Ronald Arculli has a problem with dark pools. Yesterday, the Hong Kong Exchanges and Clearing chairman fired off a withering salvo at the off-exchange stock trading systems rapidly gaining popularity with institutional investors in Asian equities. To hear Arculli tell it, dark pools obstruct price discovery, split liquidity, increase volatility, hamper market regulation, impair corporate governance and exacerbate systemic risks. Meanwhile, their opacity creates a two-tier market in which professionals have access to privileged information and ordinary investors are excluded as second-class citizens. 'The playing field is no longer level,' he told a lunchtime audience in the Foreign Correspondents' Club. 'The public - the retail investor - may not have access to the best prices.' Coming from a market regulator, this would be blistering criticism. But as chairman of the listed exchange HKEx, Arculli isn't exactly a disinterested participant in the debate over dark pools. Although he presented himself yesterday as the shareholders' champion, precisely which shareholders wasn't entirely clear; whether investors in general, or investors in HKEx shares in particular. In this matter, their interests are not necessarily aligned. To see why, we need to look more closely at what dark pools are and how they operate. In a nutshell, they are private electronic order-matching systems, where institutional investors can fill large orders without declaring either their identity or the volume of shares they are seeking to trade. A dark pool participant, typically a fund manager, looking to sell a large block of shares will enter into the system the full size of their order, the minimum volume they are prepared to trade and a price limit beyond which they are not ready to transact. The computer will then sweep through its order book looking for offsetting buy orders. If it finds one that matches the minimum size at the stipulated price, the system will automatically match the two orders without revealing the identity of either buyer or seller. Institutions love dark pools because they can execute big orders at a single price without moving the market against them. On an ordinary exchange, they would typically divide up their order and parcel it out to several brokers in an attempt to slip it under the radar of market participants. Even so, there is a risk that other traders will realise someone is trying to work an abnormally large order and then either front run the trade or back off from the market. Either way, the price can move significantly against the institution placing the order. Dark pools allow fund managers to avoid that danger. But regulators are worried that the bigger the volumes that get executed on these off-market systems, the greater the risk that a divergence in price could arise between dark pools and regular exchanges to the disadvantage of retail investors. 'The more prices are formed away from the exchange, the more we are concerned that retail investors are not getting the best price,' Securities and Futures Commission chief executive Martin Wheatley told dealers last year. His fears are understandable, but it is debatable whether they are justified. The main function of dark pools is not price discovery. In fact, according to dark pool operators, more than 90 per cent of orders are executed within the bid-offer spread already established on the regular stock exchange. The real job of dark pools is volume discovery. They allow big institutional investors to locate big blocks of shares and deal in them with an efficiency that would be impossible on the open market. Dark pool operators and institutions maintain this does not disadvantage retail shareholders, but rather that it benefits ordinary investors in insurance companies, mutual funds and pension schemes by ensuring they get the best possible execution. That argument has been acknowledged by regulators in the United States, who in 2005 enacted Regulation NMS obliging dealers to execute client orders in the marketplace offering the best price. The result was an explosion in alternative trading systems like dark pools, which now account for about 20 per cent of US share trading volumes. The New York Stock Exchange, Nasdaq and the London Stock Exchange have all set up their own dark pools. Recently, dark pools have been growing rapidly in Asia. Operators claim, however, that they are not stealing liquidity from established exchanges. Rather, they say that by allowing institutions to execute in volume, they permit portfolio managers to invest in smaller, less liquid stocks that would otherwise be off-limits to big buyers, benefiting all shareholders. They may have a point. Liquidnet, which operates a dark pool in Hong Kong, says its average trade size during the third quarter of the year was HK$9.46 million, compared with HK$91,000 on HKEx. Arculli, however, is unimpressed. Part of the reason may be that HKEx has been slow about establishing its own version of a dark pool. Scenting an opportunity, in August the Singapore Exchange announced its intention to set up an offshore dark pool to match institutional orders in Hong Kong-listed stocks. If Singapore gains a first-mover advantage in offering exchange-backed dark pool services, it will clearly pose a competitive threat to any future HKEx initiative. That would be bad news for HKEx's earnings and for the exchange's shareholders. So it would be interesting to know which investors Arculli was really standing up for yesterday in opposing dark pools - investors in general who could well benefit from them, or the shareholders of slow-moving HKEx, who may stand to lose out.