In the last five years, European and US equity markets have been undergoing a rapid evolution as regulatory reforms and technological developments usher in new trading venues that challenge formerly monopolistic national stock exchanges.
The pace of change in the Australian equity market has been much slower. But this looks set to change later this year with the recent announcement of a timetable to launch Chi-X, the first competitor to the Australian Securities Exchange (ASX) since the state exchanges amalgamated in 1987.
Globally, new trading venues have offered faster and innovative new trading models and brought with them a colourful array of new language including dark pools, high frequency trading and maker-taker pricing.
But there is confusion among outside observers and even for some within the markets over what these terms mean and how they affect markets and trading. It is therefore timely to consider recent developments in overseas equities markets and the potential impact of competition, dark pools and high frequency trading in Australia.
For the last 20 years, most stock exchanges have operated using electronic trading systems. The ASX uses the the most common one, the electronic limit order book, which collects and displays all buy and sell orders from investors in the market and automatically executes these orders according to price then time priority. This means that the best priced orders trade first, and if there are two orders at the same price, the order that was placed in the market first, trades first.
By giving priority to the order placed in the market first, this rule encourages investors to display their orders and provide liquidity to the market. The information displayed in the order book helps to ensure an efficient price discovery process.
Regulatory changes in the US and Europe aimed at promoting competition for trading combined with reduced technology costs has contributed to the development of many new trading venues in these markets.
The new trading venues have competed aggressively on price and on the nature and speed of their trading technology. The existence of multiple venues trading the same securities has created new opportunities and attracted new investors, leading to reduced trading fees and increased liquidity in the market. Some of these new trading venues have been extremely successful in attracting market share. In the US, two relatively unknown new entrants, BATS and DirectEdge, now account for almost 25% of trading in US equities.
In the UK, Chi-X accounts for almost 30% of trading in FTSE-100 stocks. The names most familiar to us, such as the New York Stock Exchange and London Stock Exchange, have become much less important for trading. In Australia it is unlikely that we will see the same level of growth in the number of venues that has been observed in overseas markets.
However, competition will result in lost market share for ASX. The threat of competition has already resulted in a significant reduction in ASX trading fees and ASX has also upgraded its trading technology.
Developments in overseas markets have led to significant growth in high frequency trading (HFT). Although it is often viewed as a single type of trader, HFT actually encompasses a range of different trading strategies. The common element across these strategies is that they rely on being able to trade very quickly and only hold positions for very short periods of time.
Typically these strategies involve the use of large numbers of orders, small positions and holding no positions overnight. The TABB Group estimate that for the US markets these types of trading strategies account for approximately 60% of all trading activity. These types of traders are frequently criticised in the media for not trading on fundamentals and for creating volatility.
Despite these criticisms, the academic evidence on their impact is overwhelmingly positive. Research to date suggests that they increase liquidity and reduce trading costs by narrowing bid ask spreads and do not have an adverse affect on volatility.
In early 2010, ASX estimated that HFT accounted for only 3 to 4% of trading activity in the Australian market. This proportion is thought to have increased since then, and is expected to continue grow more rapidly with the launch of Chi-X and the introduction of ASX’s second platform, PureMatch. Growth in these types of strategies will lead to further reductions in average trade sizes and increased message traffic in trading systems.
Competition between markets has also lead to developments in the types of trading mechanisms offered by markets. In particular, there has been growth in trading venues offering non-transparent trading mechanisms. These venues have come to be known as dark pools.
Dark pools allow trading to occur without orders being displayed to the market. Trades however, are reported to the market immediately after they are executed.
Initially dark pools aimed to cater to the needs of large traders. If large orders are placed in a trading screen, it is likely that the price of the stock will move against the investor that placed the order.
This is referred to as market impact costs. In order to avoid market impact costs large investors typically prefer to break up their orders into smaller trades or to trade without displaying their orders to the market. Dark pools catering to large institutional investors therefore aim to provide mechanisms which help to minimise market impact costs. One example of this type of pool is Liquidnet.
Liquidnet allows large investors to interact with one another electronically and anonymously. Once a counterparty of sufficient size is found, investors anonymously agree a price at which to execute the trade. Trades are typically executed at prices within the best prices being displayed on the primary exchange. These types of venues therefore assist in volume discovery without contributing to price discovery.
These types of non-transparent trading venues have always existed for large traders, but technology has changed the way in which the needs of these investors are met.
In overseas markets, the use of non-transparent trading mechanisms is no longer limited to large investors. For example in the US there are over 40 different dark pools catering to the needs of different types of investors. One of the areas of significant growth in the US is broker-dealer internalisation engines.
These dark pools allow brokers to execute client orders against their own capital rather than sending the orders to the displayed market. Often the order flow that is internalised is retail client order flow. The merits of non-transparent trading for small traders are not clear.
Brokers argue that they are able to offer their clients ‘price improvement’ over the orders displayed in the displayed markets. However, this ‘price improvement’ is typically only a tiny fraction of a cent per share. While the client may receive an economically insignificant benefit, this potentially causes problems for the market as a whole.
If other investors are no longer rewarded for placing their orders in the market with earlier execution it discourages the display of orders in the market. In the extreme this reduces the liquidity in the market and makes trading more costly for everyone.
While dark pools are important, it is necessary to find a balance between dark and transparent trading. It is important that a significant part of trading volume is transparent to ensure that price discovery can occur and to ensure that markets are deep and liquid. In the US, the volume of trading done in non-transparent venues has started to raise concerns both with regulators and investors in the market.
In December last year in the US about 12% of trading was done in dark pools and 18% in broker dealer internalisation engines, meaning the total amount of trading done without any pre-trade transparency is about 30%. This is up from 17% two years earlier.
Many market participants argue that the US has allowed too much trading to move into the dark. The Securities and Exchange Commission (SEC) has expressed concern about this, and has consulted with the market on potential regulatory changes to address this concern.
In Australia non-transparent trading has always been available to investors wishing to trade in large size (that is, an order over $1M or a portfolio of at least 10 stocks over $5M). This type of trading has always been an important part of the trading process in Australia and currently accounts for approximately 15% of turnover.
Brokers have also been allowed to internalise orders provided certain conditions are met. In introducing competition to the Australian market, ASIC needs to carefully consider mechanisms for maintaining the balance between transparent and dark liquidity to ensure that the price discovery process is protected and to ensure that the displayed market remains liquid.
In its consultation paper late last year, ASIC proposed limiting non-transparent trading to orders over $20,000. ASIC have recently announced that there will be further consultation on this issue later this year.
Experiences in overseas markets suggest that the Australian market is entering a period of significant change and development. We have an opportunity to learn from the experiences of these markets to ensure that the benefits of competition are maximized and risks are limited.
Market participants and investors must ensure that they are informed about these changes and actively contributing to the policy debate to ensure that the best outcome is achieved.