In the Know
IN PRINT ARCHIVE CIR Winter 2007
By Gregory H. Bauer, Research Adviser, Bank of Canada
If we think of the many roles of fund managers, one of their primary activities is to allocate funds across a wide variety of assets. To do this, they need estimates of both: (i) the expected returns on the assets; and (ii) the covariance matrix of the assets. The latter contains both the volatilities of the individual assets and the correlations between them. Research has shown that both the volatilities and the correlations are changing over time. They are thus conditional volatilities and correlations: investors must use current information to forecast them when making portfolio decisions.
Yet, despite many years of academic research, identifying the reasons why volatilities and correlations change over time remains problematic. This is especially true in international stock markets. Research has shown that identifiable public news (e.g. new publicly available information about the economies of the countries or earnings of the individual firms) has very small effects on the means and variances of international stock returns. Nevertheless, there is some factor that causes international stock markets to move together. This missing factor could be based on rational, but not yet discovered reasons, or perhaps it arises from irrational trading on the part of global investors.
For example, if a hedge fund manager hires many PhDs and runs a sophisticated trading model, the manager may gain sophisticated information about the returns of the stocks being examined. Obviously, such resources are not available to the average investor and thus the hedge fund manager has private information when compared to the average investor. When the hedge fund manager trades stocks based on this model, the change in the prices of the stocks will (eventually) reveal the private information to all of the market participants. Private information cannot stay private forever.
If large investors who trade international equities have private information about the prices of stocks from many countries, this could be part of the missing factor that explains why these stocks move together. In research with Professor Rui Albuquerque (Boston University) and Professor Martin Schneider (New York University), I have been exploring how to extract global private information from the trades of international investors. Global private information could arise in a number of way. One example is top-down analysis where hedge fund managers have proprietary macroeconomic models that give good signals for the economies of several countries. Global private information could also arise from a bottom-up analysis where large financial institutions in aggregate disperse private information from many markets.
Our work shows that such information is important. We find that two-thirds of all trades in international equities by U.S. investors are driven by private information and that half of all these trades are due to global private information. Thus, one-third of all U.S. transactions in international equity markets are due to global private information. While this does not completely explain international stock market covariation, it does go partway to resolving the puzzle. We are continuing to work in this area with the hopes of further explaining this phenomenon.
*The views in this article are those of the author and do not reflect those of the Bank of Canada.
To view Gregory H. Bauer's presentation, click here.