Introduction

The accounting literature suggests that companies that consistently make detailed, timely, and informative disclosures obtain numerous market benefits, for example, lower cost of debt capital (COD) (Mazumdar & Sengupta, 2005; Verrecchia, 1983). If the negative relationship between financial disclosure and COD (i.e. the higher disclosure is, the lower is the COD) seems to be clear, the effects of non-financial disclosure are not so univocal because the content of any specific information can differently influence appraisal of risk (Sengupta, 1998). Among all non-financial information, that on corporate social responsibility (CSR) is certainly interesting (Dhaliwal, Li, Tsang, & Yang, 2011), because of its increased demand from stakeholders and the current trend toward CSR (Arvidsson, 2011). In particular, the cost of capital can reflect CSR risks and benefits (Weber, 2008). For example, firms might benefit from lower interest rates (Goss & Roberts, 2011) and higher access to debt capital through CSR (Cheng, Ioannou, & Serafeim, 2014). Following a combined approach aimed at integrating economic theories (agency cost of debt and voluntary disclosure) with social theories (reputation and stakeholder), we expect companies receive financial benefits by acting responsibly. This study enters the debate from a unique perspective by examining the impact of a ready-to-use corporate social performance (CSP) index, which is the Datastream/ASSET4, on both the COD and access to debt financing.1 The use of a commonly available source of information allows us to analyse the link between a flow of information produced by a third party (i.e. Thomson Reuters Datastream) and used by the market, and the cost of debt of a firm. We use two proxies of cost of debt, namely, the interest rate and debt rating (hereafter, ‘rating’) and a measure of debt access (‘leverage’). We analyse this relationship through a sample of European firms listed in the 17 countries which are included in the Standard and Poor’s (S&P) Europe 350, from 2005 to 2012. Our results show a negative relationship between interest rate and CSP. Likewise, there is a positive relationship between the rating and CSP. In addition, we find CSP positively impacts debt capital access. Moreover, since during the 2008 global financial crisis companies faced the difficulties of both higher interest rates and credit constraints, we aim to investigate how the credit crisis affected the association between CSP and both COD and debt capital access. Our analysis confirms prior results (Hoepner, Oikonomou, Scholtens, & Schroder, 2016 € ), which find no evidence of the link between CSP and COD during the financial crisis years. The results obtained from multivariate analysis are robust to different models’ specifications and sensitivity tests. This study makes three contributions to the existing literature. First, it examines the impact of social performance on the cost of debt, adding new empirical evidence in a research field that has controversial results. Second, this study considers a cross-country European sample, which has rarely been undertaken in the literature to date (Girerd-Potin, Jimenez-Garces, & Louvet, 2014; Menz, 2010; Perrini, 2005). Third, the study provides evidence of the impact of social performance on the cost of debt when considering a time horizon that includes the ongoing financial crisis. The remainder of this paper is organised as follows. Section 2 reviews the relevant literature and develops the hypotheses. Section 3 discusses the research methodology. Section 4 presents our findings and Section 5 draws conclusions and implications.