something of how we think about issuance patterns, portfolio development and our investor base.
As I started to draft this it was the day before the Budget and the air of expectation was probably more about what could still be announced on Budget night, rather than how the many prior indications of new measures could be implemented. It seems to have become tradition for Governments to do a lot of their thinking ‘out aloud’ – maybe this gives a good idea as to what is in store for them when they release the details of actual decisions. Not that this should be seen as a bad thing – particularly as this approach creates an opportunity to adjust and refine policy decisions with the benefit of commentary and feedback. And it considerably expands the information set and advice on which decisions are made. Given that Government can be seen largely as in the business of making judgement calls, the art of success comes down to having clear objectives, appropriate settings to filter the advice and feedback, and an effective means of communication.
What say the AOFM?
While it is certainly true that we keep detail regarding specific market developments pretty close to our chest, I would argue that there should have been no real surprises over the past few years given the discussions in which we have engaged and the general indications given by us.
If I look back over what we have said and what we have done – there is a relatively strong correlation between the two. I would also argue that there has been a fairly close correlation between the types of market developments that have been suggested to us over the past couple of years and those which we have undertaken –although this of course is not to be interpreted as us having been long everyone else’s ideas and short our own.
It remains the case however that we try to strike a balance between signalling intentions for market development and not saying things that will lead to specific short-term market responses. Although difficult at times, we have tried to maintain a ‘no surprises’ approach because we take the view that CGS investors value transparency, regular patterns of behaviour on our part, and a reasonable degree of predictability to the changes we introduce. It shouldn’t come as any surprise that much of our investor engagement revolves around these themes.
However, I will acknowledge that at times we have been more subtle and less direct. This is in large part because our objectives have needed to be revised in the face of increasing issuance programs and this has to a degree required periodic re-thinking of the portfolio management strategy, in turn pushing us to create suitable operational flexibility. Sometimes this is interpreted as us being vague, but I can assure you that nothing could be further from what we are seeking to achieve
One exception to the ‘no-surprises’ category may have been our recent decision to launch the 2018 linker. A new 10-year line introduced just over two years ago would at the time have been seen as a substitute for a shorter line – and yes, at that time it was. Since then we hadn’t given much indication that it remained a new maturity option. But recently we were prompted by an intermediary to reconsider it, although we realised that the opportunity to introduce a 2018 TIB was all but closed. Therefore, to use the opportunity we needed to act rather than wait longer. Having previously heard the arguments from investors in favour of it, and nothing overwhelming against it – we took the decision to launch at short notice. The idea of buying back the 2015 had been under consideration for some time and we had really just needed for an alignment of circumstances on our side to trigger that. The two decisions were independent, although I understand that may not have appeared to be the case.
Turning to the question posed by the topic of today’s presentation, it is useful to look back in order to form a view about what might lie ahead.
Australian Government Bonds on Issue
You will be familiar with a chart such as this which shows a relatively quick growth in the size of the CGS market since the beginning of 2008-09 [the dark bars represent nominal bonds and the lighter bars linkers]. The amount of CGS outstanding is today at around $328 billion, and taking into account a maturity and further issuance for the remainder of this month we will see in the new financial year with a market about 6 times larger than it was prior to the GFC. We will finish this year with an issuance program in gross terms of around $87 billion – $63 billion of that reflecting net new issuance. Just prior to the GFC our issuance program was $5 billion.
This highlights how the financial world and the CGS market have changed in a relatively short time. Today the CGS market is seen as one offering depth and liquidity – so on the one hand this growth in CGS outstanding has increased Australia’s exposure to sovereign debt, but it has come with the benefit of significantly improved access to financial markets. Given this, a large part of our job remains one of enhancing the reputation of the CGS market, thereby improving our ability to achieve successive financing tasks and lowering the cost of borrowing to Government.
Putting to one side the political and policy issues surrounding Australia’s increased exposure to sovereign debt – two obvious questions arise. One relates to how we have been able to smoothly deliver increasingly large issuance programs. The other question serves to analyse motives for development of the market more generally. Together, a look at these provides useful guidance as to what could lie ahead.
Both have several dimensions to their underlying explanation. To answer the first question we should look outward rather than inward, and an instructive place to start is the steady rate of global foreign exchange accumulation.
Growth in Global FX Reserves
From this chart we can see that global reserves had been building steadily since early in the 2000s – levelling during the GFC, before taking off again at an even greater rate of accumulation in recent years. This chart also shows that it has been the developing and emerging countries in which the bulk of these reserves have been accumulating. This is an important indicator of where global growth has been centred in recent years. Overall and most importantly from our perspective the growth in foreign reserves has been instrumental in underpinning a growing offshore bid for CGS. The premise on which this has occurred is the increasing challenge for foreign reserve managers to diversify their assets. To many reserve managers Australia has looked increasingly attractive because of its stable triple-A sovereign rating, growing liquidity in the currency and the sovereign bond market, relatively high yields, and a growing supply of CGS. According to BIS data, trade in the $AUD as a proportion of global foreign exchange turnover was just over 8.5% last year compared with 3% in 1998.
Holding of CGS amongst Reserve Managers
World’s 50 Largest Foreign Currency Reserve Managers
Prior to 2008 there was around 8 reserve managers in the CGS market, motivated in most part by taking a position on currency or rate changes. Today there are around 40 – this representing just under two thirds of the 50 largest global reserve managers, and of the 30 largest that is around ¾ that hold CGS. They are no longer tactical investors but have made portfolio allocations into the market.
Having spoken directly to nearly all of them through our investor relations efforts we know that their portfolio allocations into $AUD range from 1-2% up to about 13-14%. Although we don’t have firm data on the specific level of reserve manager holdings, we have formed an internal view that is likely to be in the range of about half–two thirds of total offshore ownership. This tends to suggest that a large part of our offshore investor base looks to be ‘safe hands’.
I know I have mentioned this before, but it is important once again to note that many of these reserve managers are looking increasingly to seek duration (thereby trading across the curve), are currently or are thinking about making stock available through repo, or are relatively active traders as they adjust and re-adjust their portfolio positions. Our experience as we talk to investors and intermediaries is that the strong presence of reserve managers in the CGS market has not impacted appreciably on liquidity through a uniform ‘buy and hold’ pattern of behaviour.
Also as we talk to them we find them to be less focused on the macro economic outlook for Australia, the risks to that outlook, or the level of the currency (reserve managers largely holding our paper unhedged). Rather, they tend to be focused on the outlook for issuance and our mode of operation. Having seen various surveys of reserve managers that seek to reveal their core considerations on asset management, they tend overwhelmingly to be focussed on liquidity and capital preservation (hence the significance of triple‑A as a ‘safe haven’).
As many of them have triple-A sovereign mandates Australia’s rating is obviously core to their decision to hold CGS.
Budget Balance and Gross Debt
This chart indicates for at least two key measures of a low risk sovereign that Australia sits comfortably within the bands of its triple-A status given its current fiscal position, with a forecast for measurable improvement over the coming years – this of course having been presented in the recent Budget. The yellow point reflects the forecast Budget position for the coming year, with the green showing a planned improvement towards the end of the forward estimates.
The other thing to note is Australia’s relatively low exposure to sovereign debt when compared with the triple-A outliers in this regard – Germany and Canada. Add to that rating agency assessments of other criteria such as our institutional arrangements for example and this reinforces the reserve allocation decisions into $AUD.
Sovereign Bond Yields and Credit Ratings
Australia vs. OECD Countries – 10 year Bond (Inflation Adjusted)
And as this chart shows, on top of that there is a yield pick-up in our market compared with other triple-A sovereigns.
But it is not just reserve managers who have found the CGS market increasingly attractive. While there has for some time been a solid representation in our market of fund managers, including insurance and pension fund managers, this latter group are not only building further interest, but seeking longer duration. This has helped in our efforts to increase average term to maturity of the portfolio, in most part through lengthening the yield curve and pushing issuance longer.
This brings me to the second of the two questions I outlined earlier – our motives for market development and what we have done with the opportunity.
This chart shows the weighted tenor of new Treasury Bond issuance for the last 10 years divided into three time periods. It reflects changes to the average term of our issuance flow– not the stock outstanding, although the latter is obviously impacted overtime by the changing tenor of new issuance.
The first period covers maintenance of the market during which outstanding stock remained at around $50-55 billion. As you can see, the average tenor of issuance fluctuated significantly throughout that period – this being driven by the fact that the yield curve was at its longest around 12 years, with periodic new maturity issuance aimed at supporting the 10 and 3-year futures contracts.
The period 2008 – 2011 reflects our issuance response to the impact of the GFC, with increasingly large programs relative to the size of the market. During this period and prior to lengthening of the yield curve, the average tenor of issuance declined sharply as the strongest demand for CGS was seen in the short-medium part of the yield curve. In the period during late 2008 to mid‑2009 the amount of stock held by offshore investors didn’t change very much, with the result that as issuance was ramped up, the proportion of offshore ownership in CGS declined. Shortly thereafter existing investors began to increase their holdings and some new investors began to appear in the CGS market. This marked the beginning of the steady expansion of the offshore investor base, particularly the reserve mangers.
In 2011 we began to extend the yield curve with the establishment of a 15-year maturity (the 2027). For the next two years we pursued the objectives of building existing bond lines in order to promote liquidity and consolidating the 15-year benchmark. In our view a gradual approach to this has been rewarded with new entrants (mainly in the form of real money investors) whom have been comfortable with seeking longer duration than had been the case with typical CGS investors. At the same time a number of reserve mangers that had become familiar with the market began to follow us as we started to increase the proportion of our issuance into longer dated maturities. As a result, over the last three years we have been able to increase the average tenor of new issuance from just less than 6 years to around 9 years. This has allowed us to increase the average term to maturity of the overall nominal bond portfolio from 5 to 6 years purely through our issuance programs.
The Budget casts an outlook for steadily declining net issuance as the task of fiscal consolidation continues, although the path to fiscal surplus now planned is more gradual than when we contemplated this same issue several years ago. For us, one of the main issues to arise from this revolves largely around what size to expect for the CGS market when annual net issuance is at or approaching zero. This will colour our thinking on how long the yield curve should be, how many bond lines we think is appropriate, and whether we will be able to continue to support deep and liquid bond lines across the curve with declining gross issuance programs.
Within AOFM we are confident that for now the benchmarks should be the 3 and 10-uear futures baskets and a 20-year maturity. For us, the significance of a 15-year maturity point on the yield curve has been overtaken with the launch last year of the 2033 bond line. This means that as we move in the near term toward a more established pattern of launching new maturities the focus will be on maintaining a 20-year benchmark – which we will look to achieve by periodically issuing slightly longer than a 20-years – and stable market conditions will likely see us start this over the coming 12 months. But establishing this overall pattern will help us to think about further curve extensions and how they might be implemented.
At the same time, we will look to launch intermediate maturities to support the 10-year futures contract, similar to the way in which we have launched the 2024, the 2025 and the 2026 lines. However, with a 2027 already in place we will not have the need to do this in the coming year.
When the Budget is returned to surplus and we are faced with relatively fixed gross annual issuance programs, further extending the yield curve would necessarily require the diversion of issuance from other parts of the curve. Establishing new in-fill maturities would have the same effect. Therefore, our outlook for the shape of the portfolio is heavily dependent on our outlook for the outright size of the market.
If the Budget forecasts as they currently stand are realised, then the combined TIBs and Treasury Bond market will peak at around 24 per cent of GDP, with our expectation being that the proportion of TIBs to nominal bonds will look similar to what we see today.
We also took the step this year of re-introducing a second within year maturity – this being the October 2018. This was decided on the basis of the MYEFO projections, which suggested that even a yield curve longer than 20-years would have required short to mid-curve bond lines to reach very large levels. Faced with this our thoughts focused on: (1) the task of managing liquidity in the market when very large lines matured; (2) whether investors would adequately reward us for any additional liquidity that might arise in these very large lines; and (3) whether there was an opportunity to create a bit more diversity in the market with new mid-curve issuance points.
In the event it looked the most practical to create these new lines where there is strongest demand for CGS and this came down to a choice of additional 3 or 10-year futures basket lines. Having already created shorter additional lines during the period when issuance was increasing quickly and the yield curve was shorter, we decided that this was a strategy we could easily re-implement. We plan to repeat this again in the coming year with a new 2019 line and then again in subsequent years. However, this strategy will be our most flexible lever in the face of stable long-end demand and declining overall issuance.
Treasury Bonds – Distribution of 2013-14 Issuance
Looking back over the year you can see from this chart where we have directed our issuance. I have also highlighted how we segment the yield curve as the basis for developing an issuance strategy for the year ahead. Although we don’t make these strategies public in advance of our program, for this year (i.e. 2013-14) we had a strategy that involved, amongst other things, issuing at least 50 percent into the two longest buckets (which is currently sitting at 52 percent) and nothing into the very short-end. You can also see from this chart how we are focused on building new and smaller lines relatively quickly: (firstly) because we believe that this helps to generate liquidity in these lines, expecting they should trade better (all other things being equal); and (secondly) it helps to more quickly dissipate the new issuance premium investors expect when new lines are established. While we can’t predict market conditions for the coming year, you can expect that our overall pattern of issuance for 2014-15 should be similar in nature as to what we followed this year.
While it is clear from these last two charts that we have both extended our nominal curve and shifted the weight of our new issuance longer to meet portfolio objectives in light of market conditions, I want to talk briefly about our thinking on linkers.
Since 2009 we have tended to use TIBs as the lead for our yield curve extensions. However, I wouldn’t assume any longer that our longest line will always be inflation-linked. Leading the way with linkers was a strategy that has held us in good stead over the last five years. But with a continued lack of domestic liability-driven investors seeking very long indexed duration in large volume, we are at a point where further extensions are less obvious than they have been to date. While I don’t want to dispel the idea that the linker curve may at some point be maintained beyond 20 years, either for portfolio management or policy reasons, at present it isn’t a high priority for our portfolio management objectives, given the apparent cost. That said, we will continue to look for obvious opportunities for further development. Finally on this, we don’t see a rebalancing of issuance across the linker curve so as to meet domestic real money demand as being inconsistent with our desire to maintain our nominal issuance at the recently-established longer tenor weighting.
One definite change we should advise for the coming year is that we will be re-engaging more consistently in Treasury Notes issuance. About 18 months ago we announced that we would be reducing our reliance on Treasury Notes as one of our primary means of managing the mismatch between receipts and outlays in the Government’s cash portfolio. We took this decision in the face of a global trend of reducing reliance on short-dated maturities in order to mitigate re-financing risk. Looking around we see some debt managers rolling over very large amounts of short-dated financing. This practice increases the frequency of being in the market to issue new paper and creates a vulnerability to market dislocation or issuer specific events. Because we had been far from heavily exposed to this type of risk our previous decision was based on prudency rather than foreseeable need. However, having gradually increased our term issuance to reduce the need for Treasury Notes, something emerged that we hadn’t adequately anticipated. As the supply of Treasury Notes diminished there were increasing representations from a range of investors who relied heavily on very short-dated maturities to manage their duration targets. As we weren’t issuing bonds with maturities less than around 18 months this has created increasing difficulty for them.
Taking into account the strong underlying demand for Treasury Notes, we have reviewed our previous decision on the basis that we don’t face an appreciable re-financing risk – particularly in this segment of the market – and that there is an important service component to a core part of our investor base from maintaining regular Treasury Note issuance. Therefore, we will gradually move back towards a greater reliance on Treasury Notes as a cash portfolio management tool and this will be accommodated by a slightly lower volume of announced term issuance over the coming two years than would otherwise have been the case. But while we will be issuing more Notes we are not re-committing to a floor in the total amount of them to be outstanding, given our previous attempts to generate liquidity by doing this were, by all accounts, relatively ineffective.
In drawing to a conclusion I would like to make the following points and observations.
First, the policy decision to maintain a sovereign bond market in Australia has passed a significant test in so far as our ability to quickly and in an orderly fashion engage with investors to achieve substantial net new issuance.
Second, we have been fortunate to have faced the task of increasingly large issuance programs during years in which global reserve managers have been heavily active in a search for high quality paper and new currencies in which to diversify the management of a large and growing global accumulation of foreign exchange reserves. Without doubt this has made it possible to see increasingly large CGS issuance programs absorbed smoothly and readily into the market.
Third, a focus over the past few years on developing the linker and nominal bond markets, rather than seeking to diversify our issuance into a wider range of new instruments, has been recognised with broad investor support. In many regards our recent syndications, with very strong domestic and offshore investor responses, are a good indication of this.
Fourth, the CGS portfolio has been developed in a way that is mindful of its role in facilitating broader domestic financial market development in Australia. Longer linker and nominal yield curves, together with deep and liquid bond lines have enhanced the role of the sovereign bond market as a cornerstone of the Australian financial market. The opportunity has also been taken to lock in historical low rates of borrowing while spreading the overall refinancing profile of outstanding CGS in a manner that will make it easier for successive Governments to manage maturities as they roll forward.
Fifth we are now at a point where we are unlikely to see new reserve manager demand for CGS emerging at the rate seen over the last four or so years. This means that we will need to engage a wider range of investors if our issuance tasks continue to be as large as we have experienced in the last two years in particular. While we continue to see new interest from fund managers it seems unlikely this emerging demand could play the role played by the arrival of a growing number of large reserve managers to the CGS market. In this regard the Budget as announced sets the tone for a manageable decline in net new issuance. In turn I think this will refresh the question of an appropriate size for an ongoing CGS market. While not something to be addressed immediately investors will no doubt take a growing interest in this as the Budget position returns closer to one of being balanced.
Finally, there are still a few years ahead in which new CGS market developments will be a key consideration of the AOFM’s task. While to date our efforts in this regard have not been exclusively driven by engagement with investors and our intermediaries, they have most certainly been productively informed by them. Therefore we see ahead of us at least another two years of heavy investor relations engagement and this, together with the outlook for the size of the CGS market and how we can support that during times of very low or no net issuance, and the feedback we receive in support of further extensions of the real and nominal yield curves, will underpin our approach not just for 2014-15, but for several years hence.
Last updated: 19 August 2014