Instinet aims to cut transaction costs, boost liquidity in Asian markets

Instinet Corporation, a wholly-owned subsidiary of Reuters Group, is the world’s largest agency broker. In the past year, Instinet has raised its stock exchange memberships to three in the Asian time zone.  The company is beginning to introduce its full range of technology into the Hong Kong market, where the exchange’s new AMS/3 system now allows direct connections, and into Japan, where Instinet recently received a licence for its trading system.

Q. Japan has been the major focus of recent activity at Instinet. What were the procedures for setting up operations there?

A. Last year was an important year for us. We got our membership and built an automated link to the exchange, enabling our clients globally to route orders into the Tokyo market, making it easier and more efficient for them to trade. We also applied for our PTS (Proprietary Trading System) licence last year, and we got that a few weeks ago. That allowed us to install the technology that we’ve used successfully in the US and Europe. Here in Japan, anything that requires using computers to trade off the exchange requires a PTS licence.

Q. Instinet is well-known as one of the major electronic communications networks (ECNs) dominating trading on NASDAQ. What are the differences in Instinet’s Asian operations?

A. Our intention was not to set up an ECN here because we don’t believe ECNs work outside of NASDAQ. The reason for that is NASDAQ isn’t really an exchange, it’s really a combination of market makers and ECNs. There are now 10, I think.  We are the largest, and we were also the first. Of course we still bring our core principles of neutral, anonymous, agency broking to the Asian markets, regardless of their structure.

Q. So in NASDAQ you’re taking advantage of the absence of a central limit order book?

A. Yes. And that’s why in Japan where they do have a central limit order book (CLOB) environment – highly liquid and highly efficient, ECNs can’t survive. The only possibility could be for retail after-hours trading, but that’s not something we have a great interest in getting into. There has been some discussion in Japan among some of the retail brokers about setting up an ECN to accommodate retail order flow and provide retail investors with fills outside of exchange trading hours. They do that now in Korea and some other parts of Asia.

But what we do have an interest in doing is allowing our clients in Japan to use the same technology that clients elsewhere are using to efficiently route orders into the CLOB in Japan as well as other markets globally. We also believe that it's useful for investors to have access to what we call the upstairs pool of liquidity. It’s important to have access to both pools of liquidity, both on and off the exchange.

If you look at our business in the US, by definition 100% of our NASDAQ trading is done “off-exchange”, because there isn’t a central exchange. When you look at the Asian and European markets, with their totally different structures, it’s the other way round: nearly 100% of the business is done on the exchange, and our clients use our technology to route orders to an exchange and get immediate confirmations back. The off-exchange business serves niche markets such as block trades.

We’ve made the infrastructure investments we needed to for Japan and I think what we need to do now is capitalize on that investment to bring more order flow into the exchange from our global client base. The PTS licence allows our clients to use a computer to trade rather than have to place orders with their broker by telephone.

Q. The Japanese finance market is often thought of as being conservative when it comes to new products. What kind of response have you had so far to the terminals?

A. It’s very early days yet. In Japan you have to be very careful what you promise because if you build up expectations and can’t deliver, you lose credibility and that takes a very long time to win back here. So we kept a very low profile on marketing the technology, but we made sure our clients knew this was something we were actively pursuing with the regulators. I think, though, that the uptake might be slow. Subscribers hitting their own keys is going to take a while because it’s not like the UK and US markets where the buy side have centralized trading desks and centralized dealing rooms. Here in Japan, there are very few houses that have centralized dealing rooms and traders that actually trade their order flow. Most orders are input by fund managers to brokers for them to execute for them and they tend to hand off the responsibility to the broker to execute. I think that is going to take some time to change but I don’t doubt that investors in Japan, buy side institutions, are very keen to reduce their trading costs and I think they realize that we can help them to do that. But changing their structure to a centralized dealing desk will also help reduce trading costs.

Q. Will this require some kind of education on the less obvious costs of trading?

A. Yes. It’s beginning to happen. Plan sponsors are beginning to become more aware of the different components of investment performance, which is putting pressure on the buy side to focus on trading costs.

Q. How long did it take to educate or convince the Japanese regulators that Instinet could play a viable role in the market system?

A. I don’t think the regulators ever had a problem with us operating in Japan: we received our brokerage licence as soon as we opened the Tokyo office in the mid '90s. And of course the operating environment has changed dramatically in the last five years: we can offer services today that were just not permitted in 1995. I think that our membership of the TSE shows that we’re welcomed by both the exchange and the regulators, and that they do understand what we are and what we do to improve our clients’ investment performance.

Our intention is not to appear like a black cloud and suck all the liquidity out of the exchange, but actually the opposite. We want to use the relationships we have globally to bring order flow to the exchange and provide institutions in Japan with the necessary technology to bring order flow to the exchange.

Q. Was losing liquidity from the exchange the major concern of the regulators?

A. When I came back to Japan with Instinet in 1996 the exchange really didn’t know who we were, or what our intentions were. It was a similar thing in Europe as well at first, and I think it helped us that the exchanges in Europe had seen that we actually brought liquidity to the market and that was communicated to the Asian exchanges.

Australia is a good example of this. We couldn’t get a meeting with the exchange in the mid '90s - couldn’t get in to see anyone. But they finally welcomed us and we’ve had a membership there for two years now. Richard Humphrey, the CEO, really understood and wanted to give us membership to get more liquidity into Australia.

Q. Do you have any quantitative numbers on the amount of liquidity Instinet has helped bring into an exchange?

A. Not that I’m aware of, but there are some studies that have been done on execution quality in the US. The last numbers I saw were for 1999. We saved investors over $3 billion in transaction costs in '99.

Q. What’s your opinion on the efficiency of the Tokyo Stock Exchange?

A. The Tokyo market is very efficient in that it handles a significant volume every day very successfully. I think once the market demutualises that will be even better news for all the participants in the industry. If you look at other exchanges around the world that have demutualised, it has benefited investors. At the end of the day, we need to make sure investors are taken care of, more so than the members of the club, which is how a number of markets have operated in the past.

Q. Where is the TSE at in terms of planning a demutualisation?

A. It’s being discussed actively now. The exchange itself has a committee and the foreign brokers also have a committee. My gut feeling is that it is going to happen, but not in a short timeframe. I think the important thing that has to be recognized by the TSE is that if they demutualise they should be operating just like any other commercial enterprise. That is, the board should be elected by the shareholders, and the executive committee - the people who run the exchange - should be run by the board, which is not the way it’s traditionally been in Japan. Running as a for-profit entity, they will have to open up the exchange and try to increase liquidity. That’s their mission in life - it’s not their mission to protect their members.

I think we’ll see some definite changes going forward that will all be positive, but sometimes they need a bit of a push. We saw that happening when NASDAQ came to Japan. MOTHERS was developed quite quickly after that.

Q. What’s your outlook for other opportunities in Asia?

A. We’re already set up with exchange memberships in Hong Kong, Japan and Australia and we’re evaluating a number of opportunities in other countries in the region. Where we haven’t got memberships yet, we’ve built relationships with local brokers, or foreign brokers in those markets.

Obviously everyone who trades through us trades anonymously so we take counterparty risk. There are markets in Asia where they can’t do that, like Korea and Taiwan, where IDs have to be disclosed. But in the more sophisticated and open markets we can give our customers that anonymity, which is a strong selling point. Here in Japan, institutions feel very comfortable placing orders with Instinet because they know that we are neutral, we don’t have a proprietary trading book. We don’t trade futures. We don’t trade derivatives. Our focus is just getting the best execution. The chance of information leaking out and having a market impact is zero.