THE MODERATING EFFECT OF INNOVATION ON
THE RELATION BETWEEN CORPORATE SOCIAL
PERFORMANCE AND FIRM PERFORMANCE - AN
INDUSTRY APPROACH
Reinier Smits S2160390
University of Groningen Faculty of Economics and Business
MSc. Finance
Supervisor: B. van Oostveen January 2018
ABSTRACT
This study continues on prior research by addressing the moderating effect of innovation on the relationship between corporate social performance and firm performance and compares the influence of innovation on this relationship between the service and manufacturing sector. The first part of this study examines the relationship between corporate social performance and firm performance using a large dataset with listed companies from the United States. The second part of this study uses this dataset to test the moderating role of innovation in the two sectors and compares this moderating effect of innovation. Results show that there is not a significant effect of firm performance on corporate social performance in both sectors, neither vice versa. Furthermore, there is not a significant direct moderating effect of innovation in the manufacturing industry. This relationship is neither found in the service industry.
1
1. INTRODUCTION
Theoretical evidence states that the relationship between corporate social performance and firm performance can be positive, negative or neutral. The sign of this relationship concludes from different theories. The social impact hypothesis states that meeting the needs of more stakeholders than only the shareholders will result in an overall positive performance including firm performance. On the other hand, the trade-off hypothesis states that using resources to fulfill the needs of stakeholders other than shareholders will harm firm performance. Furthermore, literature provides evidence for hypotheses derived from both theories, as shown by the abundance of mixed results on this topic. This study will clarify the uncertainty about this relationship and the role of innovation on this relationship. By using a large dataset and both fixed and random effects models, this study addresses the moderating effect of innovation on the relationship between corporate social performance and firm performance in two different industries.
The central question in the field of corporate social performance is if a corporation can create wealth and at the same time not harm society. This question of whether a corporate can do good and do well at the same time is of interest in research over the past century (Wells, 2002). In other words, the question is if different kind of performances of companies are affected by their strategies and operations in market and non-market environments. What makes the organizations’ non-market strategies even more important is an increasing power of activist groups and media in western societies. One construct that might capture a major element of these non-market strategies is corporate social performance (Orlitzky et al, 2003). Corporate social performance can be defined as ‘a business organization’s configuration of principles of social responsibility, processes of social responsiveness, and policies, programs, and observable outcomes as they relate to the firm’s societal relationships’ (Wood, 1991).
2 Wang (2009), Peters and Mullen (2009) and Ruf, Muralidhar, Brown, Janney and Paul (2001) conclude that the relationship between corporate social performance and firm performance is positive. However, studies from for example Becchetti, Di Giacomo and Pinnacchio (2008), Shrosphire and Hillman (2007), Surroca and Tribo (2008) and Moore (2001) show a negative relationship between corporate social performance and firm performance. Furthermore, other academic papers from Richard, Ford and Ismail (2006) and Galbreath (2006) even find no positive nor a negative relationship.
3
2. LITERATURE
There are two different theories describing the responsibilities of a company. Firstly, the shareholder theory proposed by Milton Friedman (1970) claims that the solely purpose of a business entity is to increase profits for its shareholders. Secondly, the stakeholder theory introduced by Edward Freeman stated that a company owes a responsibility to a wider group of stakeholders, other than just shareholders. These two different theories are the fundament of the research conducted in the past on the relationship between corporate social performance and firm performance. Even though these theories are opposed, past research has shown mixed results, positive, neutral, and negative on the relationship between corporate social performance and firm performance. Based on the stakeholder theory, four different theories are presented which give a theoretical explanation for the sign of the corporate social performance - firm performance relationship and the causality of this relationship. These theories are combined with empirical evidence which results in the tested hypotheses. Ultimately, this relationship will be discussed in the context of innovation and different industries.
2.1 Theoretical Framework
In addition to the inevitable problems of measurement, the relationship between corporate social performance and firm performance involves two different empirical issues. One of these issues is the sign of the relationship: Are social and financial performance positively, negatively or not associated at all? The other issue is the causal relationship involved: Does social performance influence firm performance or does firm performance influence social performance (Preston and O’Bannon, 1997). Combining these two dimensions of variation yields four possible causal and different sign hypotheses discussed in the following paragraphs. The presented hypotheses are supported by the presented empirical evidence in the following section.
4 the expectations of various non shareowner constituencies will generate market fears, which, in turn, will increase a company’s risk premium and ultimately result in higher costs and/or lost profit opportunities. According to their study, when a corporation tries to serve the implicit claims of different stakeholders this will ultimately lead to a more positive company reputation in a way that has positive impact on its financial performance. On the other hand, when a company fails to meet the implicit claims of different stakeholders and thereby disappointing these stakeholder groups it can result in a negative reputation and may have a negative financial impact. According to Preston and O’Brannon (1997) the ‘social impact’ version of the stakeholder theory implies a lead-lag relationship between social and firm performance; external reputation develops first, then financial results follow. Secondly, the trade-off hypothesis claims the opposite of the social impact hypothesis. The trade-off hypothesis claims that corporate social performance is the independent variable in the relationship between corporate social performance and firm performance. Furthermore, it states that social accomplishments involve financial costs. This hypothesis reflects the classic Friedman position and is supported by the well-known early finding of Vance (1975) that corporations displaying strong social credentials experience declining stock prices relative to the market average. More recently, the trade-off hypothesis has been carefully articulated by Aupperle et al. (1985). They point out that socially responsive activities may siphon off capital and other resources from the firm, putting it at a relative disadvantage compared to firms that are less socially active. Hence, a firm’s higher level of social performance may lower its financial performance as compared to competitors and/or other norms.
5 leading corporate social performance. This relationship would provide a relevant test of this proposition. McGuire et al. (1988) found a stronger positive relationship when firm performance was viewed as the leading variable. Kraft and Hage (1990) found that the availability of slack resources, as well as the values and goals of managers, strongly influenced the level of community service undertaken by corporations. Secondly, the managerial opportunism hypothesis states that firm performance affects corporate social performance negatively. In the existing literature it is often argued that corporate managers may pursue private objectives. This is often a disadvantage to both share- and stakeholders. It is argued that managers consider their own interest of primary importance in corporate decision making. The managerial opportunism hypothesis states that there is a negative relationship between firm performance and corporate social performance. The hypothesis concluded that the pursuit of private managerial targets, in the context of compensation schemes closely linked to short-term profit and stock price behavior, affects the relationship between firm performance and corporate social performance negatively. There is a simple reasoning behind this conclusion. In a situation where the firm performance is strong, managers may attempt to cash in by reducing social expenditures in order to take advantage of the opportunity to increase their own short-term private gains. Conversely, when financial performance weakens, managers may attempt to offset, and perhaps appear to justify, their disappointing results by engaging in conscientious social programs (Preston and O’Brannon, 1997).
2.2 Empirical evidence
The first section will present empirical evidence related to the relationship between corporate social performance and firm performance and present the supportive hypotheses. Furthermore, the second part of this section will discuss this relationship in association with innovation and present empirical evidence on the differences between innovation in the service and manufacturing industry. This section is concluded with two other hypotheses. The pressure firms endure to maximize their corporate social performance as well as their firm performance is increasing (Grow, Hamm, and Lee, 2005). The large number of empirical studies that examine the relationship between corporate social performance and firm performance bring forth mixed results (Orlitzky et al, 2003).
6 hypothesis. Corporations trying to improve their corporate social performance draw funds and management effort away from the core business. This will result in an overall lower profit. In this theory, it is impossible for managers to perform on both corporate social operations and firm operations at the rightful time. Surroca and Tribo (2008) state that managers possible collude with non-shareholder stakeholders in order to reinforce their entrenchment strategy. This will have a negative impact on firm performance. Moore (2001) suggests that positive firm performance will lead to good corporate social performance, which then distracts firms for the core business. This will consequently lead to poor firm performance. Becchetti, Di Giacomo and Pinnacchio (2008) suggests that focus on corporate social performance is a shift from shareholder wealth to stakeholder wealth. They find a negative relationship between corporate social performance and firm performance Furthermore, there is empirical evidence that the relationship between corporate social performance and firm performance is a not significant relationship (McWilliams and Siegel, 2000). Although there is no clear consent on the relationship between corporate social performance and firm performance, most of the research in this field of research does support a positive relationship. For instance, a study by Peter and Mullen (2009) argues that there is a positive relationship between corporate social performance and firm performance. They study the effect of long-term cumulative corporate social performance on future firm performance. They stated that long-term positive corporate social performance is positive for both stakeholders as shareholders. Janney and Paul (2001) found a positive short-term relationship between corporate social performance and firm performance. The study finds support for the tenet in stakeholder theory that shareholders benefit when the needs of multiple stakeholders are met. A positive change in corporate social performance is associated with growth in sales for the current and subsequent year. Thus, taking into account the different theoretical position on the relationship between corporate social performance and the empirical evidence, we expect for both causal relationships a positive relationship:
H1: Corporate social performance positively affects firm performance H2: Firm performance positively affects corporate social performance
7 old products or processes when new innovations enter a market are likely to fail. On the other hand, firms that offer new innovations into the market are likely to grow as the older products or processes are replaced. The study of managing innovations has been of interest of academic research for decades (Burns and Stalker, 1961) and these studies have shown that innovation has a positive effect on firm performance when corporate social factors are taken into account (Chandler, Keller and Lyon, 2000). It is suggested that any study that examines the drivers of firm performance or corporate social performance should definitely include innovation to avoid missing its effects (McWilliams and Siegel, 2000). In this study, taking into account the empirical evidence on the effect of innovation on firm performance and corporate social performance we expect a positive relationship between innovation and firm performance and corporate social performance.
8 performance together with the relationship between corporate social performance and firm performance and the differences of innovation in the manufacturing and service sector we will test the following two hypotheses:
H3: The moderating effect of innovation on the corporate social performance and firm performance relationship is positive for the manufacturing industry
H4: The moderating effect of innovation on the corporate social performance and firm performance relationship is positive for the service industry
3. DATA AND DESCRIPTIVES STATISTICS
The study presented in this paper distinguishes itself from existing empirical literature on the relationship between corporate social performance and firm performance because it includes and empirically tests the influence of innovation on this relationship. Furthermore, this study acknowledges the differences of innovation in different sectors and test these differences. Additionally, the data set used can be considered an improvement when compared to previous research as it captures a longer time frame of eight years (2008-2016) and only focuses on the United States where other studies focus on other markets or combined markets. This dataset enables us to test the unique hypotheses explained in the previous section. According to Campbell and Minguez-Vera (2008), panel data is preferred over cross-sectional data because it leads to a more reliable analysis and it migrates some of the biases regarding unobservable heterogeneity and omitted variables.
9 from the year 2008 till 2016. The data available in Thomson Reuters Datastream for the chosen variables were all presented yearly. Therefore, the data used in this study is yearly. Since the data we will use will both contain time series and cross-sectional elements it is a panel dataset. In the following section we will explain the variables used in the model to test the previous stated hypotheses. The first variables that we will elaborate on are the dependent variables. As a dependent variable we will use firm performance in the first model and corporate social performance in the second model. In previous literature examples of used measurements of corporate social performance are the MSCI KLD index for U.S. securities and corporate social performance disclosure measurements (Brown and Perry, 1994; Graves and Waddock, 1994; Griffin and Mahon, 1997; Wiseman, 1982; Wolfe, 1991).The measurement we will use for corporate social performance is the numerical score provided for the different pillars, corporate governance, social and environmental, in the Asset4 Database. We have chosen for this measurement because of the large availability of the data for the used companies. We will use two control variables to control for the size of the firm. These proxies for firm size are number of employees and net revenues or sales. Normally the model should be controlled for sector effects but since we will focus explicitly on the differences between the two sectors this control variable is not needed.
3.1 Variable Operationalization
10 profitability of the firm. It is a widely used measure for firm performance and is of great interest for shareholders which makes it an interesting measurement (Ruf, et al., 2001). On the other hand, volatile market-based measures seem not to be appropriate for this study due to the large time horizon of eight years.
Corporate social performance - The corporate social performance of a firm can be measured in several ways (Wood, D.J., 2010). The Thomson Reuters Datastream data measures corporate social performance in four different pillars. The pillars are corporate governance, economic, environmental and social. The corporate governance score is described by the Thomson Reuters Datastream as follows: “The corporate governance pillar measures a company’s systems and processes, which ensure that its board members and executives act in the best interest of its long term shareholders. It reflects a company’s capacity, through its use of best management practices, to direct and control its rights and responsibilities through the creation of incentives, as well as checks and balances in order to generate long term shareholder value.” The economic score is described as: “The economic pillar measures a company’s capacity to generate sustainable growth and a high return on investment through the efficient use of all its resources. It is the reflection of a company’s overall financial health and its ability to generate long term shareholder value through its use of best management practices.” The environmental score is described as follows: “The environmental pillar measures a company’s impact on living and nonliving natural systems, including the air, land and water, as well as complete ecosystems. It reflects how well a company uses best management practices to avoid environmental risks and capitalize on environmental opportunities in order to generate long term shareholder value.” The social score is described as follows: “The social pillar measures a company’s capacity to generate trust and loyalty with its workforce, customers and society, through its use of best management practices. It is a reflection of the company’s reputation and the health of its licence to operate, which are key factors in determining its ability to generate long term shareholder value.” The economic pillar is excluded from the measurement for corporate social performance in the study because the economic pillar measures the same as firm performance. We created an index of corporate social performance by calculating the index as the sum of the remaining three scores. By creating this index all three scores have equal weights.
11 social performance and firm performance innovation is included in the interaction term. In the first two models it is modelled together with firm performance and in the last two models it is modelled together with corporate social performance. As a proxy measure for innovation we will use the general research and development expenditures (Hull and Rothenberg, 2008). General research and development expenditures are often used as a proxy for innovation in comparable research (Donnelly, 2000). The individual data for the used companies of general research and development expenditures were provided by the Thomson Reuters Datastream database. This database describes this variables as: “Research and development expenses represent all direct and indirect costs related to the creation and development of new processes, techniques, applications and products with commercial possibilities.”
12 3.2 Descriptive Statistics
For all variables used in this study, the descriptive statistic are presented below in table 1 for the manufacturing industry and in table 2 for the service industry.
Table 1
Descriptive statistics of variables manufacturing industry
In this table all the variables, corporate social performance score, the return on assets, the return on equity, the general research and development expenditures, the natural logarithm of number of employees, the natural logarithm of sales, used in the models for the manufacturing industry are presented. The following properties of the variables are presented: number of observations, the mean, standard deviation, minimum and maximum.
Variable Obs Mean Std. Dev. Min Max
Corporate social performance score
2617 57.88 91.55 0.00 285.51
Return on assets 2617 4.71 11.24 -92.46 55.01
Return on equity 2617 8.77 22.57 -98.22 98.50
General research and development expenditures
1753 136295.9 626399.6 0.00 8124000
Number of employees (ln) 2271 8.25 1.81 3.26 12.32
Sales (ln) 2526 13.70 2.14 0 18.93
Table 2
Descriptive statistics of variables service industry
In this table all the variables, the corporate social performance, the return on assets, the return on equity, the general research and development expenditures, the natural logarithm of number of employees, the natural logarithm of sales, used in the models for
the service industry are presented. The following properties of the variables are presented: number of observations, the mean,
standard deviation, minimum and maximum.
Variable Obs Mean Std. Dev. Min Max
Corporate social performance score
2369 32.22 66.27 0 285.61
Return on assets 2369 1.75 15.45 -86.59 73.61
Return on equity 2369 2.18 28.81 -99.97 97.65
General research and development expenditures
1281 150624.6 703498.5 0 15800000
Number of employees (ln) 2023 6.87 1.83 3.22 12.47
Sales (ln) 2369 12.19 2.41 0.00 18.29
13
Table 3
Pair wise correlation matrix manufacturing industry
Corporate social performance Return on assets Return on equity
General research and development expenditures Number of employees (ln) Sales (ln) Corporate social performance - Return on assets 0.1059 - Return on equity 0.2040 0.7472 -
General research and development expenditures 0.2611 0.0412 0.0973 - Number of employees (ln) 0.6141 0.1111 0.1939 0.3185 - Sales (ln) 0.5748 0.2820 0.3177 0.3387 0.9075 -
In the correlation matrix of the variables used for the manufacturing industry all pair wise correlation are relatively low. As can be seen in the table the correlation between both the control variables, the natural logarithm of the number of employees and the natural logarithm of sales, is 0.9075. This is not surprising since it is logical that sales grows together with the number of employees. Therefore there is no reason to adjust the model for both the variables. In addition, the correlation between return on equity and return on assets is 0.7472. Both variables measure profitability of the firm so the high correlation is no surprise. Since these variables are not used together in the model there is no need to make adjustments to the variables.
14
Table 4
Pair wise correlation matrix service industry
Corporate social performance Return on assets Return on equity
General research and development expenditures Number of employees (ln) Sales (ln) Corporate social performance - Return on assets 0.1449 - Return on equity 0.1646 0.8747 -
General research and development expenditures 0.1390 0.0821 0.0831 - Number of employees (ln) 0.4212 0.0879 0.0865 0.1572 - Sales (ln) 0.3786 0.0674 0.0936 0.8200 0.4777 - 4. METHODOLOGY
The aim of this study is to analyse both directional relationships between corporate social performance and firm performance and the moderating effect of innovation on both relationships. More precisely, we investigate whether general research and development expenditures, being a proxy for innovation, increase the effect of firm performance on corporate social performance and vice versa. By doing this study for both service and manufacturing industry, we can compare the influence of innovation on this relationship between these industries. The two equation that will be estimated for both the industries are:
α α innovation α fp α innovation p α employ α sales c v ε
CSPit = 0 + 1 it + 2 it+ 3 * f + 4 + 5 + i + t + it (1) (2) α α innovation α csp α innovation sp α employ α sales c v ε
F Pit = 0 + 1 it + 2 it + 3 * c + 4 + 5 + i + t + it
15 corporate social performance and the control variables employ and sales have the coefficient vectors α4and α3. The estimation method that is used is an ordinary least squares regression. The term c i is a firm-fixed effect, v t is a time-fixed effect and εit is the error term. For firm performance two different measurements are used to estimate the above-mentioned equations. The proxies for firm performance are the following: return on assets and return on equity. Both equation (1) and (2) are combined with these two different measurements of firm performance. This yield four different equations for both service and manufacturing industry. Models 1a and 1b are the estimations of equation 1 for the manufacturing industry with use of return on assets and return on equity respectively as proxies for firm performance. Models 2a and 2b are the estimations of equation 2 for the manufacturing industry. This is equal for the service industry with respectively models 3a and 3b for equation 1 and models 4a and 4b for equation 2.
16
5. RESULTS
This research consists of two parts examining a different relationship. The first part presents the results of the ordinary least squares regression for the models where corporate social performance is regressed on firm performance and the interaction term plus the control variables. In the second part the two types of different firm performances are regressed on corporate social performance and the interaction term plus the control variables. These different relationships are all performed for both the service and manufacturing industry. The results are presented in pairs of the same model for the service and manufacturing industry.
Table 5
Corporate social performance and the impact of firm performance and innovation
This table shows the results of the fixed effects regression estimations of both model (1a) and model (3a). The first column
presents the results of this estimation for the manufacturing industry. The second column presents the results of this exact same
estimation for the service industry. For both model (1a) and (3a) financial data is collected from Thomson Reuters Datastream
and the data on corporate social performance is collected from the Asset4 Database. All the data is from the United States
covering the period between 2007 - 2016. Column 1 shows the results from corporate social performance regressed on general
research and development expenditures, return on assets and the interaction term of general research and development expenditures and return on assets where the number of employees (ln) and sales (ln) is controlled for. Column 2 shows the same regression estimation for the service industry. All variables are measured in the last quarter of the year. ***, ** and * represent
significance at the 1%, 5% and 10% levels respectively. The numbers between brackets are the t-values for estimations of the
coefficients.
Model (1a): Manufacturing Industry Model (3a): Service Industry
Corporate social performance Corporate Social performance General research and development
expenditures 0.000 (0.01) -0.000*** (-2.05) Return on assets 0.002 (0.01) -0.027 (-0.56) Return on assets * General research
and development expenditures
0.000 (1.32) -0.000 (-0.06) Number of employees (ln) 5.928 (1.05) 2.934 (0.80) Sales (ln) 17.557*** (3.19) 18.095*** (3.82) Observations 1570 1154 Number of groups 243 212 Overall r2 0.3878 0.3577
17 logarithm of sales are both significance at a 1% significance level. Since the coefficients of the interaction terms for both industries are not significant no significant moderating effect of innovation on the relationship between return on assets and corporate social performance is observed.
In table 6 the results of the fixed effects ordinary least squares estimations for both model (1b) and model (3b) are presented.
Table 6
Corporate social performance and the impact of firm performance and innovation
This table shows the results of the fixed effects OLS estimations for both model (1b) and model (3b). The first column presents
the results of this estimation for the manufacturing industry. The second column presents the results of this exact same estimation
for the service industry. For both model (1b) and (3b) financial data is collected from Thomson Reuters Datastream and the data
on corporate social performance is collected from the Asset4 Database. All the data is from the United States covering the period
between 2007 - 2016. Column 1 shows the results from corporate social performance regressed on general R&D expenditures,
return on equity and the interaction term of general R&D expenditures and return on equity where the number of employees (ln)
and sales (ln) is controlled for. Column 2 shows the same regression estimation for the service industry. All variables are
measured in the last quarter of the year. ***, ** and * represent significance at the 1%, 5% and 10% levels respectively. The numbers between brackets are the t-values for estimations of the coefficients.
Model (1b): Manufacturing Industry Model (3b): Service Industry
Corporate social performance Corporate Social performance General R&D expenditures 0.000**
(2.00) -0.000 (-1.03) Return on equity 0.089 (1.35) -0.006 (-0.15) Return on equity * General R&D
expenditures 0.000 (0.83) 0.677 (-0.42) Number of employees (ln) 6.240 (1.10) 0.000 (0.14) Sales (ln) 16.666*** (3.02) 0.000 (1.10) Observations 1570 1154 Number of groups 243 212 Overall r2 0.382 0,080
18 In table 7 the results of the fixed effects OLS estimation models for model (2a) and model (4a) are presented. This model regresses the return on equity on the previous explained variables. In model (2a) the coefficients for the natural logarithm of sales and the number of employees are significant at a 1% and 10% significance level respectively. Furthermore, the coefficient for the general R&D expenditures is significant at a 10% significance level. As can be seen in the table the coefficients of the independent variables and the control variables in model (4a) are not significant at a 1%, 5% or 10% significance level.
Table 7
Firm performance and the impact of corporate social performance and innovation
This table shows the results of the fixed effects OLS estimations of model (2a) and model (4a). The first column presents the
results of this estimation for the manufacturing industry. The second column presents the results of this exact same estimation for
the service industry. For both model (2a) and (4a) financial data is collected from Thomson Reuters Datastream and the data on
corporate social performance is collected from the Asset4 Database. All the data is from the United States covering the period
between 2007 - 2016. Column 1 shows the results from the return on assets regressed on general R&D expenditures , corporate
social performance and the interaction term of general R&D expenditures and corporate social performance where the number of
employees (ln) and sales (ln) is controlled for. Column 2 shows the same regression estimation for the service industry. All
variables are measured in the last quarter of the year. ***, ** and * represent significance at the 1%, 5% and 10% levels respectively. The numbers between brackets are the t-values for estimations of the coefficients.
Model (2a): Manufacturing Industry Model (4a): Service Industry
Return on assets Return on assets
General R&D expenditures -0.000* (-1.78)
-0.000 (-1.90) Corporate social performance 0.000
(0.05)
0.000 (0.02) Corporate social performance * General
R&D expenditures 0.000 (1.63) -0.000 (-0.09) Number of employees (ln) -2.339* (1.89) -0.000 (-1.84) Sales (ln) 5.376*** (3.37) -0.000 (1.44) Observations 1570 1154 Number of groups 243 212 Overall r2 0,037 0,004
19 variables are both significant. The natural logarithm of sales and the natural logarithm of number of employees are significant at a 1% and 10% significance level respectively. Furthermore, general R&D expenditures is the only independent variable with a significant coefficient. In model (4b) for the service industry the same variables are significant but at other significance levels as can be seen in the table below.
Table 8
Firm performance and the impact of corporate social performance and innovation
This table shows the results of the fixed effects OLS estimations for model (2b) and model (4b). The first column presents the
results of this estimation for the manufacturing industry. The second column presents the results of this exact same estimation for
the service industry. For both model (2b) and (4b) financial data is collected from Thomson Reuters Datastream and the data on
corporate social performance is collected from the Asset4 Database. All the data is from the United States covering the period
between 2007 - 2016. Column 1 shows the results from the natural logarithm of return on equity regressed on general R&D
expenditures (ln), corporate social performance and the interaction term of general R&D expenditures (ln) and corporate social
performance where the number of employees (ln) and sales (ln) is controlled for. Column 2 shows the same regression estimation
except that the variables are regressed on return on equity, so not on the natural logarithm of the variable. Furthermore, the
corporate social performance is incorporated in the model as a natural logarithm. All variables are measured in the last quarter of the year. ***, ** and * represent significance at the 1%, 5% and 10% levels respectively. The numbers between brackets are the t-values for estimations of the coefficients.
Model (2b): Manufacturing Industry Model (4b): Service Industry
Return on equity Return on equity
General R&D expenditures -0.000* (-1.73)
-0.000** (-1.99) Corporate social performance 0.015
(1.21)
-0.008 (-0.33) Corporate social performance * General
R&D expenditures 0.000 (1.43) 0.000 (0.34) Number of employees (ln) -5.431* (-1.96) -0.000*** (-2.69) Sales (ln) 11.503*** (3.76) 0.000* (1.77) Observations 1570 1154 Number of groups 243 212 Overall r2 0,070 0,005 6. CONCLUSION
20 social performance does not significantly influence firm performance. A possible explanation for this result is the ban of a lag between corporate social performance and firm performance. As explained in the literature section some argue for such a lead-lag because there is no direct result of corporate social performance on firm performance. A favourable corporate social performance needs time to yield a positive firm performance. We can reject hypotheses 1 and can conclude that there is no direct causal relationship of corporate social performance on firm performance. Furthermore, hypothesis 2 can be rejected for the two measurements of firm performance. There is no direct causal relationship found between firm performance and corporate social performance.
Furthermore, the research shows that innovation in the manufacturing industry does not significantly affects the influence of firm performance on corporate social performance. In other words, there is no evidence found that the higher the general research and development expenditures the more positive the effect of firm performance on corporate social performance is. This part of the research shows that in the service industry the effect of innovation on the relationship between corporate social performance and firm performance can be both positive and negative depending on the measurement of firm performance but none of the estimations is significant. Therefore we can conclude that for the manufacturing industry innovation does not significantly amplified positively the effect of firm performance on corporate social performance. In the service industry this effect can neither be seen. However, the theory that explains this relationship is difficult to determine. As explained in the literature section, different researches explain in detail the differences between innovation in the manufacturing industry and the service industry. However the reason why the effect of innovation on the relationship between corporate social performance and firm performance is different is not elaborated on. One might think that innovation in the manufacturing industry effects the relationship with a smaller lead-lag due to the so-called ‘hard’ innovations. In the service industry it can be expected that the so-called ‘soft’ innovations or process innovations has a positive effect on the relationship between corporate social performance and firm performance on a longer term. This study focuses on the direct effect of general research and development expenditures on corporate social performance and firm performance and therefore this longer term effect is ignored.
21 significantly. In other words, the general research and development expenditures do not amplify the effect of corporate social performance on firm performance either positive or negative. This is the result in the two different models in both the manufacturing as the service industry. Therefore we can conclude that the measurement of firm performance does not affect the influence of innovation on the relationship of corporate social performance and firm performance. Furthermore, not only the interaction term is not significant in this model also the corporate social performance variable solely is not significant. In the first section of the research the independent variables of firm performance were significant. Therefore, there is reason to believe that in this research, with this specific dataset, corporate social performance has no significant effect on firm performance. However, as explained in previous sections there is theoretical and empirical evidence that claims that this specific causal relationship does exist. A possible explanation for the results is that we ignored lags in this study. The causal relationship explained above needs probably a lag to measure the effect of innovation on the extent of effect of corporate social performance on firm performance. The investments in corporate social performance are expected not directly to be seen in a change of firm performance. These investments are a longer term investment and therefore the effect of innovation and corporate social performance is not seen directly in firm performance.
Taking into account both the first and second part of this study, we can conclude that the third hypothesis can not be rejected for the specific causal relationship where corporate social performance is the dependent variable and firm performance the independent variable with innovation as the moderator. The reverse causal relationship is not significant. The fourth hypothesis can neither be rejected for the causal relationship where corporate social performance is the dependent variable and firm performance is the independent variable with innovation as the moderator. The reverse causal relationship is not significant
23
REFERENCES
Acs, Zoltan J., and David B. Audretsch, 2003, Innovation and Technological Change, Handbook of Entrepreneurship Research 1, 55-79
Aupperle, Kenneth E., Archie B. Carroll and John D. Hatfield, 1985, An Empirical Examination of the Relationship between Corporate Social Responsibility and Profitability, The Academy of Management Journal 28, 446-463
Becchetti, Leonardo, Stefania Di Giacomo and Damiano Pinnacchio, 2008, Corporate Social Responsibility and Corporate Performance: Evidence from a Panel of US Listed Companies, Applied Economics 40, 541-567.
Brown, Brad and Susan Perry, 1994, Removing the financial performance halo from Fortune's "most admired companies", Academy of Management Journal 37, 1346-1359
Burns, Tom E. and G.M., Stalker, 1961, The Management of Innovation (Oxford University Press, Oxford)
Campbell, Kevin and Antonio Minguez-Vera, 2008, Gender Diversity in the Boardroom and Firm Financial Performance, Journal of Business Ethics 83, 435-451
Chandler, Gaylen N., Chalon Keller and Douglas W. Lyon, 2000, Unraveling the determinants and consequences of an innovation-supportive organizational culture, Entrepreneurship: Theory and Practice 25, 59-67
Chih, Hsiang-Lin, Hsiang-Hsuan Chih and Tzu-Yin Chen, 2010, On the Determinants of Corporate Social Responsibility: International Evidence on the Financial Industry, Journal of Business Ethics 93, 115-135.
24 Cohen, Wesley M. and Steven Klepper, 1992, The Anatomy of Industry R&D Intensity Distributions, American Economic Review 82, 773-799
Cornell, Bradford and Alan C. Shapiro, 1987, Corporate Stakeholders and Corporate Finance, Financial Management 16, 5-14
Donnelly, George and Ron Fink, 2000, A P&L for R&D, CFO Magazine 16, 44–50.
Friedman, Milton, 1970, The Social Responsibility of Business is to increase its Profits, The New York Times Magazine, September 13
Galbreath, Jeremy, 2006, Does primary stakeholder management positively affect the bottom line?: Some evidence from Australia, Management Decision 44, 1106-1121.
Graves, Samuel B. and Sandra A. Waddock, 1994, Institutional owners and corporate social performance, Academy of Management Journal 37, 1034-1046
Griffin, Jennifer J. and John F. Mahon, 1997, The corporate social performance and corporate financial performance debate: twenty-five years of incomparable research, Business & Society 36, 5-31
Gronroos, Christian, 1990, Service Management: A Management Focus for Service Competition, International Journal of Service Industry Management 1, 6-14
Grow, Brian, Steve Hamm and Louise Lee, 2005, The Debate Over Doing Good,Business Week 39, 47-76
Hull, Clyde E. and Sandra Rothenberg, 2008, The Interactions of Corporate Social Performance with Innovation and Industry Differentiation, Strategic Management Journal 29, 781-789
25 McGuire, Jean B., Alison Sundgren and Thomas Schneeweis, 1988, Corporate Social Responsibility and Firm Financial Performance, The Academy of Management Journal 31, 854-872
McWilliams, Abagail and Donald Siegel, 2000, Corporate Social Responsibility and Financial Performance: Correlation or Misspecification?, Strategic Management Journal 21, 603-609 Mishra, Supriti and Damodar Suar, 2010, Does Corporate Social Responsibility Influence Firm Performance of Indian Companies?, Journal of Business Ethics 95, 571-601.
Moore, Geoff, 2001, Corporate Social and Financial Performance: An Investigation in the U.K. Supermarket Industry, Journal of Business Ethics 34, 299-315.
O'Reilly, Charles A. and Michael Tushman, 2004, The Ambidextrous Organization, Harvard Business Review 82, 74–81
Orlitzky, Marc, Frank L. Schmidt and Sara L. Rynes, 2003, Corporate Social and Financial Performance: A Meta-analysis, Organization Studies 24, 403-441
Padgett, Robert C. and Jose I. Galan, 2010, The Effect of R&D Intensity on Corporate Social Responsibility, Journal of Business Ethics 93, 407-418
Peters, Richard and Michael R. Mullen, 2009, Some Evidence of the Cumulative Effects of Corporate Social Responsibility on Financial Performance, Journal of Global Business Issues 3, 1-14.
Pires, Cesaltina P., Soumoudip Sarkar and Luisa Carvalho, 2008, Innovation in services - how different from manufacturing? The Service Industries Journal 28, 1339-1356
26 Preston, Lee E. and Douglas P. O'Bannon, 1997, The Corporate Social-Financial Performance Relationship: A Typology and Analysis, Business and Society 36, 419-429
Richard, Orland, David Ford and Kiran Ismail, 2006, Exploring the performance effects of visible attribute diversity: the moderating role of span of control and organizational life cycle, International Journal of Human Resource Management 17, 2091-2109.
Ruf, Bernadette, Krishnamurty Muralidhar, Robert M. Brown, Jay J. Janney and Karen Paul, 2001, An Empirical Investigation of the Relationship Between Change in Corporate Social Performance and Financial Performance: A Stakeholder Theory Perspective, Journal of Business Ethics 32, 143-156.
Shropshire, Christine and Amy J. Hillman, 2007, A Longitudinal Study of Significant Change in Stakeholder Management, Business and Society 46, 63-87.
Sundbo, Jon and Faïz Gallouj, 1999, Innovation as a loosely coupled system in services, International Journal of Services Technology and Management 1, 15-36
Surroca, Jordi and Josep A. Tribó, et al., 2008, Managerial Entrenchment and Corporate Social Performance, Journal of Business Finance & Accounting 35, 748-789.
Ullman, Arieh H., 1985, Data in search of a theory: A critical examination of the relationships among social performance, social disclosure, and economic performance of US firms, Academy of Management Review 10, 540-555
Van Ark, Bart, Lourens Broersma and Pim den Hertog, 2003, Services innovation, performance policy: A review, synthesis report in the framework of the project structural information provision on information in services, Strategy, Research & International Cooperation Department Directorate-General for Innovation, Den Haag
27 Wells, Harwell, 2002, The cycles of corporate social responsibility: An historical retrospective for the twenty-first century, Kansas Law Review 51, 77–140.
Wiseman, Joanne, 1982, An evaluation of environmental disclosures made in corporate annual reports, Accounting, Organizations and Society 7, 53-63
Wolfe, Ruth, 1991, The use of content analysis to assess corporate social responsibility, Research in corporate social performance and policy 12, 281-307
Wood, Donna J., 1991, Corporate social performance revisited, Academy of Management Review 16, 691–718.
28
APPENDIX
A I. Results of the Hausman test Table AI
The outcome of the Hausman test for the six different models for both the manufacturing and the service industry.
***, **, * Represent significance at the 1%, 5% and 10% levels respectively. The null hypothesis is that the individual-level effects are adequately modeled by a random-effects model. The complete models can be found in the section below; under the corresponding number.
Model Chi2(5) Prob>chi2 Type of model preferred
