Paper Authors Kee-Hong Bae, Rene M. Stulz, and Hongping Tan

Paper Title: Do Local Analysts Know More? A Cross-Country Study of the Performance of Local Analysts and Foreign Analysts

Paper date: October 2005

Working Paper Number: 11697

Paper Website

Student Author: Matt Hostetler

Review date: 2005-11-26

Revision date: 2005-12-12

Local analysts more accurate in forecasting earnings

In Do Local Analysts Know More? A Cross-Country Study of the Performance of Local Analysts and Foreign Analysts, Bae, Stulz and Tan set out to investigate if local analysts have an advantage over foreign analysts in forecasting earnings. Out of the 32 countries represented in their sample there was a positive local analyst advantage in 26 of those countries. It may be the case that local analyst’s have more access to word-of-mouth information or have some kind of relationship with insiders of the firm giving them access to information the public and foreign analysts don’t have access to.

The local analyst advantage is defined as the difference between the accuracy of local analysts and the accuracy of foreign analysts. To examine if this advantage exists a sample is constructed using the S&P’s Transparency and Disclosure dataset (TD dataset). This dataset contains information on firms from over 40 countries. The dataset also provides rankings of each firm’s corporate reporting practices, which allows the authors to examine if information asymmetries because of more information disclosure are a cause of the local analyst advantage. The data chosen for the sample from the TD dataset had to meet these three requirements:

  1. The countries selected must have more than 50 firms followed by I/B/E/S analysts (The Institutional Brokers Estimate System (I/B/E/S) is a unique service which monitors the earnings estimates on companies of interest to institutional investors. The I/B/E/S database currently covers over 18,000 companies in 60 countries.)
  2. The most recent earnings estimates are used for each analyst’s forecasts in each year for each firm.
  3. Only firms with foreign and local analysts are selected for the sample.

Their final sample contained 2,563 local and 1,920 foreign analysts providing a total of 20,425 annual earnings forecasts on 611 firms from 32 different countries. The data showed a difference in forecast accuracy between local analysts and foreign analysts of 7.8% expressed as a percentage of the average price-scaled (scaled by the latest monthly stock price) forecast error. The magnitude of the local analyst advantage varied from country to country. Out of the 32 countries represented in the sample the local analyst advantage was positive in 26 countries, and of those, 10 countries had a significantly positive analyst advantage.

The authors then try to determine the underlying reason for this local analyst advantage. Using information from the S&P Transparency and Disclosure index, the authors found that there was less of a local analyst advantage when the quality of information disclosed by a firm was above average (above the sample median of quality of information). More specifically, after running multiple regressions they conclude that the local analyst advantage is strong in countries where:

  1. Information disclosure is weaker.
  2. Institutional investors are less important, because when they are important it usually requires firms to disclose more information.
  3. Corporate ownership is more concentrated, more information isn’t needed to be made as accessible because owners are local
  4. US investor participation is low, indicating the quality of information disclosed is poor and US investors don’t know enough about the companies.

Because the local analyst advantage is found to be strong in countries that are underweighted in US portfolios, it can be assumed that the same information asymmetries that cause local analyst advantage, also contribute to the reason why investors tend to invest more heavily in domestic firms (the home bias). When company information isn’t made known to the public and the investments are riskier due to the lack of information, less people are likely to invest. In conclusion, the underlying reason for the local analyst advantage found in some countries seems to stem from information asymmetries, as local analysts apparently have access to some kind of information that foreign analysts do not have access to, which leads to more accurate forecasting.

Does real distance matter?

The empirical evidence presented supports the case for the local analyst advantage. However, the paper doesn’t address the issue of real distance, observations are only made from country to country. If real distance were playing a role, then looking solely at the United States would show a local analyst advantage. By gathering and examining the data the way they did it may simply be cross country information barriers that are giving local analysts this advantage as oppose to actual distance from the company headquarters.

I would be interested in finding out exactly why local analysts predict better. We know that when more and better information is released by a company the local analyst advantage is less, but what kind of information, when a company’s disclosure of information is poor, leads local analysts into making more accurate predictions. Is it because local analysts have close relationships with the owners? Do they understand more about how businesses operate in their region? Maybe it is because information of executive operations spread by word of mouth to local analysts. These are some examples of topics that I would like to see explored further.

Also, I would be interested to learn how this local analyst advantage has evolved over time. It is possibly the case that because of advances in communication and globalization, the local analyst advantage was historically much larger. If there has indeed been a decline in local analyst advantage, we could then predict when, if at all, this advantage would diminish all together.

Back to Matt’s Papers

Back to Titles of All Student Papers