1) Market Maker – that is, firms which provide quotes and then act as the opposite side of all client trades, taking client trades on “their books”, versus firms which act as
2) Agent (or “No-Dealing-Desk firms”), which (typically) aggregate quotes from third party sources (such as banks, or other Forex firms), then pass on those quotes to their client traders, usually adding a markup in the form of additional spread to those aggregated quotes. Clients open and hold an account with the Agent firm, but each client trade is either “passed on” to one of those third party firms, or is mirrored by the Agent firm with one of the third party firms. In that way, the Agent firm takes on no position itself, and no (or very little) trading and market risk.
The theoretical argument goes that Agent firms have no conflict of interest with their client traders, while Market Maker firms do as they are the opposite side of every client trade. And, as such, traders should do better in an Agent firm environment. The more sinister version of this (theoretical) argument is that Market Maker firms – as the opposite side of every client trade – only make money off their clients’ losses, and as such are somehow incentivized to hope their clients lose money on their trades.
To jump ahead to our conclusion, the data out there simply does not support this theory at all. If anything, traders seem to do just as well, if not better, at Market Maker firms. Why that is requires some theorizing and explaining, of course. But first, the data:
FxPro results converted from British Pounds. FxPro volume estimated from chart in IPO memorandum. Gain Capital net profit excludes gain/loss from embedded derivative liability attributable to the redemption feature of outstanding preferred stock. Client Profitability data as reported in Forex Magnates.
With the recent IPOs of two large US-based firms, FXCM (Agent) and Gain Capital / Forex.com (Market Maker), and the attempted IPO of Cyprus-based firm FxPro (Market Maker), we for the first time received a lot of data and details which can help form the basis of our comparison. In addition, thanks to new NFA rules (in the US) we now have information on client account profitability at the various US firms. [Note that there are other public firms such as IG Group and London Capital Group, but their volumes are a combination of “traditional” Forex/CFD trades and spread betting, making apples-to-apples comparisons more difficult.]
A few quick observations from the data table above:
- Market Makers do earn more revenue per volume. The two Market Maker firms, Gain and FxPro, both earn about 2.7 pips per (round-trip) client trade, versus just 2.2 pips for Agent firm FXCM. Based on other research and surveying we’ve done in the industry, those numbers are indeed representative of a 20%-30% revenue premium per-volume earned by Market Maker firms.
- Market Maker firms are more profitable. While Gain Capital’s 20% net profit margin is below FXCM’s 31%, we believe that FxPro’s 41% net profit margin is really more indicative of what Market Maker firms can achieve if everything is running properly. Another data point supporting this conclusion is IG Group in the UK, which as a Market Maker over the past two years has been achieving 34%-35% net profit margins.
- Clients at Agent Firms are no more profitable than at Market Maker firms. Clients at Agent firm FXCM (24% of accounts profitable) did a little better than at Market Maker firm Gain (22%) over the first 9 months of 2010, although that reversed in Q4-2010, where FXCM clients (23%) did much worse than Gain clients (28%). Note that the overall industry average of 28%-29% profitability is made up mostly of Market Maker firms, putting clients at Agent firm FXCM well behind the industry average. [Note – kudos to the people at Forex Magnates for compiling and reporting this data].
So, if Market Maker firms earn more revenue-per-trade, and are more profitable than Agent firms, why do clients of Market Maker seem to be no worse off (or even better off) than clients at Agent firms? We offer a few possible explanations:
1) Less mouths to feed. An Agent firm means, by definition, that someone other than the eventual liquidity provider / market maker (providing the tradable quotes) also needs to earn a fee. While the Agent firm often receives “wholesale” quotes from its liquidity providers, once you add on the Agent’s fee (in the form of additional spread) the total spread to the client is, on average, inferior to the quotes provided by Market Maker firms. Simply put, in a Market Maker situation only the Market Maker needs to make money to run a profitable business. With an Agent firm both the Agent and the liquidity provider to that Agent need to make money. And who pays in the end? The client.
2) Less variability in spreads. Several (but not all) Market Maker firms offer their clients fixed spreads. As “market” spreads widen out during times of high volatility or slow volume, Market Makers often keep their spreads fixed or tighter than the market, while Agent firms (by definition) just pass on the wider “market” spreads to their clients. The result can be the closing out of client positions at a loss (or the triggering of Stop orders) at firms where the spreads widen, even if the market does not appreciably move.
Again, these are just theories we posit in trying to explain some of the data observed. We would welcome your comments on the topic as well – please send us your comments on the Feedback Section of our website, or email us at firstname.lastname@example.org.