Within certain markets, contracts for difference (CFDs) have experienced marketing-fueled popularity and a commensurate level of regulatory attention. Despite this popularity—and notoriety—academic research regarding CFDs has been limited because of the dominance of opaque proprietary OTC networks. The authors focus on CFDs listed on the Australian Securities Exchange to examine CFD investor trading performance.
What’s Inside?
Contracts for difference (CFDs) are recently developed financial derivatives that allow an investor to pay the counterparty the difference between the current and initial value of the contract. Contract values are set with reference to an underlying security price. For CFDs listed on the Australian Securities Exchange (ASX), trades and quotes are transparent.
Not only are there regulatory concerns about CFDs, but also the literature on stocks and derivatives paints a picture of poor individual investor performance and even gambling-motivated behavior. But the authors find some evidence of individual CFD investor performance and sophistication, particularly in the short term.
How Is This Research Useful to Practitioners?
Using percentage return–based measures, the authors find that for a one-day (up to the next day’s settlement price) holding period, “portfolios” of market order buy trades significantly outperform (p 0.05) those of sell trades, inclusive of the bid–ask spread but not after financing costs. For one-month, half-year, and one-year holding periods, buy trades fail to outperform sell trades before financing costs and significantly underperform them (p 0.01) after those costs.
The authors also find that for small trades ($10,000 nominal) with a one-day holding period, buys significantly outperform sells (p 0.10) even after financing costs. This result is comparable with the buy versus sell outperformance that they find for large trades (>$20,000 nominal) with an intraday (until settlement price) holding period.
The authors examine market-timing returns of CFD trades and find that they significantly fall short (p 0.01) of the risk-free rate for the half-year and one-year holding periods, even before financing costs. After financing costs, the market-timing returns of CFD trades also significantly fall short (p 0.05) for the one-month holding period.
How Did the Authors Conduct This Research?
The authors obtain trade and quote data on ASX-listed CFDs and their underlying stocks from Thomson Reuters Tick History (TRTH). They use data from November 2007 to June 2010 for all 71 of the exchange-traded share CFDs from the initial listing on the ASX. Daily settlement prices and returns of the underlying stocks are obtained from the Securities Industry Research Centre of Asia–Pacific Core Research Database, and index returns are from TRTH.
In their analysis, the authors use buyer- and seller-initiated trades (i.e., market orders) as a proxy for investor performance. They determine whether trades are initiated by the buyer or seller by matching each trade to the prevailing bid and ask quotes, using a zero-second time delay. They use small signed trades to identify individual investor trades.
To measure investor performance, the authors use portfolios of buys minus sells and measure returns from the traded price. Because the authors use the traded price, they implicitly incorporate bid–ask spreads into their holding period return measures.
Financing costs consist of the Reserve Bank of Australia’s (RBA’s) target overnight cash rate and the Open Interest Charge (OIC). The RBA rate plus OIC is paid if an investor holds a CFD long position; the RBA rate less OIC is earned if an investor holds a CFD short position overnight. The OIC is fixed at 1.5% a year.
The authors examine investor performance using both percentage return–based and dollar-profits-earned measures, and they test for significance using Newey–West-adjusted t-statistics.
Abstractor’s Viewpoint
The findings presented by the authors indicate that CFD investor trading performance is not as poor as prior research and heightened regulatory awareness suggests, at least in the short term and in the case of exchange-listed CFDs. In other words, these investors are not total dummies.
Those seeking additional background and findings regarding CFDs should benefit from reading Brown, Dark, and Davis (Journal of Futures Markets 2010). As noted in that article, CFD research should be of interest to securities exchanges and regulators considering the introduction of similar products, as well as those interested in market microstructure.