Skip to main content
SpringerLink
Log in

Understanding CSR champions: a machine learning approach

  • Original Research
  • Published:
Annals of Operations Research Aims and scope Submit manuscript

Abstract

In this paper, we study champions of corporate social responsibility (CSR) performance among the U.S. publicly traded firms and their common characteristics by utilizing machine learning algorithms to identify predictors of firms’ CSR activity. We contribute to the CSR and leadership determinants literature by introducing the first comprehensive framework for analyzing the factors associated with corporate engagement with socially responsible behaviors by grouping all relevant predictors into four broad categories: corporate governance, managerial incentives, leadership, and firm characteristics. We find that strong corporate governance characteristics, as manifested in board member heterogeneity and managerial incentives, are the top predictors of CSR performance. Our results suggest policy implications for providing incentives and fostering characteristics conducive to firms “doing good.”

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. In this paper, we use CSR and ESG terms interchangeably as prior literature has pointed out that the only difference between the two terms is that ESG includes a governance component explicitly while CSR implicitly includes it into social responsibility of the firm (Gillan, Koch, & Starks, 2021).

  2. According to the search on Google Scholar platform conducted on 03/16/22.

  3. We performed a robustness check by excluding all firms that were dropped during the sample period. Specifically, we define dropped companies as voluntarily delisted companies (companies delisted at the company’s request) and mandatory delisted companies (companies delisted by exchanges). We identify those firms using CRSP delisting codes. We screen out firms with delisting codes in 400 range (“liquidations”) and the 500 range (“dropped”), excluding firms with delisting codes of 501–503 (“stopped trading on current exchange to move to NYSE, AMEX, or NASDAQ”). During our sample period, 83 out of 9825 firm-year observations fall into this category. We drop those 83 observations and reconstruct the seven machine learning models. The out-of-sample model performance is consistent with the main results and is presented in Panel B Table 3A of Online Appendix.

  4. As a robustness check, we conduct analysis on a subsample of firms that are not operating in the “sensitive” industries. Specifically, we follow Cho and Patten (2007) and define environmentally sensitive industries (ESI) as companies with a primary SIC code of 13xx (oil exploration), 26xx (paper), 28xx (chemical and allied products), 29xx (petroleum refining), or 33xx (metals). During our sample period, 1198 out of 9825 firm-year observations fall into this category. We redid the main analysis on the subsample excluding ESI companies and included the results in Online Appendix 5A. The results are consistent with our main results confirming that our analysis is not driven by only environmentally sensitive industries.

  5. We acknowledge that ESG scores are available from multiple data providers and, according to a recent study by Berg, Koelbel, & Rigobon (2022) those scores are based on different measurement techniques that result in less than perfect correlation among them. Thus, our results are not expected to be exactly the same across all ESG scores from various sources, although we can reasonably expect some positive correlation. Our chosen ESG scores dataset (Refinitiv/Asset4) has several advantages. The Refinitiv/Asset4 has the most comprehensive coverage with roughly the largest 3,000 companies in U.S. and across more than 450 different ESG metrics (Amiraslani, Lins, Servaes, & Tamayo, 2022). This data has been used widely used in prior studies in the field (Albuquerque, Koskinen, Yang, & Zhang, 2020; Dorfleitner, Kreuzer, & Sparrer, 2020; Drempetic, Klein, & Zwergel, 2020; Havlinova & Kukacka, 2023). This makes it a good fit for our purposes to ensure the credibility of results and consistency with the prior literature.

  6. Since the industry component is reflected in ESG scores, we do not control for the industry in constructing independent variable groups.

  7. Diebold-Mariano statistics for each pair of models are calculated according to the following steps. First, gather the out-of-sample predictive error for both models. Second, compute the absolute values of these errors and the mean of the difference of these absolute values. Third, compute the covariogram for lag/lead length of the out-of-sample prediction errors for the vector of the differences of the absolute values of the predictive errors.

  8. tenfold cross validation is applied when training each model.

  9. For example, the first model that predicts ESG scores for firms in 2012 is trained using 2007–2011 as a training and validation set. We repeat this process annually till the testing set reaches the end of the sample period, the year 2018.

  10. To make the plot clear, we normalize the feature importance of all variables so that the most important feature gets the value 1.

  11. The model construction methods of different algorithms are very different, which also brings about their different dependence on each predictor. Out of the seven models we tested, OLS, PLS, Lasso, Ridge, and ENet are linear models, while MARS and XGBoost are nonlinear models. The results in Fig. 1 show that ESG performance is highly dependent on corporate governance across all models. Generally speaking, feature importance is more informative for models with better out-of-sample predictive performance. Since feature importance indicates how much each feature contributes to the model’s prediction, the feature importance in the model with the best performance (XGBoost in our case, as shown in the results in Table 3A) is the most instructive.

  12. Additionally, to assess the explanatory power of the four groups of variables, we report the out-of-sample performance of models based on the subsets of variables in Online Appendix Table 4A.

  13. Since the model predicts annual company ESG scores, the proposed portfolio is reallocated annually at the end of the year, so we do not adjust for the momentum factor (Carhart, 1997).

References

  • Adams, R. B., Hermalin, B. E., & Weisbach, M. S. (2010). The role of boards of directors in corporate governance: A conceptual framework and survey. Journal of Economic Literature, 48(1), 58–107.

    Article  Google Scholar 

  • Athey, S., & Imbens, G. W. (2019). Machine learning methods that economists should know about. Annual Review of Economics, 11, 685–725.

    Article  Google Scholar 

  • Albuquerque, R., Koskinen, Y., Yang, S., & Zhang, C. (2020). Resiliency of environmental and social stocks: An analysis of the exogenous COVID-19 market crash. The Review of Corporate Finance Studies, 9(3), 593–621.

    Article  Google Scholar 

  • Amiraslani, H., Lins, K. V., Servaes, H., & Tamayo, A. (2022). Trust, social capital, and the bond market benefits of ESG performance. Review of Accounting Studies, pp. 1–42.

  • Bali, T., Goyal, A., Huang, D., Jiang, F., & Wen, Q. (2020). Different strokes: Return predictability across stocks and bonds with machine learning and big data. Swiss Finance Institute: Research Paper Series No., 20, 110.

    Google Scholar 

  • Bénabou, R., & Tirole, J. (2010). Individual and corporate social responsibility. Economica, 77(305), 1–19.

    Article  Google Scholar 

  • Berg, F., Koelbel, J. F., & Rigobon, R. (2022). Aggregate confusion: The divergence of ESG ratings. Review of Finance, 26(6), 1315–1344.

    Article  Google Scholar 

  • Bertrand, M., & Schoar, A. (2003). Managing with style: The effect of managers on firm policies. The Quarterly Journal of Economics, 118(4), 1169–1208.

    Article  Google Scholar 

  • Borghesi, R., Houston, J. F., & Naranjo, A. (2014). Corporate socially responsible investments: CEO altruism, reputation, and shareholder interests. Journal of Corporate Finance, 26, 164–181.

    Article  Google Scholar 

  • Campbell, J. L. (2007). Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of Management Review, 32(3), 946–967.

    Article  Google Scholar 

  • Cao, J., Liang, H., & Zhan, X. (2019). Peer effects of corporate social responsibility. Management Science, 65(12), 5487–5503.

    Article  Google Scholar 

  • Carhart, M. (1997). On Persistence in Mutual Fund Performance. Journal of Finance, 52(1), 57–82.

    Article  Google Scholar 

  • Chen, T., Dong, H., & Lin, C. (2020). Institutional shareholders and corporate social responsibility. Journal of Financial Economics, 135(2), 483–504.

    Article  Google Scholar 

  • Cho, C. H., & Patten, D. M. (2007). The role of environmental disclosures as tools of legitimacy: A research note. Accounting, Organizations and Society, 32(7–8), 639–647.

    Article  Google Scholar 

  • Cronqvist, H., & Yu, F. (2017). Shaped by their daughters: Executives, female socialization, and corporate social responsibility. Journal of Financial Economics, 126(3), 543–562.

    Article  Google Scholar 

  • Davidson, R. H., Dey, A., & Smith, A. J. (2019). CEO materialism and corporate social responsibility. The Accounting Review, 94(1), 101–126.

    Article  Google Scholar 

  • Deckop, J. R., Merriman, K. K., & Gupta, S. (2006). The effects of CEO pay structure on corporate social performance. Journal of Management, 32(3), 329–342.

    Article  Google Scholar 

  • Diebold, F. X., & Mariano, R. S. (2002). Comparing predictive accuracy. Journal of Business & Economic Statistics, 20(1), 134–144.

    Article  MathSciNet  Google Scholar 

  • Dimson, E., Karakaş, O., & Li, X. (2015). Active ownership. The Review of Financial Studies, 28(12), 3225–3268.

    Article  Google Scholar 

  • Dyck, A., Lins, K. V., Roth, L., & Wagner, H. F. (2019). Do institutional investors drive corporate social responsibility? International evidence. Journal of Financial Economics, 131(3), 693–714.

    Article  Google Scholar 

  • Dorfleitner, G., Kreuzer, C., & Sparrer, C. (2020). ESG controversies and controversial ESG: About silent saints and small sinners. Journal of Asset Management, 21(5), 393–412.

    Article  Google Scholar 

  • Drempetic, S., Klein, C., & Zwergel, B. (2020). The influence of firm size on the ESG score: Corporate sustainability ratings under review. Journal of Business Ethics, 167, 333–360.

    Article  Google Scholar 

  • Endrikat, J., De Villiers, C., Guenther, T. W., & Guenther, E. M. (2021). Board characteristics and corporate social responsibility: A meta-analytic investigation. Business and Society, 60(8).

  • Erel, I., Stern, L. H., Tan, C., & Weisbach, M. S. (2021). Selecting directors using machine learning. The Review of Financial Studies, 34(7), 3226–3264.

    Article  Google Scholar 

  • Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1), 3–56.

    Article  Google Scholar 

  • Fama, E. F., & French, K. R. (2015). A five-factor asset pricing model. Journal of Financial Economics, 116(1), 1–22.

    Article  Google Scholar 

  • Fabrizi, M., Mallin, C., & Michelon, G. (2014). The role of CEO’s personal incentives in driving corporate social responsibility. Journal of Business Ethics, 124(2), 311–326.

    Article  Google Scholar 

  • Ferrell, A., Liang, H., & Renneboog, L. (2016). Socially responsible firms. Journal of Financial Economics, 122(3), 585–606.

    Article  Google Scholar 

  • Flammer, C., Hong, B., & Minor, D. (2019). Corporate governance and the rise of integrating corporate social responsibility criteria in executive compensation: Effectiveness and implications for firm outcomes. Strategic Management Journal, 40(7), 1097–1122.

    Article  Google Scholar 

  • Freeman, R. E. (1984). Strategic Management: A Stakeholder Approach. Pitman.

    Google Scholar 

  • Friedman, M. (1970). The social responsibility of business is to increase its profits. New York Times Magazine 13 September: 32–33.

  • Gillan, S. L., Koch, A., & Starks, L. T. (2021). Firms and social responsibility: A review of ESG and CSR research in corporate finance. Journal of Corporate Finance, 101889.

  • Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. The Review of Financial Studies, 33(5), 2223–2273.

    Article  Google Scholar 

  • Gupta, A., Fung, A., & Murphy, C. (2021). Out of character: CEO political ideology, peer influence, and adoption of CSR executive position by Fortune 500 firms. Strategic Management Journal, 42(3), 529–557.

    Article  Google Scholar 

  • Hastie, T., Tibshirani, R., & Friedman, J. H. (2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). Springer.

    Book  Google Scholar 

  • Hambrick, D. C., & Mason, P. A. (1984). Upper echelons: The organization as a reflection of its top managers. Academy of Management Review, 9(2), 193–206.

    Article  Google Scholar 

  • Havlinova, A., & Kukacka, J. (2023). Corporate social responsibility and stock prices after the financial crisis: The role of strategic CSR activities. Journal of Business Ethics, 182(1), 223–242.

    Article  Google Scholar 

  • Hegde, S. P., & Mishra, D. R. (2019). Married CEOs and corporate social responsibility. Journal of Corporate Finance, 58, 226–246.

    Article  Google Scholar 

  • Hong, B., Li, Z., & Minor, D. (2016). Corporate governance and executive compensation for corporate social responsibility. Journal of Business Ethics, 136(1), 199–213.

    Article  Google Scholar 

  • Huang, S. K. (2013). The impact of CEO characteristics on corporate sustainable development. Corporate Social Responsibility and Environmental Management, 20(4), 234–244.

    Article  ADS  Google Scholar 

  • Ikram, A., Li, Z. F., & Minor, D. (2019). CSR-contingent executive compensation contracts. Journal of Banking and Finance, 105655.

  • Jones, S. (2017). Corporate bankruptcy prediction: A high dimensional analysis. Review of Accounting Studies, 22(3), 1366–1422.

    Article  Google Scholar 

  • Krüger, P. (2015). Corporate goodness and shareholder wealth. Journal of Financial Economics, 115(2), 304–329.

    Article  Google Scholar 

  • Lins, K. V., Servaes, H., & Tamayo, A. (2017). Social capital, trust, and firm performance: The value of corporate social responsibility during the financial crisis. The Journal of Finance, 72(4), 1785–1824.

    Article  Google Scholar 

  • Lundberg, S. M., & Lee, S. I. (2017, December). A unified approach to interpreting model predictions. In Proceedings of the 31st international conference on neural information processing systems (pp. 4768–4777).

  • Manner, M. H. (2010). The impact of CEO characteristics on corporate social performance. Journal of Business Ethics, 93(1), 53–72.

    Article  ADS  Google Scholar 

  • Masulis, R. W., & Reza, S. W. (2015). Agency problems of corporate philanthropy. The Review of Financial Studies, 28(2), 592–636.

    Article  Google Scholar 

  • McCarthy, S., Oliver, B., & Song, S. (2017). Corporate social responsibility and CEO confidence. Journal of Banking & Finance, 75, 280–291.

    Article  Google Scholar 

  • Nofsinger, J. R., Sulaeman, J., & Varma, A. (2019). Institutional investors and corporate social responsibility. Journal of Corporate Finance, 58, 700–725.

    Article  Google Scholar 

  • Pastor, L., Stambaugh, R., & Taylor, L. A. (2021). Sustainable investing in equilibrium. Journal of Financial Economics, 142(2), 550–571.

    Article  Google Scholar 

  • Pastor, L., Stambaugh, R., & Taylor, L. A. (2022). Dissecting green returns. Journal of Financial Economics, 146(2), 403–424.

    Article  Google Scholar 

  • Petrenko, O. V., Aime, F., Ridge, J., & Hill, A. (2016). Corporate social responsibility or CEO narcissism? CSR motivations and organizational performance. Strategic Management Journal, 37(2), 262–279.

    Article  Google Scholar 

  • Radu, C., & Smaili, N. (2021). Alignment Versus Monitoring: An Examination of the Effect of the CSR Committee and CSR-Linked Executive Compensation on CSR Performance. Journal of Business Ethics, pp 1–19.

  • Refinitiv (2021) Environmental, social and governance (ESG) Scores from Refinitiv. RE1291648/3–21.

  • Shaukat, A., Qiu, Y., & Trojanowski, G. (2016). Board attributes, corporate social responsibility strategy, and corporate environmental and social performance. Journal of Business Ethics, 135(3), 569–585

  • Udayasankar, K. (2008). Corporate social responsibility and firm size. Journal of Business Ethics, 83(2), 167–175.

    Article  Google Scholar 

  • Waldman, D. A., & Siegel, D. (2008). Defining the socially responsible leader. The Leadership Quarterly, 19(1), 117–131.

    Article  Google Scholar 

  • Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT press.

  • Yuan, Y., Tian, G., Lu, L. Y., & Yu, Y. (2019). CEO ability and corporate social responsibility. Journal of Business Ethics, 157(2), 391–411.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Fordham University, New York, USA

    Alona Bilokha, Mingying Cheng, Mengchuan Fu & Iftekhar Hasan

  2. Bank of Finland, Helsinki, Finland

    Iftekhar Hasan

  3. University of Sydney, Sydney, Australia

    Iftekhar Hasan

Authors
  1. Alona Bilokha
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mingying Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mengchuan Fu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Iftekhar Hasan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Iftekhar Hasan.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 391 kb)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bilokha, A., Cheng, M., Fu, M. et al. Understanding CSR champions: a machine learning approach. Ann Oper Res (2024). https://doi.org/10.1007/s10479-024-05839-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10479-024-05839-3

Keywords

JEL Classification

Search

Navigation