Journal of Network Theory in Finance

Welcome to the first issue of Volume 4 of The Journal of Network Theory in Finance.

A paper by Duc Thi Luu and Thomas Lux, titled “Identifying patterns in the bank–sector credit network of Spain”, opens this issue and fills an important gap in the literature: the analysis of the topological properties of bipartite credit networks. This is a data-driven contribution that uses a data set of Spain’s undirected binary bank–firm credit network over the period 1997–2007 (including a weighted version), and looks at the deviation of the observed topological features from those expected under various sensible null models. This paper also investigates the role of the intrinsic heterogeneity in observed degrees and/or strengths on the higher-order properties of these networks. In the authors’ own words: “This paper not only contributes to the literature on the identification of higher-order patterns in real-world bipartite networks in general, but also provides some important implications for the reconstruction of real credit networks from limited information.” Indeed, the findings from this study provide important suggestions for additional extensions to the fitness-induced configuration models. This paper encourages further studies and future work on many other bipartite networks. Thank you, Duc Thi and Thomas, for inspiring us!

Our second paper, “Debt, information asymmetry and bankers on board” by João Amaro de Matos and João Mergulhão, is another valuable contribution to the literature on financial networks. Its main finding suggests that the connectedness of bankers plays a key role in reducing information asymmetry, and that firms with relatively low debt levels are more affected by this state of affairs. This paper suggests that firms can use connected bankers on the board to reduce information asymmetry. The presence of connected bankers is shown to increase (on average) the debt level of the analyzed US firms in a sample that includes board information for S&P 1500 firms from 1996 to 2013, with data on more than 11 000 directors per year and several other variables: this includes debt information from Compustat/CRSP. The paper is well written and provides a very clear message: bankers contribute to the reduction of information asymmetry, transmitting to the market their perception of debt capacity use. Results from this paper will be beneficial to both academics and practitioners.

The issue concludes with “News-sentiment networks as a company risk indicator” by Thomas Forss and Peter Sarlin. The third and final paper in this issue introduces a sentiment-based company risk indicator. The authors analyze crowdsourced news articles about companies and create networks in which nodes represent companies and links denote that the companies were mentioned together in the same paper (co-mention networks). They find that information extracted from these networks can be used to improve the predictions of stock movements on different time scales. This paper contributes to an emerging body of research analyzing large amounts of unstructured data for risk management or alpha generation in asset management. Its findings have the potential to be developed into a prescriptive risk tool that could be used to warn companies when they are at higher risk of stock price decreases: something that would be directly useful to finance industry experts such as portfolio managers, investors and traders.