The Asian Infrastructure Investment Bank may be the newest triple-A rated multilateral bank to enter the public sector bond market but it has not been backwards in coming forwards. It has spread far and wide across the capital markets since beginning its bond market funding with a dollar benchmark in 2019. It now raises around $10bn a year from the bond market, issuing dollar, euro and sterling benchmarks.
Two other markets that the AIIB has explored in particular are those for Panda and Hong Kong dollar bonds. Its head of funding, Darren Stipe (pictured), appeared on a special, sponsored edition of the GlobalCapital Podcast in late November, as part of the bank’s celebrations of 10 years in operation, to discuss its experiences in both arenas, including in 2025 becoming the first issuer to price a publicly syndicated Hong Kong dollar deal — the first ever Wonton bond.
GlobalCapital: The AIIB is marking a decade of operations, but although it hasn’t been in the bond market for quite that long, it nonetheless issues an array of different products, and we’re going to focus on a couple of those now. The first is Panda bonds, which are bonds sold onshore in China by international borrowers. The AIIB has been a keen issuer of Panda bonds. Tell us about your programme in that market and how you’ve developed your investor base there.
Darren Stipe, AIIB: Just starting with the first part of that question, the Panda market itself is a fascinating market. Often, I’m in conversations about cross-border connectivity of Asian markets, and Panda is very relevant in that regard.
Maybe just outside of Panda, if you look at China’s recent blockbuster dollar issuance, that one had $234bn of bids, I think, for a $4bn print. It’s just an indicator of the fact that investors do have a strong demand for Asian assets and Panda offers an alternative that’s really fitting.
It helps the domestic investor, because they get access to issuers they otherwise would have no access to at all and I think it helps the international investors as well, because they may already have CNY exposures, and it’s a good complement to what’s usually a [Chinese government bond (CGB)] holding.
In AIIB’s case, we’re 0% risk-weighted, and we’re recognised as High Quality Liquid Asset Level 1 [under the Basel Liquidity Coverage Ratio], so it’s a good fit for the typical Panda investor, which is usually a bank treasury. And we’re also issuing around 30bp wide of that CGB level — so also compelling.
And then to say a little bit more about our programme, AIIB issues one or two times per year, ranging from a two to five year tenor and in sizes from Rmb1bn-Rmb3bn.
We could do more if we had a larger borrowing programme, so it’s not to say that the market isn’t deep enough for us to do more.
And then we typically swap to Shibor or Sofr. But either way, our proceeds are going to sustainable development projects within the Asia region, so we’re doing all that issuance from the perspective of having a sustainable development bond label.
And in terms of that domestic investor base, how did they respond to the idea of buying a triple-A rated asset? Because obviously that’s much tighter than some of the offerings they have looked at before.
Yes, it’s really relevant to ask about the investor base, because it’s central to our objective of issuing Panda in the first place. So we’re trying to get access to an investor base that we wouldn’t otherwise have.
The investor base is split between onshore and offshore investors, of course; that’s the whole nature of the Panda market, as you just said.
Our philosophy is that success in the long run as a Panda issuer relies on a stable onshore investor base, so that’s what we cater to. And when we approach that onshore investor base, we find that it’s really granular, but also that there’s a process of promoting awareness that’s important.
Investors onshore need to know AIIB: who we are as an issuer; why they should care about an offshore triple-A rating, so the credit quality and why is that relevant to them; and then what is a sustainable development bond?
To get that awareness out into the market we’ve had over 100 investors met. And so part of being a successful issuer, I think, is just getting on the ground and getting in front of investors.
But in terms of promoting awareness, it is helpful to educate them on why the Panda market is relevant to them, why AIIB is relevant to them as an issuer. But really what they’re concerned about, I think, mostly, more than anything else, is liquidity. So the real way of attracting investors, I think, onshore or offshore, is going to be in the way you’re transacting, and it has to be in a way that’s encouraging secondary market liquidity. So, we issue in benchmark size, we’re issuing frequently, and we’re pricing in a way that we actually do pay a premium for the liquidity difference between our bonds and the CGB. And we also account for the difference in taxation compared to CGB in the way we price — so really hitting a market clearing price.
And when we do that in the Panda market, you get secondary market flows. So we see in our bonds healthy two-way flows — that’s with onshore and offshore investors. We are seeing about five dealers actively trading our bonds, so it’s really healthy. And that’s what attracted that domestic investor.
An adjacent market is that for Hong Kong dollar bonds. Issuers have been printing privately placed bonds in that market for decades. But this year, the AIIB sold its first, or the first, public bond in that market, a Wonton bond. Firstly, can you outline the difference between the two ways of issuing in the Hong Kong market — both private and public — and secondly, how did your experience in the Panda market inform your decision to do a Wonton deal?
Yeah, I think of the Hong Kong dollar case [that], as the IFI [international financial institution] community, we kind of got caught up in Hong Kong dollars being part of the mix of our private placement business. It’s not a currency where IFIs are lending, so we think of the Hong Kong dollar market as a complement to the overall private placement business, where we’re shown levels across the variety of currencies, and Hong Kong dollars just would have been one of them.
There’s also a dynamic where there’s a smaller set of dealers covering the Hong Kong domestic market, who also cover tier one SSA issuers and so syndication just never really came up.
I think if you look at the size of the Hong Kong dollar private placement issuance in 2024 it’s about HK$36bn, if you just focus on that tier one MDB peer set.
It was large enough in our minds that it did make us wonder: is there enough demand where a syndicated deal could be a success?
And so when we started exploring that with dealers who are on the ground and close to Hong Kong domestic investors, we thought we really could come up with a transaction that would be benchmark volume, still price close to those private placement levels and also attract a broader investor base than we’re getting out of the private placement business.
Maybe we can cover it later, but there’s no difference between the bond itself [and our private placements] — so same documentation, it’s just the fact that it’s a syndicated transaction of a larger size that’s going to attract investors in who otherwise wouldn’t be able to participate in a private deal.
We talked a little bit about the characteristics of the Panda bond investor base. What’s particularly different or similar in the Wonton market? And is being a triple-A rated, tightly priced issuer with a limited borrowing programme there a hindrance or a help?
If we’re comparing Panda to Hong Kong dollars, both of those investor bases are dominated by bank treasuries, so that’s quite similar.
Hong Kong dollars is more domestic-driven. Panda, of course, is meant to attract both onshore and offshore participation, so especially central bank, official institution-type investors are attracted to the Panda market because they’re managing their currency reserves.
That isn’t the case for Hong Kong dollars, but the fit for that domestic Hong Kong dollar investor is compelling. So, I was saying the volume of what got issued last year — around HK$36bn of private placements. Already this year, when we introduced syndicated Wonton bond issuance, the volume has already reached HK$23bn.
We still had HK$44bn this year issued under the private placement format, so that still is relevant, but the overall size of Hong Kong dollar issuance for our peer set has almost doubled, and the syndicated format is already 30% of the market right out of the gates, so it’s quite interesting.
For us, the size we can issue is limited by our programme size, it’s true. I think there’s strong enough demand that we could print more, we just have to become a larger issuer, and that’s going to happen for AIIB as our balance sheet continues to grow.
What was AIIB’s approach to pricing in the Wonton market? Did you prioritise cost of funding or growing the investor base? It’s always an interesting thing to think about with an inaugural deal like that, I think. Some issuers will be focusing on the arbitrage funding they can achieve, others will be looking to start a new market and driving pricing over time by growing the market. What was your approach?
Yes, well, we were very conscious that we were going to be the first international issuer in the Hong Kong dollar market from a syndicated perspective, and so we wanted to set a precedent for both pricing and size.
On the on the size side, we looked at domestic issuer behaviour and really landed on HK$4bn as a size. So, it was not too large — we didn’t want to be too aggressive with an inaugural trade — but also not too small. You still want to get the benchmark size out of it, that’s the whole point.
HK$4bn — I would say it wasn’t a size for the timid. Having done 13 private placements in 2024, those were averaging a size around HK$400m, so it was quite a bit larger than our normal private placement business.
And then the on the pricing, it had to make sense for both issuer and investor. So, for us it wouldn’t have made sense to issue if it was going to be much more expensive than the private placement; then we would have just preferred to continue with private placements. And for the domestic investor it wouldn’t have made sense if it wasn’t offering some type of pick-up over the more liquid government bond.
So it’s fortunate to have a sweet spot where investors were getting about 40bp over the government bond, whereas AIIB was still well within our US dollar cost of funds when we swapped the proceeds back to Sofr.
The Panda market is known for how long-winded the approval process can be before a borrower is allowed to issue. Hong Kong is obviously a very differently regulated capital market traditionally, but what are the features that issuers need to consider before issuing their own Wonton deals?
Yes, it’s one of the things that makes the Wonton market quite attractive — [it’s] a lot lighter a lift for sure. In our case, we had everything in place. We already had an MTN programme which we issued from, so no new documentation for us or investors to consider.
We already had our sustainable development bond framework, so there wasn’t anything new about our labelling.
We were already 0% risk-weighted under Basel [and] Hong Kong investors already knew it. And we also already had [Central Moneymarkets Unit (CMU)] clearing and settlement — that’s the natural means of clearing for any domestic investor, which is going to be the bulk of any Hong Kong dollar deal — so no change required from their side either. So, it was just really coming to the market with the syndication set-up is where the difference was.
I think any issuer approaching a Wonton trade with one or more of those elements missing might have a different experience than AIIB had in terms of size or pricing, or their ability to attract a wider investor base. It’s going to be a headwind, I think, if you don’t have all those elements in place.
For us, it worked out really well, and that’s why we’d expect to come back to the market in 2026 with another trade.
We have a fresh programme for next year, and expect to be back with another benchmark Wonton bond.
