Examining Relationship Of Banks Buying And Selling Of Government Bonds And Daily Price Changes

The buying and selling of government bonds by banks explains 25 per cent of the daily price changes in this market — meaning order flows can be used to predict future price changes, say new research.

Financial markets do not always behave the way classical economic theory prescribes. Among other things, this theory posits that new and relevant information that has a bearing on prices in a given market (whether they be bonds or shares) are immediately reflected in prices. This means that prices for, for example, Norwegian government bonds, should always reflect all relevant information.

The reality, however, is often quite different. Researchers are often fascinated by mysteries such as these, and the possibilities they present of finding new pieces in the financial puzzle.

Examines trading

For her doctoral thesis at the BI Norwegian Business School, financial researcher Siri Valseth has examined trading in Norwegian government bonds between 1999 and 2005, concentrating on the most active players during this time period, which constitute seven financial institutions.

Studies of the microstructure of financial markets show that new information is impounded into prices via two channels; one direct and one indirect.

Via the direct channel, prices change as an immediate response to new, public information, such as new figures on consumer prices or growth in the gross domestic product.

Via the indirect channel, prices are adjusted gradually, in line with purchases by market players who may possess private information that has a bearing on the market. Private information is information that is not available to everyone. This may be knowledge about circumstances that may affect investors’ risk willingness or liquidity in the market. Private information may also constitute market players’ qualified understanding of financial data.

25 per cent via indirect channels

For financial researchers, Norway, with its access to high-quality and detailed data, represents an ideal laboratory for gaining a better understanding of the mysteries of the financial markets.

Valseth has been interested in finding out how much of the daily price changes (measured by interest rates) in Norwegian government bond occurs via the indirect channel. To find the answer, she has followed order flows, which represent the sum of all transactions initiated by buyers and sellers in the course of one day of trading.

“Order flows explain one fourth of the daily change in interest rates in the Norwegian government bonds market. The market players contribute in different ways to this process, and the contributions they make depend not only on how big they are, but also on the degree to which they initiate trade with others.

Predicts changes

Valseth also proves that information that lies in the market players’ buying and selling of government bonds can be used to predict changes in the prices of bonds. Daily order flows can be used to predict bond rates for the following day, while monthly order flows can be used to predict monthly changes in interest rates: “If you trade according along these lines, you may be able to earn money,” Siri Valseth points out.

The financial researcher has also been able to examine the seven market players very closely, to find out whether they differ in any way. The order flows for different market players have varying abilities to predict changes in interest rates. “Some banks are better than others,” says Valseth, though without revealing who is best in the Norwegian government bonds market.

According to Valseth, there are two reasons why some banks are better than others: “One of them is that they have customers who possess private information. The other is that they are better at interpreting order flows and what effect public information has on future bond prices.”