ABSTRACT
This paper performs topic modeling using all publicly available CSR (Corporate Social Responsibility) reports for all constituent firms of the major stock market indices of 15 industrialized countries included in MSCI Europe for the sample period from 1999 to 2016. Our text mining results and LDA analyses indicate that ‘employees safety’, ‘employees training support’, ‘carbon emission’, ‘human right’, ‘efficient power’, and ‘healthcare medicines’ are the common topics reported by publicly listed companies in Europe and the UK. There is a clear sector bias with industrial firms emphasizing ‘employee safety’, Utilities concentrating on ‘efficient power’ while consumer discretionary and consumer staples highlighting ‘food waste’ and ‘food packaging.’ To produce these results, we used a battery of python code to organize the hundreds of reports downloaded from Bloomberg and the internet, the latest R-algorithm to estimate LDA (Latent Dirichlet Allocation) model and the LDAvis interactive tool to visualize and refine the LDA model.
Acknowledgments
We would like to thank Chris Adcock, Martin Carpenter, Andreas Hoepner, and participants at the workshop on ‘Recent Developments in Econometrics and Financial Data Science’ at ICMA Centre, Reading University in November 2017 for helpful comments. The research funding from DigitalHumanities@Manchester is gratefully acknowledged.
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
Irina Goloshchapova http://orcid.org/0000-0001-9201-4327
Ser-Huang Poon http://orcid.org/0000-0002-7297-9401
Phil Reed http://orcid.org/0000-0002-4479-715X
Notes
1 See https://www.unpri.org/about/the-six-principles. The first three principles center on incorporating ESG issues in investment and decision making, ownership policies and practices, and ESG disclosure, while the last three focus on promoting ESG principles in the industry, implementation and report on progress.
2 The Pensions Act 1995 and 2005 investment regulations place a range of investment duties on trustees. For example, trustees must secure appropriate asset diversification and invest in a manner ‘calculated to ensure the security, quality, liquidity, and profitability of the portfolio.’ Trustees are also required to include a statement in the scheme's Statement of Investment Principles (SIP) regarding the extent they take ESG and ethical considerations into account.
3 For ESG reporting, these organizations include most notably GRI, SDG, IRRC, and WBCSD. For special interest groups on climate and environment, there are CDP, CDSB, TCFD, the CEO Water Mandate, ICMM, IIGCC, and CERES. For country based jurisdiction, there are SASB who works closely with SEC, UK's FRC, and a whole barrage of professional organizations such as IR Society, IFC, IFAC, CCI, WICI, COSO, EFFAS, and DVFA.
4 In this paper, we will use the two terms, CSR and ESG, interchangeably.
5 The six trends are: (i) the shift towards longer-term investment policy by incorporating, e.g., climate risks or social inequality, or by tilting towards more resilient assets with a longer investment horizon; (ii) the increased awareness of physical risk, especially that relates to, e.g., climate change, and the encroaching scarcity of water; (iii) a growing trend in Asia for the level of the commitment to ‘Responsible Investment’ to match its Western counterparts; (iv) the increased importance of ESG factors as a performance indicator; (v) the new ambition to invest with real impact and the disclosure of such an intention or information to help achieve Sustainable Development Goals; and (vi) the surge of innovation in sustainable development projects and initiatives in China, India and other emerging markets to attract foreign capital.
6 Austria ATX 20, Belgium BEI 20, Denmark OMXC 20, Finland OMXH 25, France CAC 40, Germany DAX 30, Iceland ISEQ 20, Italy FTSEMIBN 40, Netherland AEX 25, Norway OBX 25, Portugal PST 20, Spain IBEX 35, Sweden OMX 30, Switzerland SMI 20, and UK UKX 100.
7 The SIC code represented are 10: Energy, 15: Materials, 20: Industrials, 25: Consumer Discretionary, 30: Consumer Staples, 35: Health Care, 40: Financials, 45: Information Technology, 50: Telecommunication Services, 55: Utilities, 60: Real Estate, and 99: Funds & Investment Co.
8 LDAvis R package is a web-based interactive visualization of topics estimated using Latent Dirichlet Allocation that is built with D3.js, a very popular javascript visualization tool (see Sievert and Shirley (Citation2014)).
9 There is a lack of word count for topic 4 in Real Estate in the later year after an upsurge just before 2010.
10 The dip in 2017 is probably due to a reporting lag since our data collection stopped in July 2017.