Analyst Recommendations And International Stock Market ReturnsVW Staff
Analyst Recommendations And International Stock Market Returns
University of Auckland – Business School
University of Auckland – Business School
July 1, 2016
This paper shows that analyst recommendations aggregated at the country level predict international stock market returns. A trading strategy based on past country-level recommendations yields an abnormal return of around 0.9 percent per month. Aggregate analyst recommendations also predict one-quarter ahead gross domestic product, industrial production and one-year ahead aggregate earnings. Overall, our results suggest that analyst recommendations aggregated at the country level provide useful information to guide international asset allocation.
Analyst Recommendations And International Stock Market Returns – Introduction
This paper examines whether analyst recommendations aggregated at the country level have predictive value. We define the aggregate recommendation for a country as the average of all outstanding recommendations for shares listed in the domestic stock market. By averaging across firms, we eliminate firm-specific information contained in analyst recommendations and obtain a country-level recommendation. This country-level recommendation is used in our main trading strategy which consists of buying the MSCI market index (in USD) of countries in the quintile with the highest average recommendation and selling the MSCI market index of the quintile of countries with the lowest average recommendation. Depending on the international asset pricing model used, this strategy yields an abnormal return of 0.6 to 0.9 percent per month.
Prior studies show that analyst recommendations provide valuable information at the firm level. For example, Barber, Lehavy, McNichols, and Trueman (2001) find that buying stocks with the most favourable consensus analyst recommendations and short selling stocks with the least favourable consensus recommendations, yields annual abnormal returns of more than four percent. Jegadeesh, Kim, Krische, and Lee (2004) show that changes in consensus recommendations over the previous quarter predict future returns of individual firms. A small number of studies take an international perspective. For example, Jegadeesh and Kim (2006) evaluate the value of analyst recommendations in the G7 countries. They show that calendar time trading strategies that buy upgraded stocks and sell downgraded stocks are profitable in six of the seven countries.
This study extends the literature on analyst recommendations by examining the information content of analyst recommendations at the country level. Our research is related to Howe, Unlu, and Yan (2009) and Boni and Womack (2006). Howe et al. (2009) show that changes in aggregate analyst recommendations for US stocks forecast future excess returns for the US stock market. However, their evidence with regard to the ability of industry-aggregated analyst recommendations to predict industry returns is substantially weaker. Examining the same issue, Boni and Womack (2006) conclude there is no predictive value in industry-aggregated analyst recommendations.
We use IBES analyst recommendations for stocks from 33 different countries for the period from January 1994 to June 2015, to construct a monthly country-level recommendation measure. In our base case, we focus on the average consensus forecast using three-month outstanding recommendations and one-month-ahead stock market returns. Buying the market index of the quintile of countries with the highest average recommendation and selling the market index of countries with the lowest average recommendation yields a monthly abnormal return of 0.9% based on the international asset pricing model in Brusa, Ramadorai, and Verdelhan (2015) and a monthly abnormal return of 0.7% based on the international five factor asset pricing model in Fama and French (2015b). Additional tests show these results are robust to changes in research design and also hold in the more recent period after the regulatory changes that affected the brokerage industry around the world in 2002 and 2003.
To provide insight into the channel through which analyst recommendations predict international stock market returns, we examine whether country-level recommendations predict future growth in gross domestic product (GDP) and industrial production (IP), and whether country-level recommendations predict aggregate earnings growth for the countries in our sample. We find a highly significant relation between country level analyst recommendations and next quarter’s GDP growth and next quarter’s IP growth in a model that includes country fixed effects. To construct measures of aggregate annual earnings changes for the countries in our sample, we follow Kothari, Lewellen, and Warner (2006). We relate these aggregate earnings measures to country-level recommendations and show that firms in countries for which analysts have a more optimistic view tend to experience higher aggregate earnings growth. Again, these results hold after controlling for country- and year fixed effects. Overall, our results suggest that aggregate analyst recommendations contain information that predicts unexpected changes in cash flows at the country level, and that this is an important reason why country-level recommendations predict international stock market returns.
We contribute to the literature by showing that aggregate analyst recommendations for individual countries contain information about the cross-section of future international stock market returns. We also show that country-level recommendations predict GDP-growth, IP-growth and aggregate earnings growth. We thus contribute to a small emerging literature that focuses on the information content of firm-specific variables that are aggregated at the market level (see for example, Anilowski, Feng, and Skinner (2007) for earnings guidance; Kothari et al. (2006) for aggregate earnings surprises; and Hirshleifer, Hou, and Teoh (2009) for aggregate accruals and aggregate cash flows). In line with these studies, we show that by aggregating firm-level information (in our case at the country level), useful information that is not yet reflected in prices can be obtained.
See full PDF below.