• No results found

The relationship between corporate social responsibility performance and firm performance: The moderating effects of internationalization and stock market development

N/A
N/A
Protected

Academic year: 2021

Share "The relationship between corporate social responsibility performance and firm performance: The moderating effects of internationalization and stock market development"

Copied!
30
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The relationship between corporate social responsibility

performance and firm performance:

The moderating effects of internationalization and stock market

development

By

Karam Abdallah

Master Thesis

Rijksuniversiteit Groningen

Faculty of Economics and Business

February 2019

Name: Karam Abdallah

Student Nr: S3533042

(2)

Abstract

This thesis explores the relationship between firm- and CSR performance, by considering a sample of 4978 firms covering a period from 2002-2017. It specifically examines the impact that prior Tobin´s q has on environmental, social, and governance scores individually. This relationship is examined by considering the possible moderating effects of internationalization and stock market development. The findings of the main relationship indicate that firm performance has a significant positive impact on social- and governance scores, and thus delivers support for the slack resource theory. The theory implies that a change in CSR performance is a result of an increase in available funds which are allocated to improve CSR-related issues, and eventually result in an increase in CSR performance. The results of the moderating effect of internationalization, indicate a negative moderating effect of internationalization, on the relationship between firm performance and CSR performance. This delivers weak support for the pollution haven hypothesis, which implies that firms belonging to energy intensive industries are inclined to foreign direct investment in countries that offer less stringent environmental regulations.

(3)

1. Introduction

This thesis examines the relationship between firm – and CSR performance. I specifically examine the impact that prior firm performance has on CSR performance and utilize the link to the slack resource theory. The theory implies that relatively higher financial performance results in the abundance or “slack” of resources, which in turn creates an opportunity for investing in CSR related issues Waddock and Graves (1997). Following from this, firms with higher firm performance should be able to allocate more funds to CSR-related issues and ultimately increase their CSR performance. This would represent a competitive advantage over firms with lower firm performance. Adding to this, I examine the moderating effect that internationalization and stock market development have on the main relationship. Internationalization refers to the amount of foreign sales that firms generate as a percentage of their total sales. The results of this thesis make a strong case for the support of the “slack resource theory” which is important, since the causality of the relationship between CSR performance and firm performance is still ambiguous and widely discussed throughout past and present literature. In other words, it is not clear whether an increase in firm performance leads to an increase in CSR performance, or vice versa. Moreover, this thesis examines the relationship between prior firm performance and all three components of the environmental, social, governance framework (ESG hereafter), whereas most of the past studies focus on either one component Combs et al. (2006), Rubera and Kirca (2012), Endrikat et al. (2014), Stam et al. (2014) and Rosenbusch et al. (2007), or two components Allouche and Laroche (2005), Wu (2006), Margolis et al. (2009). The research question I seek to answer is the following: “Does an increase prior firm performance result in an increase in CSR performance?

(4)

meeting the implications by Waddock and Graves (1997), to examine the consistency of the relationship between corporate financial and corporate social performance throughout time. This thesis covers all industries, except for the financial industry, and all countries represented in the ASSET4 global database on Datastream, ultimately comprising a sample of 4969 firms. I examine the relationship between prior firm performance and E, S, and G scores by employing OLS regressions. The findings of this thesis indicate that when including prior firm performance poxied by lagged values of Tobin´s q as dependent variables, a highly significant relationship (1% threshold) is observable between prior Tobin´s q and social scores as well as moderately significant (5% threshold) relationship between Tobin´s q and governance scores. Adding to this, a significant moderating effect of internationalization on the relationship between firm performance and environmental scores is observable. Lastly, I find a highly significant (1% threshold) moderating effect of the stock market development on the relationship between firm performance and environmental as well as governance scores. The remainder of this thesis adheres tothe following structure. In section 2 I introduce related past literature and develop my hypotheses. In section 3 I present my data and research design followed by the respective results in section 4. In section 5 I draw conclusions.

2. Related literature and Hypotheses

2.1 Relationship between firm performance and CSR performance

(5)

decrease in CSR performance and vice versa. 5 seminal theories exist which support the different views with regards to the direction of causality and sign of the relationship Preston and Bannon (1997). Slack resource theory states that firms which experience an abundance or slack in available resources or funds are able to allocate these to CSR-related issues and thus, increase their CSR performance Waddock and Graves (1997), Preston and Bannon (1997). Hence, an increase in firm performance leads to an increase in CSR performance. In contrast to this, managerial opportunism theory states that due to the desire towards personal enrichment, managers exploit an increase in firm performance by decreasing investments in CSR-related issues in order to benefit personally. Hence, an increase in firm performance leads to a decrease in CSR performance. good management theory which is based on both stakeholder- and legitimacy theory states that good management, such as meeting the demands of stakeholders with regards to CSR- related issues leads to the legitimization of a firm and ultimately results in an increase in firm performance. Hence, an increase in CSR performance leads to an increase in firm performance. Opposing this, trade-off theory states that investment in CSR-related issues consumes a firm´s funds and represents a competitive disadvantage, which ultimately results in a decrease in firm performance. Hence, an increase in CSR performance leads to a decrease in firm performance. Keeping this in mind, this thesis focuses on the relationship between firm performance and CSR performance in which firm performance plays a leading role. Hence, it is concerned with the changes in CSR performance that occur as a result of a change in prior firm performance. In order to consider firm performance as playing a leading role, values of Tobin´s q are lagged. In other words, the impact of firm performance in one year on CSR performance in the following year is examined.

(6)

corporate financial performance have a significant and positive impact on CSP. Hence, CSP is dependent on CFP and increases as a result of an increase in CFP. Preston and O´Bannon (1997) examine the same relationship yet focus on ROA as a proxy for corporate financial performance and include it as a lagged independent variable and 3 individual CSP measures related to employee retention, quality of products and services, and community and environmental responsibility as dependent variables. Their results show that ROA has a positive and significant leading impact on all three measures of CSP. McGuire et al. (1988), also find a significant relationship between prior accounting performance proxied by ROA and CSR rankings. The aforementioned results all deliver strong support for the slack resource theory in that they deliver evidence for the fact that an increase in CSR performance results from an increase in prior firm performance. In accordance with these results, I formulate the following hypothesis:

Hypothesis 1: An increase in prior Tobin´s q has a positive impact on E, S, and G scores.

2.2 Internationalization as a firm-level moderator

(7)

costly Zahra and Garvis (2000). Over time, the increase in media and analyst coverage adds pressure to such demands Hong and Kacperczyk (2009). As a result, firms address the respective demands by intensifying their employment of CSR-related measures in order to meet the expectations of different stakeholders Brammer, Scharfman et al. (2004), and Pavelin and Porter (2009). Moreover, internationalization reduces managerial risk-taking, by reducing conflicts with regulators, activists and consumers, and allows for increased focuse on CSR Kang (2013). Strike et al. (2006), Examine the relationship between international diversification of U.S. firms and CSR and corporate social irresponsibility (CSI) and find a significant positive relationship between international diversification and CSR performance when including international diversification as an independent variable. In line with the aforementioned theoretical explanations and empirical evidence I formulate the following hypothesis:

Hypothesis 2a: An increase in internationalization has a positive effect on the impact that Tobin´s q has on E, S, and G scores.

(8)

Strike et al. (2006), also find a significant positive relationship between international diversification of U.S. firms and CSI. Moreover, Kolstad and Xing (2002), examine the impact of the strictness of environmental regulations on the location of U.S. energy intensive industries. Their results show that a negative relationship exists between foreign direct investment (FDI) of U.S. chemical and metal industries, and the strictness of environmental regulations existent in the respective foreign target country. In accordance with the theoretical explanations and empirical evidence, I formulate the following hypothesis:

Hypothesis 2b: An increase in internationalization has a negative effect on the impact that Tobin´s q has on E, S, and G scores.

2.3 Stock market development as a country-level moderator

(9)

may not be as reliant on the improvement of CSR-related issues, since information asymmetry is relatively low. Therefore, stakeholders do not have as strong of an expectation towards CSR performance compared to stakeholders in countries with relatively low stock market development. As a result, firms present in countries with relatively low stock market development will allocate less funds generated through an increase in firm performance towards the improvement of CSR-related issues. In accordance with the aforementioned arguments, I formulate the following hypothesis:

Hypothesis 3: An increase in stock market development has a negative effect on the impact that Tobin´s q has on E, S, and G scores.

3. Methodology 3.1 Data and sample

I start by gathering firm-level data from Datastream. I follow Cheng, Iannou and Serafeim (2013), and utilize the ASSET4 global database comprised of 6377 firms, to extract data regarding firm ESG ratings as well as firm-level control variables. Adding to this, I collect country-level data from the World Federation of Exchanges database available on the World Bank website. I specifically collect data on stock market development. I include data of all firms available in the ASSET4 database for the period between 2002-2017. I eliminate financial firms with SIC codes between 6000-6999, due to firm unique characteristics. The final sample is comprised of 4978 firms and 37,033 firm-year observations.

3.2 Methodology, Variables and regression models

(10)

importance of E, S, and G scores has gained popularity and significance throughout time. Limiting the sample period to a single year or a short time period would not capture the true relationship between firm performance and E, S, and G scores. In order to address endogeneity concerns, I lag the independent variables by 1 year. Moreover, I employ firm and year fixed effects.

Dependent and independent variables

I follow Baghat and Bolton (2008), and Gomper, Ishii and Metrick (2003), and include Tobin´s q as a measure of firm performance. I calculate Tobin´s q by adding the market capitalization of a firm, which is calculated as the year-end market price of a firm multiplied by the number of common shares outstanding to total liabilities. I then divide the sum of this calculation by total assets which is comprised of the sum of total current assets, long-term receivables, investment in unconsolidated subsidiaries, other investments, net property, plant, and equipment (PPE) and other assets.

TOBINSQ = (Market capitalization + total liabilities) / Total assets (1)

Environmental score:

The environmental score is comprised of three different categories. The first category being “Resource Use”, is based on 19 different indicators. The second category being “Emissions”, is based on 22 different indicators, and the last category being “Innovation”, is based on 20 different indicators (Thomson Reuters, 2018)

ENV = Environmental score / 100 (2)

Social score:

(11)

different indicators, and the last category “Product responsibility” is based on 12 different indicators (Thomson Reuters, 2018).

SOC = Social score / 100 (3)

Corporate Governance score:

The governance score is comprised of three different categories. The first category being “Management”, is based on 34 different indicators. The second category being “Shareholders”, is based on 12 indicators, and the third indicator being “CSR Strategy”, is based on 8 different indicators (Thomson Reuters, 2018)

GOV = Corporate governance score / 100 (4)

Control variables

(12)

decreases Jensen and Meckling (1976), and thus represents an inducement to exploit corporate resources. This in turn increases agency costs and a firm´s risk premium.

SIZE = Natural logarithm of total assets in dollars (5)

LEVERAGE = Total debt % Total assets /100 (6) R&D = Research and development / total assets in local currency (7)

Firm-level moderator

Based on the examination of 17 studies that deal with the relationship between the level of internationalization of a firm and firm performance by Sullivan (1994) in which he observes that each of these studies applied the ratio of foreign sales to total sales as a proxy for the level of internationalization of a firm, I include the ratio of international sales to total net sales. If the number equals 0.2 or more, the firm is categorized as a multinational corporation (MNC).

FOREIGN = 1 if % of foreign sales to total sales >= 20, 0 otherwise (8)

Country-level moderator

This thesis seeks to investigate the moderating effect of stock market development on the relationship between prior firm performance and E, S, and G scores. I collect data for the level of stock market development from world federation of exchanges database available on the World Bank website. The level of stock market development of a country is calculated by adding up the market capitalization of each firm listed on the domestic stock exchange and dividing the sum by the respective country´s GDP at the end of the year. The market capitalization for each individual firm is calculated by multiplying the share price at the end of the year with the number of shares outstanding. The proxy for stock market development of a specific county i in a specific year t is:

(13)

Regression models

Equation 10 represents the analysis of regressing E, S, and G scores (*SCORE) individually as dependent variables on lagged values of Tobin´s q (TOBINSQ) as independent variables, while controlling for firm size (SIZE), firm leverage (LEV), R&D expenditures (R&D), and the moderating effect of internationalization (FOREIGN*TOBINSQ). Equation 11 analyses the same relationship, while also considering the moderating effect of stock market development (MKTDEV*TOBINSQ). I employ firm and year fixed effects in both equations.

*SCOREi,t = b0 + b1TOBINSQi,t-1 + b2SIZEi,t-1 + b3LEVi,t-1 +b4R&Di,t-1 + b5FOREIGNi,t-1 +

b6FOREIGN*TOBINSQi,t-1 + µi,t (10)

*SCOREi,t = b0 + b1TOBINSQi,t-1 + b2SIZEi,t-1 + b3LEVi,t-1 +b4R&Di,t-1 +b5FOREIGNi,t-1 +

b6FOREIGNi,t-1 * TOBINSQi,t-1 + b7MKTDEVj,t-1 + b8MKTDEVj,t-1 * TOBINSQi,t-1 + µi,t (11)

I test hypotheses H1 by considering equation 10 and specifically the coefficient b1, which

signifies the response of either the E, S, or G score to variations in TOBINSQ. If the coefficient is positive and significant, I can conclude that prior values of Tobin´q have a positive and significant impact on E, S, or G scores. If the coefficient is negative and significant, I can conclude that prior Tobin´s q has a negative and significant impact on E, S, or G scores. In order to test hypothesis 2a and 2b, I consider the coefficient b6 in equation 10. If the coefficient

(14)

E, S, or G scores. In order to test hypothesis 3, I consider the coefficient b8 of equation 11. If

the coefficient is positive and significant, I can conclude that stock market development has a positive and significant moderating effect on the relationship between TOBINSQ and E, S, or G scores. Hence, an increase in stock market development increases that impact that Tobin´s q has on E, S, or G scores. If the coefficient is negative and significant, I can conclude that stock market development decreases the impact that TOBINSQ has on E, S, or G scores.

4. Results

4.1 Descriptive statistics

Table 1 reports the descriptive statistics of the firm- and country-level variables employed in the empirical analysis. In order to address the existence of outliers, all variables except the country-level variable stock market development proxied by MKTDEV are winsorized at 1% of both tails. The dependent variable TOBINSQ reports a mean, median, and standard deviation of 1.87, 1.43 and 1.36 respectively. The independent variable ENV reports a mean, median, and standard deviation of 0.52, 0.52, and 0.32 respectively. The independent variable SOC reports a mean, median and standard deviation of 0.52, 0.53, and 0.31 respectively. The independent variable GOV reports a mean, median, and standard deviation of 0.50, 0.55, and 0.30 respectively.

Table 1. Descriptive statistics

Variable N Mean Median Minimum Maximum Std. Dev

TOBINSQ 37033 1.87 1.43 0.60 9.55 1.36 ENV 37033 0.52 0.52 0.94 0.97 0.32 SOC 37033 0.52 0.53 0.05 0.96 0.31 GOV 37033 0.50 0.55 0.01 0.96 0.30 SIZE 37033 15.24 15.27 8.96 18.58 1.57 LEV 37033 0.25 0.24 0 0.89 0.18 R&D 37033 0.02 0 0 0.43 0.05 FOREIGN 37033 0.55 1 0 1 0.51 MKTDEV 37033 131.29 100.79 0 1109.26 187.19

(15)

4.2 Correlation analysis

Table 2 reports the correlation coefficients of the dependent as well as lagged independent, control and moderating variables employed in the empirical analysis. The results indicate that the environmental (ENV) and social (SOC) scores are negatively correlated with firm performance (TOBINSQ). On the other hand, the governance score (GOV) is positively correlated to firm performance (TOBINSQ). Firm size (SIZE) and leverage (LEV) are both negatively correlated to firm performance (TOBINSQ), whereas research and development (R&D) is positively correlated to firm performance (TOBINSQ). The moderating variables of internationalization (FOREIGN) and stock market development (MKTDEV) are both negatively correlated to firm performance (TOBINSQ).

Table 2. Sample correlation

TOBINSQ ENV SOC GOV SIZE LEV R&D FOREIGN MKTDEV

TOBINSQ 1 ENV -0.1611* 1 SOC -0.0906* 0.8054* 1 GOV 0.0687* 0.1770* 0.3126* 1 SIZE -0.3735* 0.4360* 0.4016* 0.0674* 1 LEV -0.1981* 0.0547* 0.0573* 0.0351* 0.2738* 1 R&D 0.3205* 0.0125* -0.0159* 0.0089 -0.3045* -0.1558* 1 FOREIGN -0.0717 0.2658* 0.2475* 0.1232* 0.2519* -0.0117* 0.0335* 1 MKTDEV -0.0009* -0.1400* -0.1270* -0.0053 0.0153* -0.0123* -0.0220* -0.0168* 1

Note: This table includes information regarding the correlation of the variables employed throughout the regression analyses. * denotes statistical significance at the 5% level.

4.3 Multivariate analysis

(16)

TOBINSQ throughout model 1-5 are insignificant, I reject hypothesis 1 since accepting it would imply the significance of the relationship between TOBINSQ and all three types of scores. The positive and significant (1% threshold) coefficients of size throughout models 1-5, indicate that firm size (SIZE) has a positive and significant relationship with the environmental score (ENV). This implies that an increase in firm size has a positive impact on the environmental score. The positive yet insignificant coefficients of LEV, indicate that an increase in firm leverage would have a positive effect on environmental scores. Nevertheless, it would be inappropriate to make any inferences due to the insignificance of the coefficients. The positive and significant (1% threshold) coefficients of R&D, indicate that an increase in R&D expenditures has a positive impact on the environmental score (ENV).

(17)

Table 3. Firm performance, environmental scores and internationalization Model 1 ENV b/se Model 2 ENV b/se Model 3 ENV b/se Model 4 ENV b/se Model 5 ENV b/se TOBINSQ 0.002 [0.001] [0.001] 0.002 [0.001] 0.002 [0.001] 0.002 [0.001] 0.002 SIZE 0.045*** [0.002] 0.045** [0.002] 0.045*** [0.002] 0.045*** [0.002] 0.046*** [0.002] LEV 0.006 [0.009] [0.009] 0.006 [0.009] 0.006 [0.009] 0.006 R&D 0.183*** [0.051] 0.183*** [0.051] 0.186*** [0.051] FOREIGN -0.001 [0.003] [0.004] 0.008 FOREIGNxTOBINSQ -0.005** [0.002] Constant -0.336*** [0.032] -0.336*** [0.032] -0.350*** [0.032] -0.350*** [0.032] -0.358*** [0.032]

Year FE Yes Yes Yes Yes Yes

Firm FE Yes Yes Yes Yes Yes

Adj R2 0.184 0.184 0.184 0.184 0.184

Observations 36106 36077 36077 36077 36077

Note: This table reports the results from regressing the environmental score divided by 100 (ENV) on firm performance (TOBINSQ), controls and the firm-level moderator level of foreign sales (FOREIGN). TOBINSQ equals the sum of a firm´s market capitalization and total liabilities divided by the firm´s total assets. SIZE is the natural logarithm of a firm´s total assets in $. LEV equals a firm´s total debt to total assets. R&D equals a firm´s total R&D expenses divided by total assets. FOREIGN equals 1 if the % of foreign sales to total sales is equal or larger than 20, 0 otherwise. I report the standard errors below each coefficient estimate in brackets. The asterisks ***, **, and * signify the statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 4 reports the results of the moderating effect of stock market development. The highly significant (1% threshold) and negative coefficient of the interaction term MKTDEVxTOBINSQ in model 7 indicates that stock market development has a negative moderating effect on the relationship between the firm performance (TOBINSQ) and the environmental score (ENV). It specifically implicates, that an increase in the level of stock market development decreases the positive impact of firm performance (TOBINSQ) on the environmental score (ENV). This is in line with hypothesis 3, which implies that an increase in stock market development has a negative effect on the impact that firm performance has on E, S, and G scores. Adding to this, the significant (1% threshold) and positive coefficient of MKTDEV in model 6 indicates that the level of stock market development has a positive impact on the environmental score (ENV).

(18)

indicate that firm performance has a positive impact on the social score (SOC). The positive and significant (1% threshold) coefficients of SIZE indicate that firm size has a positive impact on the social score (SOC). The negative and significant (5% threshold) coefficients of LEV indicate that firm leverage has a negative impact on the social score (SOC). The positive and significant (1%) coefficients of R&D indicate that R&D expenditures have a positive impact on the social score (SOC). The negative coefficient of the interaction term FOREIGNxTOBINSQ in model 12 indicates that throughout MNCs, the positive impact of TOBINSQ on the social score is decreased. Since the coefficient is not significant, it is not appropriate to make any final inferences regarding the moderating effect of internationalization. The insignificance of the interaction term also leads to the rejection of hypothesis 2b, which predicts that the level of internationalization negatively effects the relationship between Tobin´s q and E, S, and G scores. The positive and significant (10% threshold) coefficient of FOREIGN in model 11 indicates that the level of internationalization has a positive impact on the social score (SOC).

Table 5. Firm performance, social scores and internationalization

Model 8 SOC b/se Model 9 SOC b/se Model 10 SOC b/se Model 11 SOC b/se Model 12 SOC b/se TOBINSQ 0.005*** [0.001] 0.005*** [0.001] 0.005*** [0.001] 0.005*** [0.001] 0.006*** [0.001] SIZE 0.043*** [0.002] 0.043*** [0.002] 0.044*** [0.002] 0.044*** [0.002] 0.044*** [0.002] LEV -0.024** [0.009] -0.023** [0.009] -0.024** [0.009] -0.024** [0.009] R&D 0.178*** [0.049] 0.176*** [0.050] 0.178*** [0.050] FOREIGN 0.007* [0.003] 0.012** [0.004] FOREIGNxTOBINSQ -0.003 [0.002] Constant -0.339*** [0.031] -0.342*** [0.031] -0.356*** [0.031] -0.354*** [0.031] -0.358*** [0.031]

Year FE Yes Yes Yes Yes Yes

Firm FE Yes Yes Yes Yes Yes

Adj R2 0.254 0.254 0.254 0.254 0.254

Observations 36106 36077 36077 36077 36077

(19)

Table 6 reports the results of including the level of stock market development (MKTDEVxTOBINSQ) as a moderator of the relationship between TOBINSQ and the social score (SOC). The negative coefficient of the interaction term MKTDEVxTOBINSQ in model 14, indicates that in countries with a more developed stock market, the positive impact of firm performance (TOBINSQ) on the social score (SOC) is mitigated. Yet, due to the insignificance of the interaction term, one can infer that the level of stock market development does not moderate the relationship between TOBINSQ and SOC. On the other hand, the positive and significant (1% threshold) coefficient of MKTDEV in model 13 indicates that the level of stock market development has a positive impact on the social score (SOC).

(20)

Table 7. Firm performance, governance scores and internationalization Model 15 GOV b/se Model 16 GOV b/se Model 17 GOV b/se Model 18 GOV b/se Model 19 GOV b/se TOBINSQ 0.003** [0.001] 0.003** [0.001] 0.002** [0.001] 0.002** [0.001] [0.001] 0.003* SIZE 0.039*** [0.002] 0.039*** [0.002] 0.040*** [0.002] 0.040*** [0.002] 0.040*** [0.002] LEV -0.008 [0.008] [0.008] -0.008 [0.008] 0.007 [0.008] -0.008 R&D 0.115** [0.043] 0.117** [0.043] 0.117** [0.043] FOREIGN -0.006* [0.003] [0.004] -0.005 FOREIGNxTOBINSQ -0.001 [0.002] Constant -0.271*** [0.27] -0.269*** [0.27] -0.278*** [0.27] -0.279*** [0.27] -0.280*** [0.27]

Year FE Yes Yes Yes Yes Yes

Firm FE Yes Yes Yes Yes Yes

Adj R2 0.072 0.072 0.072 0.072 0.072

Observations 36026 35997 35997 35997 35997

Note: This table reports the results from regressing the governance score divided by 100 (GOV) on firm performance (TOBINSQ), controls and the firm-level moderator level of foreign sales (FOREIGN). TOBINSQ equals the sum of a firm´s market capitalization and total liabilities divided by the firm´s total assets. SIZE is the natural logarithm of a firm´s total assets in $. LEV equals a firm´s total debt to total assets. R&D equals a firm´s total R&D expenses divided by total assets. FOREIGN equals 1 if the % of foreign sales to total sales is equal or larger than 20, 0 otherwise. I report the standard errors below each coefficient estimate in brackets. The asterisks ***, **, and * signify the statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 8 reports the results of including stock market development (MKTDEVxTOBINSQ) as a moderator of the relationship between TOBINSQ and GOV. The negative and significant (1% threshold) coefficient of the interaction term MKTDEVxTOBINSQ in model 21, indicates that the level of stock market development decreases the positive impact of TOBINSQ on the governance score. Hence, an increase in stock market development decreases the positive impact of firm performance (TOBINSQ) on the governance score (GOV). This is in accordance with hypothesis 3 which implies that an increase in stock market development has a negative effect on the impact that Tobin´s q has on E, S, and G scores. The positive and insignificant coefficient of MKTDEV in model 20, indicates that stock market development has a positive yet insignificant impact on the governance score (GOV).

4.4 Robustness analysis

(21)

aforementioned equations. I specifically excluded data from the U.S. and Japan, since the three largest stock markets are based in both respective countries. I also excluded data from the UK since it is known as the financial center of Europe.

(22)

when excluding data from U.S., UK, and Japan. In comparison to previous results of the relationship between stock market development (MKTDEV) and the social score (SOC), has become insignificant. The results of model 24 indicate that the relationship between R&D expenditures (R&D) and the governance score (GOV) has become insignificant in the country subsample. Overall the results of excluding data from the U.S., UK, and Japan holds except for the impact of prior firm performance on the social score (model 23) and a slight decrease in the significance thresholds of some relationships.

5. Conclusions

(23)

performance follows. Specifically, as a result of an increase in Tobin´s q, social and governance scores increase. These finding delivers empirical evidence for the support of the slack resource theory which implies, that an increase in CSR performance is a result of an increase in the availability of funds which can be allocated to the improvement of CSR-related issues. In contrast to this, the findings of the moderating effect of internationalization on the main relationship oppose the stakeholder- and legitimacy theory, and rather provide weak evidence for the pollution haven hypothesis. This is the case, since an increase in foreign exposure increases the number of stakeholders which in turn increases the amount of demands and expectations which have to be satisfied in order to gain legitimacy. Due to this, MNC´s seek to increase their involvement in CSR related issues. The findings of this thesis on the other hand, show that internationalization has a negative effect on the relationship between firm performance and environmental scores. Hence, firms allocate less funds generated through an increase in firm performance to environmental issues. This might be the case since firms are inclined to locate in foreign locations that offer less stringent environmental regulations in order to save costs otherwise necessary to comply with such regulations. Nevertheless, this finding only delivers weak support for the pollution haven hypothesis, since the main sample is not solely comprised of firms belonging to energy intensive industries but rather covers a broad range of industries.

(24)
(25)

References

Aliyu, A., Ismail, N., 2015. Foreign direct investment and pollution haven: does energy consumption matter in african countries? International Journal of Economics and Management 9(s), 21-39.

Allouche, J., Laroche, P., 2005. A meta-analytical investigation of the relationship between corporate social and financial performance. Revue de Gestion Des Resources Humaines 57, 1-18.

Attig, N., Boubakri, N., El Ghoul, S., Guedhami, O., 2014. Firm internationalization and corporate social responsibility. Journal of Business Ethics 134, 171-197.

Aupperle, K, E., Carroll, A, B., Hatfield, J, D., 1985. An empirical examination of the relationship between corporate social responsibility and profitability. The Academy of Management Journal 28(2), 446-463.

Berger, A, N., di Patti, E, B., 2006. Capital structure and firm performance: a new approach to testing agency theory and an application to the banking industry. Journal of Banking & Finance 30, 1065-1102.

Brammer, S, J., Pavelin, S., Porter, L, A., 2009. Corporate charitable giving, multinational companies and countries of concern. Journal of Management Studies 46(4), 575-596.

Carpenter, R, E., Peterson, B., 2002. Ist the growth of small firms constrained by internal finance? Review of Economics and Statistics 84(2), 298-309.

Cheng, B., Ioannou, I., Serafeim, G., 2014. Corporate social responsibility and access to finance. Strategic Management Journal 35, 1-23.

Combs, J., Yongmei, L., Hall, A., Ketchen, D., 2006. How much do high-performance work practices matter? a meta-analysis of their effects on organizational performance. Personell Psychology 59(3), 501-528.

Demirgüç-Kunt, A., Maksimovic, V., 1996. Stock market development and financing of firms. The World Bank Economic Review 10(2), 341-369.

Endrikat, J., Guenther, E., Hoppe, H., 2014. Making sense of conflicting empirical findings: a meta-analytic review of the relationship between corporate environmental and financial performance. European Management Journal 32(5), 735-751.

Geringer, J, M., Beamish, P, W., DaCosta, R, C., 1989. Diversification strategy and internationalization: implication for MNE performance. Strategic Management Journal 10. 109-119.

Griliches, Z., 1979. Issues in assessing the contribution of R&D to productivity growth. Bell Journal of Economics 10(1), 92-116.

Hofstede, G., 2011. Dimensionalizing cultures: the Hofstede model in context. Online Readings in Psychology and Culture 2(1), 1-26.

(26)

Jensen, M, C., Meckling, W, H., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financal Economics 3, 305-360.

Jones, T, M., 1995. Instrumental stakeholder theory: a synthesis of ethics and economics. The Academy of Management Review 20(2), 404-437.

Kang, J., 2013. The relationship between corporate diversification and corporate social performance. Strategic management journal 34(1), 94-109.

Levine, R., 2005. Finance and growth: theory and evidence. In: Durlauf, S, A., Aghion, P. (Ed.), Handbook of economic growth. Vol. 1a, Amsterdam, Netherlands, pp. 865-934.

Lichtenberg, F., Siegel, D., 1991. The impact of R&D investment on productivity: new evidence using linked R&D-LRD data. Economic inquiry 29, 203-228.

Lin-Hi, N., Müller, K., 2013. The CSR bottom line: preventing corporate social responsibility. Journal of Business Research 66, 1928-1936.

Margolis, J, D., Elfenbein, H, A., Walsh, J, P., 2009. Does it pay to be good … and does it matter? a meta-analysis of the relationship between corporate social and financial performance. SSRN Electronic Journal, 1-68

McGuire, J, B., Sundgren, A., Schneeweis, T., 1988. Corporate social responsibility and firm financial performance. The Academy of Management Journal 31(4). 854-872.

Mitchell, R, K., Agle, B, R., Wood, D, J., 1997. Toward a theory of stakeholder identification and salience: defining the principle of who and what really counts. The Academy of Management Review 22(4), 853-886.

Morck, R., Yeung, B., Yu, W., 2000. The information content of stock markets: why do emerging markets have synchronous price movements? Journal of Financial Economics 58, 215-260.

Rosenbusch, N., Bausch, A., Galander, A., 2007. The impact of environmental characteristics on firm performance: a meta-analysis. Academy of Management Annual Meeting. Proceedings 1-6.

Sharfman, M. P., Shaft, T. M., Tihanyi, L., 2004. A model of the global and institutional antecedents of high-level corporate environmental performance. Business and society 43(1), 6-36.

Stam, W., Arzlanian, S., Elfring, T., 2014. Social capital of entrepreneurs and small firm performance: a meta-analysis of contextual and methodological moderators. Journal of Business Venturing 29(1), 152-173.

Strike, V., Gao, J., Bansal, P., 2006. Being good while being bad: social responsibility and the international diversification of US firms. Journal of International Business Studies 37(6), 254-280.

(27)

2018. Thomson Reuters ESG scores methodology. Thomson Reuters Eikon, 1-18.

Waddock, S., Graves, S., 1997. The corporate social performance-financial performance link. Strategic Management Journal 18(4), 303-319.

Wu, M, L., 2006. Corporate social performance, corporate financial performance, and firm size: a meta-analysis. The Journal of American Academy of Business 8(1), 163-171.

Xing, Y., Kolstad, C., 2002. Do lax environmental regulations attract foreign direct investment? Environmental and Resource Economics 21, 1-22.

Zahra, S, A., Garvis, D, M., 2000. International corporate entrepreneurship and firm performance: the moderating effect of international environment hostility. Journal of Business Venturing 15, 469-492.

(28)

Appendices

Appendix A Table 4. Firm performance, environmental scores and stock market development

Model 6 ENV b/se Model 7 ENV b/se TOBINSQ 0.004** [0.001] 0.006*** [0.002] SIZE 0.045*** [0.002] 0.045*** [0.002] LEV 0.006 [0.009] [0.009] 0.006 R&D 0.186*** [0.051] 0.186*** [0.051] FOREIGN 0.007 [0.004] [0.004] 0.008 FOREIGNxTOBINSQ -0.005** [0.002] -0.005** [0.002] MKTDEV 0.000*** [0.000] 0.000*** [0.000] MKTDEVxTOBINSQ -0.000*** [0.000] Constant -0.356*** [0.032] -0.358*** [0.032]

Year FE Yes Yes

Firm FE Yes Yes

Adj R2 0.185 0.185

Observations 36077 36077

Note: This table reports the results from regressing the environmental score divided by 100 (ENV) on firm performance (TOBINSQ), controls and the firm-level moderator level of foreign sales (FOREIGN). TOBINSQ equals the sum of a firm´s market capitalization and total liabilities divided by the firm´s total assets. SIZE is the natural logarithm of a firm´s total assets in $. LEV equals a firm´s total debt to total assets. R&D equals a firm´s total R&D expenses divided by total assets. FOREIGN equals 1 if the % of foreign sales to total sales is equal or larger than 20, 0 otherwise. MKTDEV equals the total market value of domestic firms listed on the domestic stock exchange as a % of GDP and is interacted with the independent variable TOBINSQ. Information for the level of stock market development in Taiwan over the sample period and the UK post 2008 is not available on the World Federation of exchanges database. I report the standard errors below each coefficient estimate in brackets. The asterisks ***, **, and * signify the statistical significance at the 1%, 5%, and 10% levels, respectively.

Appendix B Table 6. Firm performance, social scores and stock market development

Model 13 SOC b/se Model 14 SOC b/se TOBINSQ 0.006*** [0.001] 0.007*** [0.002] SIZE 0.043*** [0.002] 0.043*** [0.002] LEV -0.024** [0.009] -0.024** [0.009] R&D 0.177*** [0.050] 0.177*** [0.050] FOREIGN 0.011* [0.004] 0.011** [0.004] FOREIGNxTOBINSQ -0.002 [0.002] [0.002] -0.003 MKTDEV 0.000* [0.000] 0.000** [0.000] MKTDEVxTOBINSQ -0.000 [0.000] constant -0.357*** [0.031] -0.357*** [0.031]

Year FE Yes Yes

Firm FE Yes Yes

Adj R2 0.254 0.254

Observations 36077 36077

(29)

Appendix C Table 8. Firm performance, governance scores and stock market development Model 20 GOV b/se Model 21 GOV b/se TOBINSQ 0.003* [0.001] 0.010*** [0.001] SIZE 0.040*** [0.002] 0.039*** [0.002] LEV -0.008 [0.008] [0.008] -0.007 R&D 0.117** [0.043] 0.116** [0.043] FOREIGN -0.005 [0.004] [0.004] -0.004 FOREIGNxTOBINSQ -0.000 [0.002] [0.002] -0.001 MKTDEV 0.000 [0.000] 0.000*** [0.000] MKTDEVxTOBINSQ -0.000*** [0.000] Constant -0.279*** [0.027] -0.283*** [0.027]

Year FE Yes Yes

Firm FE Yes Yes

Adj R2 0.072 0.076

Observations 35997 35997

(30)

Appendix D Table 9. Firm performance – E, S, and G scores – Country subsample comparison All countries excluding (U.S. + UK + JP) Model 22 ENV b/se All countries excluding (U.S. + UK + JP) Model 23 SOC b/se All countries excluding (U.S. + UK + JP) Model 24 GOV b/se All countries Model 7 ENV b/se All countries Model 14 SOC b/se All countries Model 21 GOV b/se TOBINSQ 0.006** [0.002] [0.002] 0.004 0.008*** [0.002] 0.006*** [0.002] 0.007*** [0.002] 0.010*** [0.001] SIZE 0.039*** [0.003] 0.041*** [0.003] 0.039*** [0.002] 0.045*** [0.002] 0.0043*** [0.002] 0.039*** [0.002] LEV -0.018 [0.013] [0.012] -0.024 [0.012] -0.019 [0.009] 0.006 -0.024** [0.009] [0.008] -0.007 R&D 0.164 [0.096] 0.269** [0.093] [0.088] 0.136 0.186*** [0.051] 0.177*** [0.050] 0.116** [0.043] FOREIGN 0.008 [0.006] [0.006] 0.006 [0.005] -0.001 [0.004] 0.008 0.011** [0.004] [0.004] -0.004 FOREIGNxTOBINSQ -0.006* [0.003] [0.002] -0.001 [0.002] 0.001 -0.005* [0.002] [0.002] -0.003 [0.002] -0.001 MKTDEV 0.000*** [0.000] [0.000] 0.000 0.000*** [0.000] 0.000*** [0.000] 0.000** [0.000] 0.000*** [0.000] MKTDEVxTOBINSQ -0.000** [0.000] [0.000] -0.000 -0.000*** [0.000] -0.000*** [0.000] [0.000] -0.000 -0.000*** [0.000] Constant -0.196*** [0.040] -0.255*** [0.039] -0.424*** [0.037] -0.358*** [0.032] -0.357*** [0.03] -0.283*** [0.027]

Year FE Yes Yes Yes Yes Yes Yes

Firm FE Yes Yes Yes Yes Yes Yes

Adj R2 0.221 0.280 0.178 0.185 0.254 0.076

Observations 17764 17764 17714 36077 36077 35997

Referenties

GERELATEERDE DOCUMENTEN

The  last  two  chapters  have  highlighted  the  relationship  between  social  interactions   and  aspiration  formation  of  British  Bangladeshi  young  people.

Moreover, in the lottery, participants who have a negative social relationship are more likely to choose an option with a larger outcome discrepancy compared to those who have

perspective promoted by these teachers is positive or negative, the very fact that students are being told that the government does not care about their identity, history and

How does the novel function as a technology to recall, create and shape prosthetic memories on the individual level of the reader and in turn create or maintain the cultural

In order to perform the measurements for perpendicular polarization, the λ/2 plate is rotated by 45°, to rotate the laser polarization by 90°.The measurements were performed

In this research, the main investigated relationship is the possible impact the two different predictors (ESG pillar scores and ESG Twitter sentiment) have on the

This result is consistent with table 3, which shows that low-rated stocks based on social screening have the highest average monthly return and NYSE Composite Index has the

But we have just shown that the log-optimal portfolio, in addition to maximizing the asymptotic growth rate, also “maximizes” the wealth relative for one