On a mid-August day, Traders Magazine sat down separately with two Bloomberg executives to discuss developments in the foreign exchange market. The two businesses are segregated for compliance reasons, even though they are accessed through the same Bloomberg terminal. That explains why they are on separate floors at Bloomberg headquarters in New York.
Bloomberg Tradbook is a broker-dealer that offers electronic trading in FX. It provides algorithmic trading solutions to clients and an ECN that features anonymous trading with firm orders.
Tradebook is separate from Bloomberg, which has its own FX offering through its terminals. But the cornerstone of Bloomberg is its analytics, which allows clients-from financial institutions to central banks to corporations-to make more informed decisions on their trading strategies and risk management. Bloomberg’s FX trading system is a multi-dealer platform called FXGO. The platform has more than 200 banks making markets in currency pairs to clients on a fully disclosed basis, so banks know their counterparty going into the trade.
Tod Van Name is the global head of Bloomberg’s foreign exchange, economics and commodities product. He oversees business strategy, product development and content for FX, economics and commodities on the Bloomberg Professional service. Van Name, who has spent his career in FX and derivatives, joined Bloomberg in 2006. He’s traded currencies and FX options for 19 years at several major banks, including management roles at Fuji Bank and Fortis Bank.
Gary Stone, CMT, is chief strategy officer at Bloomberg Tradebook. A former prop trader at BNP Paribas, Stone has been with Bloomberg since 2001. He was named director of trading research and strategy in 2004, but has focused on product development and strategy since joining Tradebook in 2007.
Here is what Van Name and Stone had to say about industry developments, clients and their business.
Tod Van Name, Bloomberg
In the big picture, what is your challenge at Bloomberg with FX?
The interesting thing about Bloomberg, as a product, is that it does everything from soup to nuts, which is a really hard thing to do, because it is easier to pick one aspect of the market and get really good at it. From Bloomberg’s perspective, you have to be good at everything. You can’t be good at execution and bad at data, or good at news and bad at analytics. You have to be good at all those things. What makes us distinct is we have a full suite of tools that provide information in a digestible format-from idea generation to price discovery. It is seamless, and when you’re done trading, you can do an evaluation of how well you did.
Which areas are contributing the most to your growth?
Corporations are Bloomberg’s biggest growth segment this year. Also, government agencies have adopted Bloomberg, as have regulatory bodies, such as central banks. We have 50 central banks that use Bloomberg FX as a platform, and 33 that execute over Bloomberg’s multi-bank trading platform, FXGO. Financial institutions constitute the largest group of users, but one-third of these users are not FX players, so 30 percent of the people who use FX functions are equity or fixed-income users, economists or strategists.
Are there challenges to having non-FX people using your system?
We try to make it so that if you’re a bond guy, and you’re looking at exposure to European markets, you can assess your currency risk, as well as what your interest rate or sovereign risk is. We also try to make it very easy for users to step in and find all the things they need to know about a topic, for example the Eurozone or China. The challenge is to keep it easy, because there’s so much data and analytics on the system.
And is there growth in those areas?
The number of users for FX analytics is up about 30 percent from a year ago. And we’ve seen a rise in users of our electronic trading product, FXGO, up 23 percent since January.
What are you seeing with algorithms?
We’ve experienced huge growth in the algo business. We have eight liquidity providers offering their algorithms to clients. Even though that share of the market space is relatively small, it could grow substantially, as people become more sensitive to what their actual fills are and what it costs them to do business. The algos that are most utilized are relatively simple. The benefit of an algorithm isn’t necessarily crossing the spread, because the spreads are already so small in FX. But if you have an order that’s larger than the liquidity available in the market, algorithmic trading tools enable you to identify the best time of day to trade a certain currency. For example, in New York, the deepest part of the market is around 11:00 am ET.
What is going on with trading volumes?
FX has been a growing market. In 2007, FX volume was at $3 trillion a day, and it got up as high as $4.8 trillion a day. But we’ve fallen back as of late to about $4 trillion because of what’s happening in Europe. For us, the volume on FXGO is up 50 percent just this year. And it is up about 65 percent if you look at the rolling 12 months. We’re pretty pleased.
Has that impacted your standing as a destination?
There are a handful of multi-dealer platforms. Most aren’t too dissimilar in terms of volume. We generally don’t release our volumes for one primary reason, and that is we don’t charge for trading. Our revenue model isn’t based on our volume; it is based on people who have terminals on their desks. So what I really care about are our subscribers and that they are using the analytics, not just the electronic trading platform. We don’t charge for electronic trading, which is a nice thing for our clients. It’s free to the buyside and sellside.
And this business model works for you?
Yes, our numbers are significant, but I don’t want my sales team to be motivated by trying to find volume users. I don’t need to have a high-volume trader. I need to have subscribers. We’ve found our model to be a much more stable business strategy, because volume can be fleeting. You can be the king of the hill one year and the next year someone comes out with better technology and they redirect their routers and business elsewhere.
There’s been a plethora of new trading systems in FX. What’s going on?
As a market, FX in many ways has become commoditized, particularly the spot market. Because technology is cheaper, it is not surprising there are more entrants. But these systems probably won’t all survive. It doesn’t do a liquidity provider a lot of good to distribute across 16 venues. It is very hard to manage that risk and you have to spend a lot on infrastructure technology to get your liquidity out there. What do you really get from that? Unless your clients want to deal on those platforms, it doesn’t make a lot of sense as a business strategy.
What is it that clients want?
Clients and market makers at the larger institutions are looking for a very inexpensive way to transact-inexpensive from a transaction-cost basis and from an infrastructure basis. Remember, FX is a very small-margin business. You have precision pricing in spot trades where you’re out five decimals. A tick in one million euros is $10. Ten dollars! So to be in the business, you must have high volume or provide a service to your clients where they’re going to want to deal with you.
Gary Stone, Bloomberg Tradebook
What latest development in FX interests you most?
From our perspective at Bloomberg Tradebook, the biggest change in the FX market is the move of the trading function from the back to the front office. In the past, a front-office trader would buy a foreign equity and an FX trade was done at the end of the day, by the back-office professional through a custodian. Now the responsibility is moving to the front office, so the trader is beginning to take control of the transaction, and because of that, they’re starting to demand more tools and seeking a better price. So, we’re seeing more scheduled-type algos come into vogue, and a maturation of the market from before, when FX trading was really just an afterthought.
You’ve seen a double-digit increase in algorithmic trading on your system. Why is that?
Algorithmic trading is actually one of the fastest growing areas in FX trading. We think algorithm use is increasing because algos can help minimize risk, which is a concern especially if FX is a new asset class to the front office. One way to reduce risk is to hand an order to a sales trader. But if the front-office trader’s mandate is to manage the trade himself, in other words, to take control and reduce costs, algorithms provide a way to do that.
What is the most popular algorithm for these clients?
Our average price algorithm continues to grow. In equity vernacular, it would be a soft limit. So therefore, it enables you to be able to ensure that your average price is within your limit.
How is that different from a hard limit?
It is interesting that equities haven’t responded to this idea of a soft limit. In equities, if you’ve got a limit, it’s a hard limit, meaning you won’t buy above that price. But, if you bought below that price all day, and all of the sudden the price spikes, you may want to continue buying. Thus your average price, the price the portfolio manager wants to own the position at, is the limit of the trade. In a hard-limit scenario, you may be moving away from the goals of execution and what the portfolio manager is trying to do.
Describe your customer base for algos?
The client mix trading FX through Bloomberg Tradebook is more similar to the mix we have trading futures, than the traditional group of long-only funds that trade equities. However, we do have clients that use scheduled algorithms; by far we have more clients who are putting strategies together that trade FX as an asset class. They’ll trade off of economic statistics and what’s going on in the marketplace. If the nonfarm payroll is X, then they might want to either buy or sell the dollar. It is a very hedge-fund, prop-trading type of atmosphere.
What are your clients’ objectives?
When you look at the FX market, there are two types of participants. There are those who trade it as an asset class and they’re buying and selling it as a core strategy. There are other firms for which FX is an ancillary action that occurs in response to, or because of, something they’ve done as part of the core strategy. For these clients, scheduled or average-price algorithms tend to work well. For those who use FX as an asset class, they’ll use more of the event or trigger algos.
Why has the growth in FX been so rapid?
I think there are two things to consider. First, if you look at the equities markets, the volumes have shrunk dramatically. Second, if you look at the derivatives markets-that is, options, futures and foreign exchange-they are holding up a lot better as a group than equities. I think investors are shifting their strategies and looking for more places to find alpha and more and more are looking toward FX as a result.
How are you changing things now?
Tradebook is adding a lot of liquidity providers to our platform, and we’ve created a marketplace where buyers and sellers, and different liquidity providers, benefit from interacting with each other. You can’t pair a high-frequency trader with a high-frequency trader because they have the same strategy. They go through the revolving door the same way, at the same time and nothing gets traded. I think that is one reason why we’re seeing dramatic growth in our marketplace this year, because counterparties get what they want.
Are you seeing a lot of block trading?
We have clients that trade $50 million, $100 million, $200 million at a clip. We created a view which enabled them to see not just what was at the inside of $3 million by $3 million, but look at the marketplace. ‘What is my price for $50 million? What is my price for $100 million?’ So we aggregate up all the bids and offers and depth and basically say, ‘Here is your average price for $50 million; here is your average price for $100 million, etc.’
What is the future?
Our game plan is to continue to develop a marketplace with diverse sources of liquidity. We will work with clients and help them to solve their problems. If you look at the industry and the model, the ECNs and multi-dealer platforms coexist very well because it depends upon what the client is trying to achieve. Where he wants anonymity, an ECN fits nicely. Where he wants to go out loud and get a block, that is when a multi-dealer platform is probably best.
(c) 2012 Traders Magazine and SourceMedia, Inc. All Rights Reserved.
http://www.tradersmagazine.com
