Transfer spending and sports performance on the example of selected European football clubs. The Granger causality approach

Abstract

The issue of causality in economics can be analyzed in different ways. There are many econometric methods that can be applied to study the causal relationships between the variables. A new and still not explored field is the sports economics, where the researchers consider the relationships between clubs’ financial and organizational performance. To understand these phenomena, the authors use various statistical and econometrical methods. It is clearly visible in the growing number of publications in this area. In this paper, author attempts to investigate and describe the Granger-causality concept in finance-sports performance of selected football clubs. The following paper aims to identify causality and its direction between players’ transfer values and clubs’ sports performance in football. For the purpose of the thesis, the following hypotheses had been developed. H1: there exists a causal relationship between transfer spending and points of the selected football clubs; H2: there exists a causal relationship between transfer spending and standings of the selected football clubs. The research concerns the analysis of the causality relationship between financial and sports performance on the example of five best teams from five best European football leagues. All selected clubs characterize the best sports performance in their native leagues. In the study, three variables have been used. One of them has financial nature, while two relate to sports performance. Data reflect the two rounds that appear during every football season. The research includes data from the first round of season 2003/2004 to the first round of season 2020/2021. It contains information about transfer values, the standings and points gathered by the clubs in a semi-year frequency. The research tool used for the study is GRETL software. Thank to this solution it is possible to test stationarity and develop the VAR model. This relationship between pairs of selected variables occurs only in a small percentage of cases, or it cannot be identified at all. As a final concussion, the weak utility of the approach for the practical purpose is suggested. The last part of the paper is a conclusion where the author put the list of limitations and recommendations for further researches.