Bonds Are Profitable For Dealers

Bonds Are Profitable For Dealers

In sharp contrast to equity market makers, government bond market makers realize positive positioning revenues.

That's one finding from a study co-authored by Oliver Hansch, assistant professor of finance in Penn State's Smeal College of Business Administration. The study, "Is Order Flow Information Valuable? Evidence from Government Bond Dealer Revenues," was presented at the Federal Reserve Bank of St. Louis. Hansch co-authored the paper with Victoria Saporta of the Bank of England.

"Market makers' ability to make money on their positions raises the question whether the market structure is too opaque, and whether more information-such as order flows and recent transactions-should be made publicly available to level the playing field between investors and market makers," says Hansch.

The paper investigates whether private information that is unrelated to fundamentals, such as information about past prices and volumes in a less than fully transparent market, may be valuable as well. Specifically, the researchers studied the spreads and composition of market maker revenues in the United Kingdom's government bond (gilt) market. For their research, they drew upon a unique combination of intra-day transaction data as well as daily inventory data on gilts and related interest rate derivatives.

"We find that, at the industry level, market making generates significantly positive margins," says Hansch. In the study, Hansch and Saporta investigate the sources of revenues of market makers in one of the world's most important bond markets: the market in the U.K. gilt-edge securities. They used data on all gilt transactions over a period of eight months as well as daily market maker position data on gilts and Exchange-traded derivatives instruments.

"The market structure of the U.S. government bond market is very similar to the U.K.'s and the same investment banks making markets in gilts are acting as primary dealers in the U.S. It may be possible to learn some lessons about the U.S. market from this study as well," says Hansch.

Hansch explains that a market maker's trading revenue generally stems from two sources, the bid-ask spread (spread revenues) and the appreciation of their portfolio over time (positioning revenues). A market maker's spread revenue will almost always be positive when measured over a time. Where losses to informed traders manifest themselves is in positioning profits.

"If a market maker faces informed traders, then they will on average lose money on the positions they acquire through trading, because they tend to buy overpriced and sell underpriced assets," says Hansch.

In the absence of any kind of private information, positioning revenues should be zero on average. If the market makers, Hansch says, take on a position, it should be equally likely to appreciate and to depreciate subsequently. If, however, the market maker possesses private information-such as order flow information-they should be seen as smart traders and are expected to make subsequent positioning profits.

This is believed to be the first study of any market that takes into account estimates of trading revenues in actively traded derivative instruments in the estimation of market maker gross trading revenues. It also provides the first analysis of the magnitude, distribution and source of dealer trading revenues, and is the first detailed documentation of government bond market trading activity that includes the universe of transactions.

"Our results suggest that in sharp contrast to London equity market makers, gilt-edged market makers realized positive gross trading revenues that exceeded traded spread estimates over our sample period," says Hansch.

Similar to equity market makers, gilt-edged market makers (GEMMs) make lower trade revenue margins in medium size trades than in small and very large trades, with retail trades offering the highest margins. Hansch notes that market making in the less liquid segment of the market was far less profitable than trading in the so-called "benchmark" liquid issues.

"Market makers in the gilt market incurred positive positioning revenues accounting roughly for 40 percent of their total trading revenues during our sample period. Our analysis suggests that trading in gilts is concentrated in a few actively traded stock with a small number of gilt-edged market makers participating in the majority of business," says Hansch.