Not Yet Abandoning the Single Measure for Corporate Social Performance : A Weighted CSP Measure Via DEA

While corporate social performance (CSP) plays an increasingly important role in supply chain and operations management, measuring CSP has been an important yet challenging task facing academics and managers. The challenge mainly comes from the fact that stakeholder expectations (i.e., the relative importance of different CSP dimensions) are usually unknown at the firm level but are more often known at the industry or market level. This study develops a new approach to incorporate such aggregate-level information to create a more robust and reliable single CSP measure. Focusing on the popular KLD data, this paper develops a DEA-based approach to reflect the stakeholders’ priority for different CSR issues faced by individual firms. Distinct from the traditional DEA, the new approach can incorporate constraints on average weights for different CSR issues – which are important in CSP evaluation. Empirically, I tested the new and the traditional CSP approach widely used in the literature by using data from socially responsible funds (SRF). The new approach outperforms the standard approach in the empirical test. The results show that the standard CSP scores are inversely correlated with the number and the shares of socially responsible funds invested in the firm. The new approach endogenizes the unknown CSP weights at the firm level and does not require a priori weight specification. While the CSP weights for each firm are endogenously determined in the model, the new approach allows researchers to impose a weight structure at the macro level when contextual knowledge about the stakeholder environment is available