Multi-Asset Executions Across Asia
With Kent Rossiter, Head of Asia Pacific Trading, Allianz Global Investors
It is important to remember that this is a people business and that relationships are essential regardless of the asset class being traded. Without long-established and tight broker relationships, it can be more difficult to achieve solid allocations on primary deals, or for banks to assist when an order is bigger than conditions warrant, and get pricing inside the spread during tough times.
TCA measurement of fixed income trading is as much a challenge for us as it is others in the industry. This is not helped by very sparse trading in many of the securities and the general lack of transparency. The fixed income markets have been trading OTC for decades and it is unlikely that they will transform into the ‘ticker-tapes of equity-type’ trading or become a transparent market overnight. At Allianz Global Investors (AllianzGI), we have adjusted our trading behaviour to reflect the realities of the markets. We are confident in our processes and trading discipline, and that we are able to get the best outcome for our clients when trading bonds.
Our bond brokers are evaluated not only for their trading acumen, but also their research and sales servicing abilities; there has to be a balance between both factors. In equity trading, commissions are paid and CSA programs have been set up. However, this isn’t currently available for fixed income in the same way. We have had to find ways to reward both the fundamental and economic research provided by bond brokers, but still achieve best execution. Where possible, the focus is on brokers providing the holistic fixed income service.
In FX markets, almost all our major currency trading takes place electronically. Some restricted currencies are still traded using voice trading via phone or IB chat.
Our Asian-based peers are increasingly showing interest in what AllianzGI Asia Pacific has been doing for years; that is, trading multiple assets from a single desk. From a practical point of view, this makes sense because we use the same systems and staff. With trading volumes sporadic and lumpy (i.e. some days dozens of bond trades and other days there may only be a handful), it would be inefficient to have a fully staffed trading desk so specialised that they couldn’t execute and assist on other trading products during slower times. Our traders do have designated markets and products but they are also flexible and well-rounded enough to execute other products.
A key feature of our set-up was the merger of our trading desks in Hong Kong, Singapore, Melbourne and Tokyo into one single desk, now based in Hong Kong. It made more sense and improved communication having traders side by side as this balanced out the flow on the team. Our regional bond investment team is headed by CIO for Fixed Income Asia Pacific, David Tan from our Singapore office, yet all the trading takes place out of Hong Kong. Within Asia, we execute orders from portfolio managers based in Hong Kong, Singapore, Taipei and Tokyo.
Issues of Asian USD-traded corporate bonds are usually traded electronically through various platforms. Sometimes we have more success electronically with sell orders than buys, probably because the dealers lack inventory in those names and are hesitant to sell short. It is evident that dealers continue to be risk-adverse: coming into year-end, liquidity further dries up as dealers prefer flat books. One of the benefits of electronic trading is that multiple brokers can be approached much more quickly than could be achieved manually via separate IB chats. However, electronic trading of Asian local currency bonds has not yet developed enough to trade successfully that way all the time, with the exception of perhaps some SGD or HKD bonds. Even with increased activity from our multi-asset portfolio management team based in Tokyo, we still trade much more in corporate bonds than sovereign bonds.
Without a significant change in the behaviour of investors, the traditional model of using brokers is likely to be part of the Asian fixed income landscape for years to come. The markets are ‘fractured’ mainly due to the lack of control of bond markets and the amount of inventory. There are some initiatives slowly gaining traction to match orders: either scrape the order blotter, or potentially even the inventory of institutions who are likely to be trading the same bonds, but in opposite directions. Blotter scraping for equities has grown increasingly over the last decade and is now a well-established practice. It is likely to gain a similar following in fixed income as investors become more comfortable with the idea. We are looking to invest more effort and technology into matching up potential Indications of Interest (IOIs) of investor positions. Many end-investors may not be able to imagine a market so fluid that there are market makers and genuine investors out there making bids on the positions they hold.
There are many valuable initiatives coming through but they need more take-up from users. There are also competing interests depending on where you are in the cycle. The vested interest from banks to keep bonds trading OTC resulted in relatively slow take-up, and considerable resistance to the move towards electronic trading. Politics is also a factor; for example if one government legislates that their countries’ bonds must be exchange-traded, then that is just one element of what the desks handle. The Chinese markets are a good example of how hard it is to get people to trade bonds on exchanges. The China interbank bond market (CIBM) which trades the same bonds as are listed on the Shanghai & Shenzhen bond markets, still have the lion’s share of trading volumes.
It is clear that bonds can trade in many ways, with different types of broker and market participant – there is a singular lack of consistency in this ‘fractured market’. This is quite unlike Hong Kong equities where nearly all the equities are going to the exchange with the exception of blocks, and even those get reported on the tape. The development of bond trading platforms and tools is very valuable, and we are moving in the right direction but progress is slow.
Some effort is being made to reduce the number of different bonds available, although anyone at DCM may disagree given the multitude of new issues they have to deal with daily. It will require the maturation of existing bonds and the retirement of a large number of issues before we really find much success there.
With fewer bonds, there is a better chance of being able to trade them more systematically. Right now, it is comparable to the housing market where each and every house is different and you have to look at each one differently. What is needed is a more uniform approach (as it already exists for equities) so that each “Running a multi-asset trading desk inevitably means there are more topics in which our traders have to specialise. There are so many in fact, that the work has been allocated so each trader focuses on a different subject, developing them into the ‘go-to’ trader for that area. ” and every bond isn’t another ‘house’ per se but an ‘apartment’ with known dimensions and specifications, and you can buy it or sell it without actually having to visit the property.