A New Understanding of Marketing and “Doing Good”: Marketing’s Power in the TMT and Corporate Social Responsibility

Abstract

The traditional understanding of corporate social responsibility (CSR) has largely been focused on its downstream performance implications, particularly its associations with firms’ customer market metrics such as customer loyalty, customer satisfaction and customer co-creation as well as financial ones such as firm value, return on assets etc. However, given the close relationship between CSR and marketing that literature has identified, it is surprising that the relationship between a focal upstream construct, i.e. the marketing function’s power within a firm and the firm’s propensity toward CSR has not been addressed in the literature. Examining the link between marketing’s power (MP) in a firm’s top management team (TMT) and firm CSR levels, we investigate how this fundamental TMT configuration, i.e. the distribution of marketing power in the TMT, motivates the firm’s social endeavors. Further, we formulate this relationship in a contingency-based model that incorporates the moderating effects of firm size, firm age, service intensity, and resource slack across 1569 firms operating in 63 industries. In addition to their effect on CSR, this study shows how MP in TMT may influence corporate social irresponsibility (CSI) as well as CSR capability after controlling for industry type. The inclusion on these additional dimensions of CSR (CSI and CSR capability) complements our analyses of the effect that MP has on CSR. The research contributes to a deeper understanding of CSR’s fundamental corporate determinants as well as identifies the essential role of the marketing function in firms’ CSR strategy. In this process, it yields useful implications for multiple streams of theory as well as for business practices.

Appendices

Appendix 1

See Table

Table 5 The sample distribution across industries

5.

Appendix 2

We first adopt the robust regression with White standard error with a firm cluster adjustment using:

$$\hat{V}_{cl} \left( {\hat{\beta }} \right) = (X^{\prime}X)^{ - 1} \left( {\mathop \sum \limits_{j = 1}^{j} X_{j}^{^{\prime}} \hat{\varepsilon }_{j} \hat{\varepsilon }_{j}^{^{\prime}} X_{j} } \right)(X^{\prime}X)^{ - 1}$$

where $\hat{\varepsilon }_{j}$ denotes the vector of residuals for the jth cluster from OLS estimation. This method is renowned for dealing with heteroscedasticity and autocorrelation when firm panel data are involved (Stock and Watson 2008). To ensure the robustness, we further use the Driscoll–Kraay method and Newey–West method, which also generate the heteroscedasticity and autocorrelation consistent (HAC) standard errors and are in popular use in all business research fields (e.g., Chapple and Humphrey 2014; Vogelsang 2012).

Appendix 3

$$CSI_{it + 1} = \beta_{0} + \beta_{1} \times MP_{it} + \beta_{2} \times MP_{it} \times FSize_{it} + \beta_{3} \times MP_{it} \times FAge_{it} + \beta_{4} \times MP_{it} \times SerIn_{it} + \beta_{5} \times MP_{it} \times Slack_{it} + \beta_{6} \times FSize_{it} + \beta_{7} \times FAge_{it} + \beta_{8} \times SerIn_{it} + \beta_{9} \times Slack_{it} + Control\,Variable\,Set + Time\,Dummies_{t} + Industry\,Dummies_{j} + \varepsilon_{it} .$$

Appendix 4

We use Tobin’s q as the performance measure because the q is a preferred indicator of shareholder value and is the fundamental reflection of firms’ financial goals (Singh et al. 2018). We use CSR and CSR momentum as reflected by the growth rate of CSR as the input factors. Then we use SFM to gauge how well firms can use CSR and CSR momentum to realize the performance. The model is formulated as

$$ln\left( {FPer_{it} } \right) = \alpha_{0} + \alpha_{1} \times ln\left( {CSR_{it} } \right) + \alpha_{2} \times ln\left( {CSRMomentum_{it} } \right) + \varepsilon_{it} - \eta_{it}$$

$FPer_{it}$ is the residual of a regression that partials out other factors’ influences from Tobin’ q. These factors include firm traits such as size, age, diversification, adveristing, and R&D, as well as external environmental influences such as munificence, turbulence, and competition:

$$Tobin\;q_{it} = f\left( {firm\;factors_{it} + environmental\;factors_{jt} } \right)$$

This ensures that CSR’s effects on performance in the frontier model may be gauged in purity. The reversed $\eta_{it}$ measures the capability of CSR. We then run the robust regressions with CSR capability (CSRCap) as the dependent variable and MP as the independent variable.

$$CSRCap_{{it{ + 1}}} = \beta_{0} + \beta_{1} \times {\text{MP}}_{it} + \beta_{2} \times {\text{MP}}_{it} \times FSize_{it} + \beta_{3} \times {\text{MP}}_{it} \times FAge_{it} + \beta_{4} \times {\text{MP}}_{it} \times SerIn_{it} + \beta_{5} \times {\text{MP}}_{it} \times Slack_{it} + \beta_{6} \times FSize_{it} + \beta_{7} \times FAge_{it} + \beta_{8} \times SerIn_{it} + \beta_{9} \times Slack_{it} + Control\,Variable\,Set + Time\,Dummies_{t} + Industry\,Dummies_{j} + \varepsilon_{it} .$$