Portal de Periódicos da CAPES

Corporate social responsibility and corporate performance: evidence from a panel of US listed companies

2008; Taylor & Francis; Volume: 40; Issue: 5 Linguagem: Inglês

10.1080/00036840500428112

1466-4283

Leonardo Becchetti, Stefania Di Giacomo, Damiano Pinnacchio,

Corporate Finance and Governance

Abstract We investigate whether inclusion and permanence in the domini social index (DSI) affects corporate performance on a sample of around 1000 firms in a 13-year interval by controlling for size, industry, business cycle and time invariant firm idiosyncratic characteristics. Our results find partial support to the hypothesis that corporate social responsibility is a move from the shareholders wealth to a multi-stakeholders welfare target. On the one side, permanence into the domini index (DI) is shown to increase (reduce) significantly total sales per employee (returns on equity but not when large and R&D investing firms are excluded from the sample). On the other side, lower returns on equity for Domini firms seem nonetheless to be accompanied by relatively lower conditional volatility and lower reaction to extreme shocks with respect to the control sample. An explanation for these findings, suggested by the inspection of Domini criteria, is that social responsibility implies, on the one side, decisions leading to higher cost of labour and of intermediate output, but may, on the other side, enhance involvement, motivation and identification of the workforce with company goals with positive effects on productivity. Acknowledgements Article presented ad the XIII Tor Vergata Financial Conference. Support from the Veritatis Splendor Research Project on SR is acknowledged. The authors thank Helen Alford, Michele Bagella, Laura Boccardelli, Saverio De Santo, Iftekhar Hasan, James Lothian, Ferruccio Marzano, Francesco Nucci, Alberto Pozzolo for useful comments and suggestions and Osea Giuntella for his precious research assistance. The usual disclaimer applies. Notes 1'Few trends could so thoroughly undermine the very foundations of our free society as the acceptance by the corporate officials of a social responsibility other than to make as much money for their shareholders as possible.' (…) If businessmen do have social responsibility other than making maximum profits for stockholders, how are they to know what it is? Can self private individuals decide what the social interest is? (Friedman, Citation1962) 2According to a definition of Fehr and Falk (Citation2002) 'a person exhibits social preferences if it does not only care about the material resources allocated to it but also cares about the material resources allocated to other relevant reference agents'. 3The index methodology presents advantages and drawbacks. Its advantages are that it reflects historical concerns of investors, keeps track of CSR evolution in time and includes all dimensions identified as important in CSR. Its first limit is the absence of a measure of intensity in corporate performance. Another important problem with the index is in the adoption of a 'best in class process' in which relative, but not absolute, best SR performers in some industries have been included with the specific aim of keeping the index sufficiently diversified, thereby allowing ethical fund investors to adopt well-diversified passive investment strategies. Finally, a third limit is that the index must have a constant number of constituents. Therefore, for any exit a new entry is needed, with the effect that entry timing is determined not solely by firm progress on CSR, but also by rebalancing needs. 4The weight of these funds in financial markets is growing considerably. According to 2003 Report on Socially Responsible Investing Trends in the United States, the industry of ethically managed mutual fund assets represented $2.16 trillion dollars when including all US private and institutional ethically screened portfolios. Based on these figures one out of nine dollars under professional management in the United States was part of Socially Responsible portfolios. The same report illustrates that, from 1995 to 2003 the rate of growth of assets involved in social investing, through screening of retail and institutional funds, shareholder advocacy and community investing has been 40% higher than all professionally managed investment assets in the USA (240 against 174%). 5Return on capital employed is equal to Operating income/(Shareholders' equity + Interest bearing liabilities). Its advantage is that it includes in the denominator and indicator which depends on firm indebtedness and therefore, does not suffer, like ROE, of sensitiveness to firm leverage (i.e. highly leveraged firms tend, by definition, to have significantly higher ROE than nonhighly leveraged ones). 6Among authors emphasizing the importance of testing for panel cointegration to avoid spurious regressions in panel estimates see Okunade and Karakus (Citation2001) and Gerdtham and Lothgren (Citation2002). The latter also provide one of the earlier applications of the Im et al . (2003) test which is performed also in this article. For an application of panel cointegration to financial ratios see Peel et al . (Citation2004). 7Omitted for reasons of space and available from the authors upon request. 8The test does not require model estimates because is based on the rank of covariance matrix of the disturbances driving the multivariate random walk. If this rank is equal to a certain number of common trends, this implies the presence of cointegration and vice versa. If the rank is equal to zero, as in the null hypothesis, then there are no common trends among the variables. Thus, failure to reject the null hypothesis of zero common trends is also an indication that the variables do not form a cointegrated combination. 9Preliminary tests on normality and Autoregressive Conditional Heteroschedasticity (ARCH) LM tests show that observed returns are nonnormal and have an ARCH structure. More complex (asymmetric, nonlinear) conditional heteroskedasticity models have also been estimated giving results which are not substantially different in terms of the effect of SR. Results are omitted for reasons of space and available upon request. 10Community development financial institutions primarily provide loan financing to businesses in areas that need economic development. CDFIs make loans that are generally 'unbankable' by traditional industry standards because of past credit problems, the size of the loan request, limited equity from founders or limited collateral. 11Exclusionary screens include military-weapons, alcohol, tobacco, firearms, nuclear power and gambling.