On a Gilt trip — adaptable DMO stays close to its market
To help celebrate the 20th anniversary of the setting up of the UK Debt Management Office, GlobalCapital gathered together some of the UK Gilt market’s leading traders, investors and bankers, with its CEO Sir Robert Stheeman, to discuss the state of the bond markets, how they have changed since the DMO was established in April 1998 and how the government’s investor base and bond issuance are likely to evolve over the coming years.
The discussion was held on May 16, just a day after the UK sovereign’s most recent syndicated bond issue, the £6bn 2071 Gilt. That deal, despite having a 53 year maturity, a coupon of 1.625% and having been sold amid fractious Brexit negotiations and a brewing political storm in Italy, blew out, attracting a book of £37bn.
That strong demand appeared a confirmation of the DMO’s careful approach of combining syndicated issues with auctions to maximise demand and liquidity for Gilts, as well as associated derivatives.
But investors also have stuck with the UK, despite the shock of Brexit.
The way the DMO manages debt issuance is likely to continue to evolve, in unforeseen ways, but topics on the roundtable participants’ radar included the syndication process, prospects for issuance linked to CPI inflation, green bonds and 100 year maturities.
Participants in the roundtable were:
(Click on hyperlinked names to view video interviews)
Mark Button, head of sterling inflation trading, Nomura
Kari Hallgrimsson, head of sterling rates, JP Morgan
Lloyd Harris, fund manager and lead financials analyst, Old Mutual Global Investors
David Henness, head of flow rates trading EMEA and APAC, NatWest Markets
Craig Inches, head of rates and cash, Royal London Asset Management
Peter Millington, head of rates trading, Santander
David Parkinson, head of UK rates sales, RBC Capital Markets
Sir Robert Stheeman, chief executive, UK Debt Management Office
Guy Winkworth, head of European inflation and sterling rates trading, Morgan Stanley
Toby Fildes,moderator, GlobalCapital
GlobalCapital: With the UK having issued a £6bn 53 year bond only yesterday, that raised £37bn of demand, it would seem churlish not to start with that. Sir Robert, how did it go?
Sir Robert Stheeman, UK Debt Management Office: I think it went very well. It’s the longest bond, now, that we have out there. In particular, the amount of duration which we were putting into the market was very, very significant, given the low coupon of 1.625%. It’s worth comparing the 2071 with the 2068, which was priced with a 3.5% coupon. There’s a big difference there, and that implies a lot of risk for the market to absorb. But it was a very stable market.
Genuinely — and I would say this on most of our deals, but I think yesterday in particular — all credit to the lead managers and all credit to the duration manager.
We think it went well. We also feel that it was good for the market, and that’s actually quite important from our perspective. Yes, we want to get a good price. Yes, we believe we got a good price, but we also want to make sure that these transactions work well with the wider market.
We got the strong sense that there was a lot of unsatisfied demand still out there but these things are not always straightforward. They’re difficult for the leads when they have to make the allocation. I’m aware there are investors in the room when I say this, but it’s also important that the performance of the issue immediately afterwards really is good and that these investors have the sense that they want more rather than less.
GlobalCapital: What did the underlying demand and execution tell us?
Stheeman, DMO: The underlying demand came primarily from where we would have expected it— mainly from what we refer to as the core UK domestic investor base, to a large extent the pension funds, the life insurance companies. They were the ones for whom this was primarily intended. Having said that, we also had a relatively large number of orders. Approximately 90% was domestic UK. We would expect that in this sort of maturity. But it was also a pretty diverse order book.
For us, the key thing is that the core domestic investor base, which is ultimately the mainstay of the long end of the Gilt market, were given supply, which they’ve needed, and which has been signalled to us for months now.
Demand right now is clearly at the longer end of the conventional curve.
David Henness, NatWest Markets: Sir Robert touched on it there — the strength of the syndication process. The market worked very well together, both the DMO, the investors, obviously the lead group and the other GEMMs [Gilt-edged market makers]. It shows how an issuer engaging with its investors and engaging with its market makers can lead to a pretty significant and smooth transaction.
Craig Inches, Royal London Asset Management: The size of order book is one of the big changes we have seen since the syndication process really began, in 2005, and properly in 2009. We’re seeing large books for both nominal and inflation-linked issues, whereas typically what we used to find in the nominal space was that books weren’t as large, so the allocations we received were in the region of 70% of our original order.
On the inflation side the demand for syndicated bonds was large, typically from pension funds, and the size of the deals were slightly smaller, which generally netted us a lower allocation than a nominal syndication. What we’re noticing now is that as these books get bigger and bigger, a good example being yesterday, our allocation continues to fall. We got just over 50% allocation in yesterday’s deal, which is one of our lowest ever allocations.
One thing that’s interesting is that this quarter we had [an auction of] 2057s, closely followed by the syndication. This resulted in quite a lot of concession on the curve and in the bond ahead of the auction, and when the syndication came yesterday, the market was a lot more stable as the prep work had already been done. So, yes, you can get bigger size done via syndication, but you’re now seeing more concession for auctions.
So it’s a double-edged sword. We’re not a large LDI [liability-driven investment] house, so a lot of the asset buying we do is not pension fund-led, it’s more active management, to try and add value for the end customer, either in relative value trades on the curve or versus overseas markets.
So to have that volatility around some of the supply process is quite good for an active investor, as opposed to stability being more important for a passive investor.
Peter Millington, Santander: You’ve got these massive duration events taking place in the UK — I’m not only thinking about this syndication but also the linker syndication earlier this year — and they are incredibly smooth, when you look at the price volatility in the market. The syndications are giving investors and banks a signal and a space to find the demand, to find the supply and to match it up. We’ve seen a significant drop in volatility around these big events in recent years.
Mark Button, Nomura: I’d make a couple of comments. One is about the headline size of the book. The DMO always kindly publishes how much is lead manager-related, but doesn’t publish what the rest of the co-leads, the rest of the GEMMS, have done.
So taking the £37bn headline number can be slightly misleading, because obviously real money is what matters in these deals and real money demand was obviously massive in itself. But the market doesn’t know whether it was massive to the tune of £7bn or massive to the tune of £15bn. We don’t know. So maybe having that information available is something for the DMO to consider.
The second thing is volatility around syndications. We’ve found that as LDI schemes mature, they’ve told us the syndication process is less important than it used to be, say five years ago, when they were going from 20% hedge ratios, very rapidly up to where they are now, at 80% hedged. The additional risk they need to add to the market is small enough that they can use the auction process.
Obviously yesterday blows that theory out of the water, because it was taken down in very large size. But I do think that as schemes mature, there will be less need for syndications and more for the regular auctions.
Stheeman, DMO: On the size of the book, I confess we are slightly torn on this issue. Of course the headline number is impressive — the £37bn number generates great headlines, highlighting that it’s a record and so on. But very genuinely, we do not seek to have massive order books, and actually they leave us feeling, perhaps, slightly uncomfortable. It begs the question of what happens next time if we ‘only’ have £20bn, with which I would be comfortable, but others might not see it that way and might question whether there was a problem.
We don’t make the allocation decisions. But it makes it very difficult for those people who do, that’s to say the leads, to decide. What is a real order? Even with real money, where we tend to see much less order inflation, the inflation trend is beginning to come in because people are looking to try to secure the allocations they need.
That leaves us with a sense that while the process appears to be working very well, it isn’t working perhaps quite as well for everyone. My sense is there’s something there which can still be improved on. How you do it is another matter.
Guy Winkworth, Morgan Stanley: I think that’s a global issue — it’s not just true of the UK DMO, it’s true of many of the European syndications over the last several years. As people have got more used to the process over the last nearly 10 years books have got larger and the question has become how you deal with that phenomenon. Do you either look at increasing the amount you issue, or somehow scale the different orders?
Stheeman, DMO: I’m not making this up. There have been people who have come to me and asked: ‘Given the fact that you had a £37bn book, why didn’t you do £20bn or £30bn?’ And I have to say ‘Well, it’s a long story.’
There’s a serious point behind that, which is: what is the real information content in the size of the book? To me, the success of the deal, and I do believe it was successful, was around the stability of the market and the subsequent price performance. The extent to which we were able to provide the investor base with the sort of product they need. That’s the real measure of success.
Kari Hallgrimsson, JP Morgan: I’d like to expand a little bit on Peter’s point, on how important the syndication timing is for the whole market. We were very busy on all our GBP trading books around the time of syndication. Our options books, our swap books, our inflation books and even our short end book were very busy. Not only UK focused clients trade GBP rates around this time but most major international clients get involved in the UK market in one way or another during this period. In other words, the UK becomes the focus of the global rates market for that particular time. It brings everybody together at one point, which creates phenomenal liquidity on all products, not just for the specific bond that is that is being issued.
GlobalCapital: Let’s turn to the UK economic outlook. What’s in store?
David Parkinson, RBC Capital Markets: As a house we think Brexit has cost the UK economy about 1% of GDP — maybe a bit more, maybe a bit less. And we think it will cost another 1% or so before the process is finished. But that’s so much more modest an impact than many of the predictions before the referendum and indeed immediately afterwards.
My outlook on the economy is pretty sanguine. It’s a large, modern, well diversified economy that should be able to absorb all kinds of shocks and come out strong the other side. And I think that’s ultimately what we’ll see, whether it will be a little bit stronger or a little bit weaker than it might have been otherwise. I think the UK economy is in a very good shape to get through the next two or three years.
Millington, Santander: I agree, although we are seeing some softening in the data. We think the foreign exchange effects have come through pretty quickly and that we’ll see a continued benign development of CPI [consumer price inflation]. Inflation expectations are again muted and we saw that in the May inflation report. So I feel that short term rates in the UK are going to be pretty stable for the rest of the year and that the Gilt market will be supported.
GlobalCapital: What has driven that softening?
Millington, Santander: Of course Brexit will potentially be one of the drivers, but it’s not the only one.
Lloyd Harris, Old Mutual Global Investors: We’ve benefited from the rest of the world’s growth — that has kept the UK alive through this Brexit uncertainty. But as you said, the data has come off a little, just because eurozone growth has also come off a little. From here on in I’m less sanguine. The data is not that great — just look at the recent spending data from Visa. Even in April, after the bad weather, it was turgid. So we could be in for a prolonged soft patch — and it’s going to underpin Gilts.
Hallgrimsson, JP Morgan: Peter said that short dated rates will be stable, I’m not sure about that. You currently have 50/50 chance of a rate hike in August priced in — it has to go one way or the other. I actually think that’s quite exciting. If you’re trading a bond market which has decent volatility, it can create good opportunities.
Inches, Royal London: If you look back a number of years, you had abnormally low interest rates, either from excessively low monetary policy or stimulus through QE, and the fundamental economics were just irrelevant. Nobody really cared about the fundamentals. It was all about cleaning up the balance sheet etc.
Whereas now, people are beginning to focus back towards economic fundamentals — you’ve got the governor of the Bank telling you to look at the data.
But what’s more interesting is you’ve got volatility in bond markets. You’ve got US yields at the front end close to 3%. So if you’re having to decide now between an equity or an absolute return product or front end US cash, you’ve got a choice. You don’t need to take long duration to get the same returns as you can get from going short.
Those questions are starting to play out in investors’ heads. Do I want the 2071 at a yield of 1.7%? Or do I want some sort of enhanced cash product that’s only got a duration of a year and can get me 1.5%? I might take my chances on 1.5%, based on the volatility coming down the track.
So from our client base we are seeing people shortening duration of interest rate products into shorter products, just because of that perceived volatility. And that’s quite interesting for the shape of the yield curve.
GlobalCapital: Why have we had such a strong performance by Gilts while all the Brexit negotiations have been going on? Is it simply down to the strong domestic bid?
Stheeman, DMO: It depends a bit what is behind that strong performance. Another way of looking at it is that rates in the UK have not risen to the same extent as they have in the US. Yet how long is it since 10 year US Treasury and 10 year Gilt yields were roughly on the same level?
That tells you something about the market’s present view on monetary policy in the UK. Which in itself goes back to the wider economic picture.
The interesting thing is each time the economy comes under pressure, the bond market — some people who live in the real world might say, perversely — seems to benefit from it. A little over 10 years ago, when our borrowing requirement was much lower, Gilt yields were 4.5%.
You could argue we are beginning to enter a new phase. But what has gone well is the overall issuance programme, because the right sort of demand in the primary market has held up. But it’s not something we take for granted.
Going back to yesterday’s transaction, one of the reasons why we were so pleased was that we felt it was good for the market in terms of providing a much needed liquidity point. All this is about building confidence in the market.
I’m struck by what Kari said about the amount of activity that occurs around some of these operations. That is really important to what we do.
We want to make sure the Gilt market remains relevant to international fixed income. We recognize that we are not the world’s reserve currency, and we need to know that this market functions as well as possible for most of the time.
That is the only way we can guarantee that we will be able to access the market when we need to on a regular basis. At the moment it works well, but we do not take that for granted and that’s why a lot of thought went into, for instance, yesterday’s transaction.
Harris, Old Mutual: Do you think the changes in demand over the last 10 years are now structural? Especially at certain points on the curve?
Stheeman, DMO: Yes, exactly. For the last 30 years there has been continued institutional demand in the long and the very long end. That’s what gave the UK, for the best part of 15 to 20 years (until the 2007-2008 financial year) a clearly inverted yield curve.
What shifted dramatically, almost exactly 10 years ago, was significant monetary policy loosening, leading to a reduction in yields in the short end of the market.
Everything suggests that the demand at the long end is there, but it’s more nuanced. For instance, right now demand in the linker space is not focused at the ultra-long end. It’s still long, but it’s shifted shorter. Which to me sounds logical: as some pension schemes mature, the average length of pension liabilities continues to shorten little by little every year.
All that suggests the future shape of the yield curve is going to be interesting. It’s interesting to note that in the nominal space, yields are at their highest not after 30 or 40 or 50 years. From 25 years outwards, we have an inverted yield curve.
Button, Nomura: There has been a marked acceleration in liability-driven investment flows into Gilts over the last few years, driven by a couple of things.
One is less confidence in the covenants of the scheme sponsors. There have been quite a few high profile schemes heading towards bankruptcy, such as BHS and British Steel, and that’s obviously worried consultants and trustees alike.
Secondly the availability of repo financing has meant that leverage has been much more available for LDI schemes And then finally, post-Brexit, the great performance of foreign equities in GBP terms has meant that pension funds have almost had a get-out-of-jail-free card in terms of their solvency ratios. That means de-risking has become much more attractive than it was, say, in the immediate aftermath of the Brexit vote.
Hallgrimsson, JP Morgan: We’ve talked about demand for Gilts, both outright and on curve, but it’s important to realize how well they performed on asset swap basis.
There are three main reasons for this: First of all, as Mark has alluded, there is plenty of balance sheet in the UK market right now. A good indicator of balance sheet availability is the repo market. Currently it is relatively cheap for investors to fund Gilt purchases in the repo market. That could explain why demand is shifting shorter on the curve. Investors no longer need to buy the Gilt with the highest duration, they can instead leverage their funds in repo market and get as much duration but in shorter dated Gilts. Often it is a better fit for their liability matching.
The second thing is that, despite the doomsayers before the Brexit vote, the Gilt market is the ultimate safe haven. So when we have an unclear outlook on the economy, Gilts seem to rally.
Finally, there is the transition out of Libor. What does that mean for Gilts? If you’re a pension fund and you have a choice between entering into Libor-based interest rate swap or buying a Gilt, you’ll probably end up buying a Gilt because you might have to restructure any Libor-based derivative in the future. This is one of the biggest drivers of how well the nominal bonds have performed on an asset swap basis in the last 18 months or so.
Parkinson, RBC: The original question asked whether the resilience of the Gilt market was just down to the strong domestic bid. It’s worth pointing out, because sometimes it’s under-appreciated, that while sterling is not the reserve currency, it is still a reserve currency.
One of the striking things about the Gilt market over the last 10 years or so is how sticky central bank reserve manager money is in the Gilt market. It’s been through two big tests — the financial crisis itself and the Brexit vote — and despite these events money has stayed in the UK, it has stayed in sterling, it has stayed in the Gilt market. After the referendum we had the biggest ever overseas buying of Gilts in a three month period we’ve ever seen.
Another point is there’s a bit of a supply illusion. While the DMO is still issuing over £100bn of Gilts every year, given redemptions and coupons, as a result of the size of the Gilts stock now, the DMO is actually not taking any more cash from the Gilt market now than in 2006-2007.
I’m sure Robert has got the exact numbers but it’s less than £20bn. So the DMO is not really having to find too much new money each year.
GlobalCapital: That’s an interesting point and it brings up the subject of shifting to running a surplus. How will that change things?
Winkworth, Morgan Stanley: The numbers the chancellor will unveil in November will show that there is still a large gross amount of Gilts to issue, which should keep us all busy for several years to come, I dare say!
Of course there is cash coming out with coupons and redemptions the whole time, but it’s the duration the market has to deal with on a rolling basis that I think is the issue.
Stheeman, UK DMO: This year we’ve got a £106bn Gilt programme. The gross financing requirement, which has had some adjustments made to it for this year, is £112bn. These two things are usually pretty close to each other.
Next year, current forecasts which come from the Office for Budget Responsibility, show a gross financing requirement of £147bn and in 2021, £150bn. But what is interesting is how the composition of the gross financing requirement changes because of redemptions.
This year you’re looking at a £41bn central government net cash requirement and £67bn of redemptions. Next year’s requirement is virtually, on current projections, identical: a £42bn cash requirement, but £99bn of redemptions. This is an issue for us next year and a challenge. What slightly worries us is the effect of those redemptions on, for instance, liquidity at the short end of the curve. We are very focused on that.
Winkworth, Morgan Stanley: The DMO has done a great job over the last few years of keeping the proportion of issuance very constant between various different sectors that are issued into. As you get larger, lumpier redemptions, that may need to become more flexible.
Inches, Royal London: When you look back historically to periods when the economy was doing better and we were close to a surplus or had a smaller deficit, gross Gilt issuance tends to drop.
We may like to focus on moving towards surplus but, like you say, in the next five to seven years you’ve got to refinance 50% of your Gilt book. If interest rates have risen quite significantly by then — and who knows what government we may have in by then, in terms of their fiscal outlook for the UK — we could be in a situation like the US, where you’ve got monetary tightening at the same time as fiscal expansion and you’ve got to refinance 50% of your debt. So it’s going to be a challenge.
Hallgrimsson, JP Morgan: Most discussions on the UK bond market tend to focus on the structural demand in the long end of the curve. But speaking to our clients, the central banks in particular, there is screaming demand for shorter bonds, five years and under. The redemption issue could have an even greater affect on that part of the curve.
Stheeman, UK DMO: That’s exactly what I was referring to. People forget that as recently as 2000-2001 the UK DMO did not issue a single Gilt with a maturity less than 15 years.
One of the consequences was that the short end, and the medium part of the market as well, was not fantastically liquid. That’s not good for the health of the wider market. It was, as some people might perhaps regard it, an unexpected benefit of the crisis and the increase of the borrowing requirement meant that we were able to supply the short end of the market with the liquidity it needed. That liquidity is very important and attractive to the international investor base.
Millington, Santander: There is also a demand at the short end of the curve from UK banks as we go into larger capital requirements. They need to invest this capital buffer in high quality liquid assets.
GlobalCapital: Quantitative easing kept a lid on volatility. And without QE we are going to have more volatility. How are we going to cope with that?
Winkworth, Morgan Stanley: It will depend how QE is unwound. If it’s rolling off the portfolio, from a flow perspective it wouldn’t have such a drastic effect. Whereas if it was, for example, a selling down of the portfolio at the same time as this relatively large gross issuance, that could have an extreme effect.
But I do think one of the reasons why the front end of the curve has been so much in demand and why there’s been a general shortage of good quality collateral globally at the front of the curve has obviously been QE.
Millington, Santander: We should look at the US, where we’ve got quantitative tightening on a steady glide path. There is also, as we’ve mentioned, over $1tr to be issued to finance the deficit. And there is a pretty clear movement upward in rates. I believe that the UK will be much more gradual in unwinding QE than the US has been.
Winkworth, Morgan Stanley: The redemptions are much lumpier in the UK than the US. Whereas the US has bonds and securities rolling off each month, we have something like three bonds maturing on average a year. That’s a challenge for the policymakers to come through, which presumably involves some sort of spreading the load, maybe selling out some holdings as bonds roll into redemption.
Henness, NatWest Markets: It’s going to depend on the global environment. Right now only the US is rolling off a little bit of QE assets. Japan and Europe are still buying.
If the UK were trying to unwind QE at the same time as some of these banks were moving out of quantitative easing, that would definitely lead to a lot more volatility in the market, and probably something the market has not been used to for quite a while.
GlobalCapital: Is volatility inevitable?
Henness, NatWest Markets: In an environment where several central banks are trying to unwind QE, that will challenge the market and will create volatility.
Parkinson, RBC: A lot will depend on how quickly it happens, how well it’s telegraphed and how orderly the process is. But I don’t think a normalisation of the rates market should be something people worry about. Higher yields will be good for the pension funds. Less rich HQLA [high quality liquid assets] will be good for banks. A little bit of volatility, rather than none at all, would probably be good for banks and asset managers.
GlobalCapital: Should we be careful what we wish for?
Inches, Royal London: As with Lloyd, we’re active managers of money, so volatility is good for us and allows us to add value. But volatility in fixed income hasn’t really had a huge impact on bid-offer spreads on Gilts, or linkers for that matter, for the past 10 years or more.
There’s been plenty of price movement on, say, some of the long duration index-linked assets, but you’ve still got a really tight spread, which is great for the client. But ultimately what’s driving the tightness of that bid-offer spread, I would argue, is the syndication process and the GEMMs competing for market share in auctions. So it’s good for us, but hard for you guys.
Parkinson, RBC: There’s a danger of being too market-focused. Higher interest rates and bond yields will be good news for ordinary savers. What we’ve been through over the last 10 years has been really unusual. There have been winners and losers, and I think over the next 10 years some of the losers of the last 10 years will start to have a better time.
Harris, Old Mutual: I’m not sure the market will really take it in its stride. I think it is happening in the US. The winding down of the eurozone programme is more problematic. It doesn’t feel like Europe has reached escape velocity yet and there could easily become a huge focus again on Italy, Spain, Greece, Portugal, as soon as the purchases end. I would just be amazed if the market took it in its stride.
GlobalCapital: So what’s your view on what the ECB will do with its QE programme?
Harris, Old Mutual: It might be more corporate bonds but less government bonds, perhaps. But raising the short end is probably better for Europe than any purchases — it would promote bank lending. The problem banks have in the eurozone at the moment is that it just doesn’t make sense to lend, because they can’t take the retail deposit rate below zero. So their margins are just getting squeezed and it just doesn’t promote lending.
Winkworth, Morgan Stanley: Rather like all aspects of the bond curve, QE holdings have much longer average duration in the UK than elsewhere, just because of the nature of the Gilt curve. So this becomes a slower problem if the QE exit is just a rundown, rather than an active selling down of the portfolio. It’s a much slower moving thing, which presumably is easier for markets to cope with, unlike the European case, where a huge chunk of the holdings are under five or six years.
Henness, NatWest Markets: You have to consider the long term holders in the Gilt market as contributors to that lower volatility. The LDI funds hold these bonds for a long time. Whereas if you start to get an increase in volatility, some investors in Europe and bank portfolios, for example, may need to adjust their holdings.
Winkworth, Morgan Stanley: A second function of the very large, long duration that the UK issues is that in any difficult or volatile situation, there is always a potential response from the issuer: to reduce the amount of duration that’s given to the market, if the market is struggling to deal with the duration.
Now of course, if it’s a balance sheet issue for investors, there’s only so much anyone can do. But in terms of the duration, whether it’s from shortening of liability hedging or from run-off or downsizing of QE, there is always a potential response from the issuer
GlobalCapital: How can we expect the syndication process to evolve?
Stheeman, DMO: It will always evolve. If you look at the way we do syndications now, compared to our first one in 2005, or even when we started the current programme in 2009, a number of things have changed.
We follow a format the market understands — they know, roughly, what we’re going to issue each quarter and the timing. The pitches we receive from banks are now compressed into about two or three days — sheer hell on our side!
But what it means is that the market is familiar with the process. That’s important for the investor base, so that they have plenty of time to set themselves up for a particular supply event and also that we, through consultation, through soundings from the banks, through our own discussions, have a reasonably good idea of what the demand is and therefore what the transaction should look like.
The process has also become much more efficient in terms of timing. We try not to leave the books open for too long. We try and wrap up pricing — yesterday is a good example — before the market opens in New York”?.
How will our programme evolve? It depends on how much we have to raise. If some years down the road we do not have to borrow quite so much in the long end, for instance, the programme might look different.
I remember very well, eight, nine years ago, in the early days of the programme, syndications were not universally liked by market participants.
Even now some people have concerns about the process. It’s not as neat and tidy as an auction. In an auction, price decides everything. With syndications, it’s a slightly different process. We need to have our ears really close to the ground to understand how we can improve the process and evolve as the market evolves.
I personally am pleased for the market, and therefore for us, that so far we’ve stuck with the programme, after the first couple of years when there was some scepticism and a sense that it was a very temporary measure. Let’s not forget, it was called a ‘supplementary method of issuance’.
We have never wavered from the idea that the auction programme should be the bigger focus. We have always had about 80% auctions, 20% syndications. But I do think that 20% for syndications has been good in getting buy-in from the entire market. In terms of how it will change, it is very hard to tell. If we could open at 9am and price by 10.30am that would be fantastic, but I’m not sure that will ever happen!
Winkworth, Morgan Stanley. For me it’s more about whether new types of instruments get issued — so CPI or CPIH [CPI including owner-occupiers’ housing costs] or green bonds, for example. Those are the liquidity points where the market as a whole — investors, market makers, the DMO — need to come out together and figure out how best to issue new classes of assets to get to the right investors in the right formats.
Obviously syndications are vital for things like yesterday’s trade, where there are a lot of switches to be done. You can’t do that through an auction. As the market evolves it will still be a very important part of what we do.
Hallgrimsson, JP Morgan: The DMO has found a very good balance between regular issuance and the flexibility around the syndications. The market and the DMO together have found that very useful.
It creates an incentive for market makers to continue to provide liquidity. And, let’s not forget, to turn up for all the auctions. Furthermore it give the DMO greater access to investors and market views.
The syndications in the last five years have been big duration events — they’ve been a way of placing that duration in a very efficient manner, without disrupting the market. Going forward I believe that they will be needed, not only to place large duration, but also to introduce new products such as the introduction of CPI-linked bonds.
GlobalCapital: Is CPI the preferred index for investors, and will we see CPI-linked Gilts in the near future?
Millington, Santander: There is certainly demand out there because there are plenty of pension fund liabilities linked to CPI and insurance companies are looking to get Solvency II capital relief by matching their assets and liabilities more precisely.
The challenge is how to increase the transparency of the CPI market, which already exists. There have been some developments this year that have increased transparency with some interdealer pricing of the RPI/CPI wedge. But how do you get an efficient and liquid market in CPI Gilts without undermining your RPI market, which is your main inflation market? That’s the challenge for us.
Button, Nomura: I don’t think a new CPI market would detract from the RPI market liquidity at all. In France they run an OATi curve as well as an OATei curve but the existence of both helps the liquidity of both and there are plenty of people who will trade that basis. Active managers such as Craig would be very willing and able to play the wedge trade between CPI and RPI.
The challenge for the DMO — and I’m not sure if Robert will respond to this — is to choose to issue CPI before the government has actually decided whether to come down on the side of CPI or CPIH as their main measure.
It’s quite sad, having seen [chancellor of the exchequer] Philip Hammond’s comments last week about not being able to switch away from RPI because they can’t afford it, that that’s the main hurdle awithin Treasury.
There have been issues in the past around the construction of CPIH and our understanding certainly was that they were waiting for more of a time series and track record of CPIH before making that decision. However, maybe the soundbite from Hammond that they couldn’t afford to move away from RPI was taken out of context.
Stheeman, UK DMO: It might indeed have been taken out of context and not reported correctly. I agree with what you said — I need to choose my words carefully here — but before anything can happen on our side, government must decide what its preferred measure of inflation is.
I don’t think it makes much sense for us to be doing something new until that happens. But we’ve always been clear, ever since the last consultation on this subject, which was in 2011, that this is being kept under review. And I can categorically state it is still being kept under review.
It is interesting what you say: that you don’t see a problem for the market. I’ve heard some people say they do see a problem for the market, but that doesn’t mean that one day it may not happen. But I do think clarity around the preferred measure of inflation needs to be achieved.
There is a wider debate — a political debate and economic debate — about the continued use of RPI throughout the economy.
For that reason I think it becomes more important for government to state what measure of inflation it prefers. The Bank of England has got one target. In the UK we have three consumer price indices, which you don’t see in other countries. It may be simpler to have that clarity come first.
If I could just add one thing on this, which is how government wants to tackle the issue raised by the OBR last year, and noted in the Autumn Budget last year: namely, the long-term exposure to inflation in the public finances, and the risks the government perceives arising out of that.
That merits consideration because in this year’s financing programme we’ve reduced the initially planned inflation-linked portion of issuance by two percentage points relative to last year’s initially planned programme. That doesn’t perhaps sound very much, but we highlighted at the time, that the government is taking this issue seriously.
In a couple of months the government will produce its response to the OBR’s fiscal risks report and it’s likely they will say more about this.
Before you start a new line of any stock, linked to whatever index is chosen, you need to be clear how much you can actually issue.
So a number of things need to be clarified before we see any particular change.
Parkinson, RBC: I was looking back at some of the responses to that consultation in 2011 and one of the common themes coming through from the likes of the National Association of Pension Funds was — and this ties in with Mark’s point about whether liquidity of the RPI market would be affected or not — along the lines of: ‘Well, just issue much more inflation, then you can issue CPI as well and RPI liquidity won’t be impacted.’
To Mark’s point, the really important question if the DMO conducts another consultation could be the one as to whether parallel issuance of CPI and RPI linkers will impair the liquidity of the latter.
Winkworth, Morgan Stanley: Like Mark, I don’t see any problem with the two living hand in glove with each other. RPI will always be the more liquid index, rather like the OATeis are in France, over the OATis.
But there are obviously ways you can get up and running quite quickly on CPI by buybacks, switches, things like that, which don’t change overall the government’s inflation-linked exposures. It would change the index, but not the overall size.
Hallgrimsson, JP Morgan: The switch auctions would be a powerful tool to build size in the CPI bonds quickly. One obvious choice would be to offer investors to switch out of the old eight month lagged RPI-linked Gilts into CPI bonds.
Winkworth, Morgan Stanley: There are £30bn-odd of those bonds that could be taken out to about the right maturity.
Hallgrimsson, JP Morgan: I agree completely with what Mark said on the French linker market — they benefit one another, it creates more liquidity for both.
GlobalCapital: It’s apt that we have used France as an example, as France has also issued a green sovereign bond. Can we expect a green bond from the DMO?
Stheeman, UK DMO: That’s a really interesting one. Ultimately these things are decisions for ministers. If ministers, whom we serve, decide that one of these things should be done, we will do it and we will obviously try and do it as well as we possibly can.
But they so far have not decided that and there are challenges. For instance, size is important for liquidity reasons. This was a bit of an issue when we sold the sukuk a few years ago. One mustn’t forget, however pleased we were with the sukuk, it was only £200m, which at the time was less than 0.1% of our annual borrowing requirement.
One of the issues around that was just structuring the sukuk. You don’t necessarily have quite the same extent of challenges around structure for a green bond, but for sukuk you needed to have a certain asset, that asset needed to be financed in a Shariah-compliant way, and the proceeds needed also to be spent in a Shariah-compliant way.
All these issues were much more complicated than one might assume. So there are various issues which would need to be addressed before any of this would be possible.
There is also the additional fact that we have a policy of issuing liquid bonds, benchmark bonds that command a liquidity premium in the market. If you get into the space of any of these innovations, that has to be a real test, to the extent: do they or do they not offer value for money? And that is something parliament scrutinises very closely, so this is a big and wider political issue.
GlobalCapital: How about our investors? Are you getting pressure to buy more of these assets?
Inches, Royal London: Probably less on the sovereign side, but on the corporate and equity side there is more demand from the client base for ESG-type products. We have looked at whether there is enough depth of market to create a purely sovereign, supra-type portfolio in that ESG space and unfortunately the answer is no, not really. The index is just not big enough, the depth of market is not there.
But going back to France, we did participate in that bond. It just trades alongside the other French bonds, relatively easily. It’s not trading at any particular discount or premium to its neighbouring bonds. Although it’s got that green characteristic to it, it’s still specifically backed by the French government. So it trades the same way. It’s similar to what we were talking about with RPI and CPI.
The big thing on ESG is how persistent the demand will be over the next five to 10 years. You hear about the change in investor base to millennials and how they feel about that type of structure. We run quite a large cash business as well and we find that a lot of our client base, who happen to be in the charity and university space, are heavily driven by this, either from the charitable aspect or from student bodies, who say they want fossil-free or green. So if that starts to pad out, away from cash products, into corporate bonds and sovereign bonds, you can see this becoming more prevalent over five to 10 years.
Harris, Old Mutual: I completely agree — we see it a lot more in corporates and equity, rather than sovereigns. What’s interesting in credit is that green bonds do tend to trade through non-green bonds.
Winkworth, Morgan Stanley: There’s talk that new pension regulations will force pension trustees into ESG-type assets, which is similar to what has happened in France. Once that happens, we’ll see what an incentivised investor base can do for yields…
GlobalCapital: What else would we like to see from the DMO? We saw a 100 year bond from Austria last year, as well as of course Argentina. Only 99 years to go for those lucky bondholders! But have we seen the end of century bonds for a while?
Harris, Old Mutual: We saw Oxford University do a 100 year bond last year too.
Stheeman, DMO: We consulted explicitly on what we called ‘super-long bonds’, basically 100 year bonds, and perpetuals in 2012. The responses to the consultation led to the chancellor deciding to call the existing perpetuals.
The response then was: yes, there is demand, but demand in the super-long area really starts to fall away once you go beyond 60 years. Of course, there is a price for everything. Could we issue 100 years? I have no doubt we could, in theory, but it goes back to my point a minute ago about value for money. It would come at a premium, and that’s a cost to the taxpayer which I don’t think makes it a particularly attractive proposition. It’s not something the investor base has asked us for. When we issue bonds, and this perhaps links to the wider issue of innovation in general and even actually around yesterday’s transaction, we’re not looking for trophies. We are looking to minimize cost for the taxpayer by giving the market what it actually requires, because we believe that is to our longer term advantage. My sense is that there’s not that much demand, certainly not right now, for a 100 year transaction that meets our requirements.
Hallgrimsson, JP Morgan: I hardly ever do this, but I’d probably disagree with Sir Robert on that. We brought two 100 year bonds last year, one for Wellcome Trust and the other for Oxford University, and we saw very strong demand for both. I have no doubt that a 100 year bond would fly from the UK DMO. The question for the DMO is though probably do they want it from a liability management point of view?
Stheeman, DMO: We may be saying the same thing. I didn’t say the demand was not there, I said the demand starts falling away, or more bluntly, the price at which we would have to do this goes up quite sharply and would not offer value for money in the context of the overall programme.
To go back to my point about constantly trying to build up benchmarks, it’s one thing to do a couple of hundred million as a one-off, as a 100 year. But we like to see most of our bonds at the moment in the region of £20bn-£30bn.
Hallgrimsson, JP Morgan: The pricing of 100 year bonds is interesting. I’m sure all others here have looked into this — when you extend the curve, what is the convexity worth and how do you extrapolate the inverted yield curve, particularly when you take it out to 100 years? I dare say this is something you need to do by a syndication.
Stheeman, DMO: Noted!
GlobalCapital: Maybe it’s not the product that will change too much, but perhaps the method of selling and trading. What will be the impact of technology, in particular automation and algorithmic trading and blockchain, on these markets?
Millington, Santander: With regulation we’ve seen an increasing electronification of the market and a move away from pure voice trading. And that’s definitely accelerated in the last 12 to 18 months. It’s here to stay — margins are being squeezed, there are fewer humans sitting around on the sales and trading desks and the market will become more and more electronic.
Winkworth, Morgan Stanley: The drivers behind Gilt demand and indeed supply are very human, though. One of the successes of the DMO has been its ongoing discussion and innovation and changing what it does, based off feedback with investors as to their wishes, demands, desires. I don’t think that will change, whatsoever. The mechanism of market making does change a lot with electronic trading, we’ve seen huge change in the last 10 years, and probably more to come for the next 10 years. But the fundamental reason for the existence of these markets is very personal.
Stheeman, UK DMO: One of the things which strikes me about debt management per se is that it’s probably, in terms of the actual decisions we make, more of an art than a science. It involves judgement. We have a debt management objective to minimise cost in the long-term, subject to risk.
We are constantly having to make trade-offs around how much risk to assume versus how much cost savings we want. These are judgements and they’re probably sometimes flawed judgements. They might look good on the day, but they may look quite foolish six months, or a year or two afterwards.
But I struggle to see how any country could actually conduct sovereign debt management without humans using their own judgement. Personally, I find that a huge relief and reassuring.
Parkinson, RBC: As the only salesperson sat around the table, it’s a question I get asked a lot: does electronification, which has certainly accelerated massively with MiFID, mean the death of the salesperson? For the time being, I’m finding I’m free to spend a lot more of my time than I did 18 months ago, on the phone, talking to customers, comparing views on markets, talking about trade ideas. And I would like to think that as a business we’ve got a better feel for the market and make better decisions because of that.
GlobalCapital: Sir Robert, you weren’t at the DMO when it began 20 years ago, but in your time, what have been the biggest changes?
Stheeman, UK DMO: I mentioned earlier the financial crisis and the consequences of it. The biggest single consequence was clearly that we’ve had to borrow so much more. Throughout this discussion we’ve talked about syndications and how they work alongside auctions. Syndications have improved massively the dialogue we have with the market. Auctions will ultimately give you the market price, but we really learn from syndications, to inform the sort of judgements I just spoke about, how demand is manifesting itself, from where it originates. I would like to think syndications have improved our decision-making process.
We’ve learnt much more about the market. We’ve always prided ourselves, and I think this occurred before I joined, on being close to the market, trying to understand the market.