Imagine paying for your groceries and receiving a receipt that shows only the total sum spent, without listing each item you purchased and its respective cost. Sounds preposterous, yet that's exactly the sort of trade confirmation (receipt) money managers are accepting from many brokers.
Information intentionally absent from trade confirmations that asset managers and pension funds receive from their brokers - information detailing proprietary trading elements related to that execution - is costing investors more than can be imagined. Without a fully disclosed confirmation that reveals how, for example, 1,000 fills on an order were each handled (agency vs. principal), institutional investors will never know the true impact of their trading costs and, more significant, when a proprietary desk at an executing broker-dealer traded against their orders.
Opaque and incomplete trade confirmations more than strongly suggest that money managers are paying too much for their executions. Greater transparency is needed. The SEC has mandated Reg NMS, decimalization, best execution and other standards to improve the capital markets and help people better understand what is happening with their investments.
Do You Have Proof?
Best execution is a federal requirement. Pension plans, as part of their fiduciary responsibility, are entrusting money managers to obtain best execution. But how do they go about demonstrating it without a complete picture of a transaction's life cycle? Trade cost analysis is left to guess work, and that's unnecessary when the information already exists.
What the buy side receives now, from those brokers trading for their clients and their own books, is an aggregate confirmation. There may be a notation on the confirmation indicating that some or all of the order was executed on a principal basis by the broker's proprietary trading desk or market making desk.
What the broker isn't providing is detail on each execution: Was it traded on an agency or proprietary basis? Without this information, the buy-side firms have no way of attaining valid transaction cost analysis (TCA) that would show if proprietary trading helped or hurt them in the execution process. But most money managers are unaware that a lack of transparency and disclosure in trade confirmations, as well as an outdated broker relationship, is costing them money, which is why they still go along with this process.
Before institutional trading morphed into the electronic environment we see rapidly evolving today, the only way to execute a large block of stock was to call a broker who would provide capital commitment. With that business model, a buy-side trader negotiated a price with the broker; you knew what you were getting. Providing a print on a transaction was straightforward.
Today, we have a variety of direct-access, algorithmic, and other electronic tools and services that slice orders throughout the day. The dynamics of the execution process have dramatically changed. Analysis of individual prints is necessary when trades are being chopped into small pieces. With multiple executions combining to create one large trade, the only way to perform complete trade cost analysis is to have information on each child order and not just the end result or parent order.
But the analysis most buy-side traders are conducting currently is based on an average price and overall volume. They're missing a critical piece of information: On those child orders, which ones were yours throughout the volume of the day? And of those trades, which ones were agency versus those executed on an internalized or proprietary basis?
What usually happens is this: Upon receiving an order from the buy side, a broker's proprietary desk may notice an opportunity to trade part of it on a principal basis to the broker's advantage, with an end result that costs the client more than it should have paid for a particular execution. For example, a buy-side trader has a spread 3 cents to 4 cents wide, and the broker sees that the market may be coming in. The broker gives his client a trade, prints 5,000 shares of the overall block on the offer and then, within the next 5 seconds, proceeds to buy it 3 cents to 4 cents cheaper on the bid because that was the direction of the market. The broker, in giving his client the benefit of that move, instead could have put that order out at the marketplace on the bid or an ECN. The buy-side client would have captured the spread instead of the brokerage firm.
It's clear there are opportunities within the spread or a given timeframe, and these brokers, in aggregate, are making money that's coming out of someone else's pocket. One only has to read in the financial press about the staggering profits Wall Street firms are generating from proprietary trading to better appreciate what is taking place. It's also worth noting that NYSE Regulation announced, prior to forming a single Self-Regulatory Organization (SRO) with NASD, an exhaustive review of algorithmic trading at large brokerage firms that conduct proprietary trading.