One of the most treated and traditional subject in financial literature is stock return predictability. Many papers are written in which different variables are constructed and theories are made. Still, there is no clear agreement between researchers and financial experts about the ability of different models and variables to predict stock returns. Therefore, I want to make a contribution to the literature by researching a new variable that is recently constructed and examined in the United States.
My research will mainly be based on Jefferson Duarte and Nishad Kapadia’s paper: ‘Davids, Goliaths, and Business Cycles’ (2014). In this paper the writers construct a simple variable that is called Goliath versus David (GVD). This variable reflects the annual change in weight of the stocks of the largest firms in the aggregate market portfolio. GVD seems able to predict market returns in the US and is apparently the best single predictor, even out of sample, in comparison to the traditional ones. Further, they found that GVD is the only single predictor that is also capable to predict some macro-economic variables. Later I will explain in more detail how this variable is constructed and what it will predict.
In my research I want to examine if GVD is able to predict European market returns. The reason that I do this research in European countries is because Duarte and Kapadia did their research in the US and I think it is interesting to see how this constructed GVD performs in Europe. I want to know if the GVD forecasting ability is better than the traditional variables and the benchmarks chosen. Thus: Do European equity markets reflect the forecasting implications of the Financial Accelerator Hypothesis, using Duarte and Kapadia’s Goliath versus David variable? This will be done in four Western European countries, namely: France, Germany, Italy and the United Kingdom. Those are all countries with different structured economies and differences with respect to large and small firms. Therefore it is interesting to find differences of the GVD variable in those countries.