An empirical study of CSR on financial performance in the
airline industry (2005 to 2015)
Amsterdam Business School
Name Iris Smit
Student number 10562788
Program Economics & Business Specialization Finance & Organization Number of ECTS 12
Supervisor Ilko Naaborg Target completion 31 / 01 / 2017
Abstract
Although corporate social responsibility (CSR) is an important aspect in the business strategies of companies these days, the relationship between CSR and the financial performance of airline companies has barely been studied. This study examines the impact of environmental-, social-, and governance performance on the return on assets and the return on equity of airline companies. Results of this study show a positive impact of environmental- and governance performance on the return on assets. Thus, airline companies that implement environmental- and governance responsible activities, usually have a higher return on assets. Moreover, the overall CSR performance has a positive impact on the return on assets. However, the return on equity of airline companies is only positively influenced by governance performance. Thus, airline companies that implement governance responsible activities generally have a higher return on equity.
2
Statement of Originality
This document is written by Iris Smit, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Table of Contents
1. Introduction ... 4
2. Literature review ... 5
2.1 Corporate social responsibility ... 6
2.2 The measurement of corporate social responsibility ... 7
2.3 Empirical findings on CSR and its impact on the CFP ... 8
2.4 Conclusions on the literature ... 10
3. Hypothesis, methodology and data ... 12
3.1 Methodology ... 12
3.2 Data and descriptive statistics ... 16
4. Analysis... 19
4.1 Empirical results ... 19
4.2 Robustness check ... 22
5. Conclusion and discussion ... 24
References ... 27
4
1. Introduction
Trade liberalization and technological change have a major impact on society and the environment (Habisch et al., 2005). According to Cowper-Smith & de Grosbois (2011), public and governments worldwide increasingly expect companies to adjust their business practices in order to improve social and environmental sustainability of their operations. Businesses should make a voluntary commitment to contribute to economic development while behaving in a social and environmental responsible way. The term corporate social responsibility (CSR) has been adopted to indicate this voluntary commitment.
McGuire, Sundgren, and Schneeweis (1988) argue that there are three views on the relationship between corporate social responsibility and corporate financial performance. One view states that companies face a trade-off between social responsibility and corporate financial performance. This means that firms would incur costs from corporate social responsibility, rather than financially gaining from it (Vance, 1975). The opposite view is that the costs of implementing corporate social responsibility strategies are minimal and that firms would financially benefit from socially responsible actions (Soloman & Hansen, 1985). The last view is that there are significant costs attached to corporate social responsibility strategies, but these costs are offset by the decrease in other costs of the company (McGuire, Sundgren, & Schneeweis, 1988). Either way, according to Idowu & Louche (2011), an increasing number of companies have implemented corporate social responsibility strategies during the first decade of the twenty-first century. Producing reports on CSR is now very common for many companies, and in the second decade of this century it is almost considered risky not to take CSR into account (Idowu & Louche, 2011).
An increasing number of studies has been done on corporate social responsibility and its impact on the corporate financial performance. However, there is only limited research available for the impact of implementing CSR strategies on the financial performance in the airline industry (Lee & Park, 2010). Meanwhile, the airline industry has a major impact on the environment and society. The industry is one of the major consumers of fossil fuels, and therefore it is under public pressure to reduce environmental impacts (Lee & Park, 2010). Moreover, the airline industry should be more conscious and responsible for the impact it has on society (Becken, 2007). Under the pressure of the public an increasing number of airlines is now implementing CSR strategies. One example of an airline implementing CSR strategies is Air France-KLM. According to the Dow Jones Sustainability Index of September 2016, Air France-KLM is one of the industry’s
5
leaders in the field of sustainability. They undertake several measures to minimize their impact on climate change, such as researching usage of biofuel instead of kerosene. (Air France-KLM, 2016). In addition, they offer an easy way for the consumer to compensate for the CO2 emission of their flight with their CO2ZERO-program. Furthermore, they support several social projects. For example, Air France Foundation helps disadvantaged children from all over the world (Air France-KLM, 2016).
Because of the importance of corporate social responsibility in decision making policies of companies, the impact of corporate social responsibility on the financial performance of companies is an important topic (McGuire, Sundgren, & Schneeweis, 1988). However, there is a lack of research done on the impact of CSR on the financial performance of airline companies. The aim of this research is to determine the impact of corporate social responsibility on the corporate financial performance in the airline industry. In this research the impact of three different pillars of corporate social responsibility (environmental-, social-, and governance performance) on the financial performance is tested. Moreover, the impact of the overall CSR performance on the financial performance is tested. All the airline companies contained in the ASSET4 database are included in the analysis for the years 2005 to 2015. The ASSET4 database contains scores for corporate social responsibility on a scale of 0 to 100, where 0 indicates a bad CSR performance and 100 means a perfect performance of CSR activities. The financial performance is measured by data retrieved from COMPUSTAT.
In the following section the main existing literature on corporate social responsibility and its impact on the corporate financial performance is reviewed. The third section describes the methodological approach that is used in this research. This section includes the model that is used, the hypotheses, and a description of the data. In section four, the main empirical results are presented and a robustness check is added. Finally, in the last section a conclusion is given. Implications are discussed, and suggestions for future research are provided in this section.
2. Literature review
An increasing number of published papers connects corporate social responsibility to the financial performance of companies. In this section the previous literature regarding corporate social responsibility and its relation with corporate financial performance is discussed. This section is divided in four subsections. In the first subsection corporate social responsibility is discussed.
6
Different definitions of corporate social responsibility are given. In the second subsection the measurement of corporate social responsibility is explained. The third subsection discusses the empirical findings of the relationship between corporate social responsibility and the financial performance in previously written literature. In the last subsection a brief summary of the main results in previous literature is given.
2.1 Corporate social responsibility
Corporate social responsibility is actually a broad term for different concepts relating to business ethics and sustainable behavior within companies (Matten & Moon, 2008). According to Matten and Moon (2008), defining CSR is therefore not easy. There is plenty evidence that the meaning of CSR varies across countries (Matten & Moon, 2008). However, all of the CSR definitions have one thing in common: they include corporate activities that have a positive impact on society and/or environment (Caroll, 1979). One of the first definitions of corporate social responsibility was published by Bowen (1953). He stated the definition:
“It refers to the obligations of businessmen to pursue those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society.”
Tsoutsoura (2004) uses a more explicit definition for corporate social responsibility:
“Policies, practices, and programs that are integrated into business operations, supply chains, and decision-making processes throughout the company and usually include issues related to business ethics, community investment, environmental concerns, governance, human rights, the marketplace as well as the workplace.”
To communicate the CSR strategies that companies implement in their business operations, CSR reports are usually used (Arvidsson, 2010; Dando and Swift, 2003). Additionally, some companies attract attention to their CSR program through press releases or by sharing information with employees and customers (Kuo et al., 2016). Also, corporate social responsibility has been studied by researchers in the last decades. An increasing number of literature connects corporate
7
social responsibility to the financial performance of companies. Previous empirical findings regarding the impact of corporate social responsibility on the financial performance is discussed in the third subsection.
2.2 The measurement of corporate social responsibility
As stated before, corporate social responsibility is a vague concept which can be interpreted in different ways. Therefore, it is not easy to measure corporate social responsibility. According to Igalens and Gond (2005), researchers usually use five different measurements for CSR. The first measurement method is based on the annual reports of the firm. Using this method, researchers usually measure the quantity of environmental, social, and governance information included in the annual reports. However, this measurement method might be subject to subjectivity, as the annual reports are produced by the companies themselves (Igalens & Gond, 2005). Another measure for CSR is a pollution index (Igalens & Gond, 2005). Pollution indexes are usually put out by independent parties. Therefore, this method is very objective. However, this measurement method is very limited as it only deals with one sub-dimension of the environmental part of CSR (Igalens & Gond, 2005). A third method for measuring CSR is a research based on questionnaires among the employees or the executives of the company (Igalens & Gond, 2005). In these questionnaires questions regarding different dimensions of CSR are included. This method is very perceptual, since it only measures employees’- or executives’ impressions of the implementation of CSR activities in the business’ strategies. Thus, this method is also highly subjective and can be manipulated. The forth method to measure CSR is making use of indicators for corporate reputation. This method is purely perceptual and derived from surveys of external persons such as financial analysts (Igalens & Gond, 2005). Finally, data produced by independent measurement organizations is a method to measure CSR (Igalens & Gond, 2005). Depending on the independent organization’s operational method, this method might be objective.
In this thesis corporate social responsibility is measured by data that is produced by an independent organization: Thomson Reuters. In this database, corporate social responsibility is operationalized and measured by the environmental, social, and governance (ESG) factors. The measurement of CSR by using the ESG factors is widely accepted by researchers and by capital markets (Sassen, Hinze, & Hardeck, 2016). Moreover, the data that is used by Thomson Reuters
8
is quantitative (Thomson Reuters corporate responsibility ratings, 2013). Therefore, this measurement method of CSR is not subject to subjectivity.
2.3 Empirical findings on corporate social responsibility and its impact on the financial performance
As stated before, an increasing number of literature connects corporate social responsibility to corporate financial performance. Those previous studies generated mixed results regarding the impact of corporate social responsibility on the corporate financial performance.
Makni, Francoeur, and Bellavance (2009) use a sample of Canadian firms from different sectors during 2004 and 2005 in their study for corporate social responsibility. They research the impact of CSR on three different financial performance measures: stock market returns, return on assets, and return on equity. They test these effects by using the Granger causality approach. This means, the impact of corporate social responsibility in 2004 and of the financial performance in 2004 on the corporate social responsibility in 2005 is measured. Makni, Francoeur, and Bellevance (2009) neither find significant impact of corporate social responsibility on the return on assets nor significant impact on the return on equity. However, they do find a significant negative impact of corporate social performance on stock market returns. They conclude that corporate social responsibility initiatives appear to be too costly for the relatively small Canadian firms. Therefore, government subsidies may be necessary to compensate for short-term negative impact of corporate social responsibility on the financial performance of Canadian firms (Makni, Francoeur, & Bellavance, 2009).
Inoue and Lee (2011) research the impact of five different CSR dimensions on the financial performance in tourism related industries (airline, casino, hotel, and restaurant). Those five CSR dimensions are: (1) community relations, (2) diversity, (3) employee relations, (4) environment, and (5) product quality. First of all, the community relation scores are based on whether a company provides help to communities through charitable giving, educational initiatives, and volunteer programs (Inoue & Lee, 2011). Second, diversity scores take into account whether a company integrates diversity into its management and operations (Inoue & Lee, 2011). Third, employee relations scores are rated on the extent to which a company is involved in employee related issues such as their health and safety (Inoue & Lee, 2011). Fourth, environmental issue scores are based on the extent to which a company takes the natural environment (e.g. clean energy) in
9
consideration (Inoue & Lee, 2011). Finally, product quality scores are based on whether a company offers quality and/or innovative products, while ensuring the safety of these products. The study of Inoue and Lee (2011) is based on a sample size of 367 observations from publicly traded U.S. companies in the years 1991 to 2007. All of those five CSR dimensions previously discussed are being used as main independent variables. Inoue and Lee (2011) use two different dependent variables: the return on assets, and Tobin’s Q. They find a significant negative effect of community relations on the return on assets (ROA) of airline companies. The other CSR dimensions have no effect on the ROA of airlines (Inoue & Lee, 2011). Furthermore, Inoue and Lee (2011) find that employee relations and product quality have significant positive effects on Tobin’s Q. They find that the other dimensions of CSR have no significant effect on Tobin’s Q. Inoue and Lee (2011) conclude that each of the five CSR dimensions affects the two different measures of financial performance in the airline industry differently.
Kohers and Simpson (2002) research the link between corporate social performance and the financial performance for the banking industry. As a result of the Community Reinvestment Act of 1977 (CRA) in the U.S., it is compulsory for authorities to examine banks by developing a CRA rating. The CRA rating is based on twelve factors concerning communication about the community’s credit needs and the ability to meet these needs, discrimination, and support of community development projects (Kohers & Simpson, 2002). Ratings are divided into four categories: (1) outstanding, (2) satisfactory, (3) needs to improve, and (4) substantial non-compliance (Kohers & Simpson, 2002). Kohers and Simpson (2002) find, for 385 commercial banks in the years 1993 and 1994, that there is a significant positive impact of the CRA rating on the return on assets.
In line with the results of Kohers and Simpson (2002), Lee and Park (2010) also find a positive relationship between corporate social responsibility and financial performance. Lee and Park (2010) argue in their paper that corporate social responsibility has a significant positive impact on the value performance of airline companies. Their paper is written on airline companies within the S&P500, Russel 1000, or Russel 2000 in the years 1991 to 2006. They conduct their research including CSR measured by using Kinder, Lydenburg, Domini Reseach & Analytics, Inc. (KLD) data. Their dataset contains rating scores based on 113 CSR-relevant indicators that cove four main categories: environmental, social, governance, and controversial business issues (Lee & Park, 2010). Lee and Park (2010) study not only the linear relationship between CSR and financial
10
performance. They also include the ‘quadratic’ and ‘cubic relationships between CSR and the financial performance1. Lee and Park (2010) state that impacts of CSR may vary according to the
degree of a company’s practice in CSR. However, they find no variations in the impact of CSR on the financial performance: neither the quadratic function nor the cubic function of CSR has a significant impact on value performance. Thus, there is a linear relationship between corporate social responsibility activities and the financial performance of airlines (Lee & Park, 2010). However, these results might be influenced by the limited sample size (only 46 observations).
On the contrary, Brammer and Millington (2008) find a nonlinear relationship between corporate social responsibility and the financial performance of companies. They did research on 537 firms listed in the London Stock Exchange. Brammer and Millington (2008) measure CSR by only one dimension: corporate charitable giving. Their findings suggest that only companies that perform unusually good or unusually bad on social responsibility have significant higher financial performance. This result suggest that companies with a higher financial performance either differentiate themselves by a high amount of charitable giving, or choose to save their resources for other purposes than charitable giving (Brammer & Millington, 2008).
2.4 Conclusions from the literature
Results of different researches seem to be inconsistent at first sight. While Makni, Francoeur, and Bellavance (2009) find a negative impact of corporate social responsibility on the financial performance, both Kohers and Simpson (2002), and Lee and Park (2010) find a positive relationship between CSR and the financial performance. Furthermore, while Brammer and Millington (2008) find a nonlinear relationship between social responsibility and the financial performance, most researchers do find a linear relationship (Inoue & Lee, 2011; Kohers & Simpson, 2002; Lee & Park, 2010; Makni, Francoeur, & Bellavance, 2009).
Inconsistencies in the results can, to some extent, be explained by the use of different financial performance measures. While, the most commonly used measure of financial performance is the return on assets (Inoue & Lee, 2011; Kohers & Simpson, 2002; Lee & Park, 2010; Makni, Francoeur, & Bellavance, 2009), some researchers use the ROE as an alternative
1 To control for a multicollinearity issue, a mean value of CSR is calculated. Subsequently, the difference between the mean
value and CSR is estimated. This difference is raised to the second and third functions to represent quadratic (CSR2) and cubic
(CSR3) relationships (Lee & Park, 2010). Thus, the model Lee and Park (2010) use is:
11
measure of financial performance (Lee & Park, 2010; Makni, Francoeur, and Bellavance, 2009), and others use Tobin’s Q (Inoue & Lee, 2011), or the Sharpe ratio (Brammer & Millington, 2008).
Another explanation for the inconsistent results is the difference in corporate social responsibility measures. KLD rating scores, based on 113 CSR-relevant indicators that cover four main categories: environmental, social, governance, and controversial business issues (Lee & Park, 2010), are extensively used as measure for corporate social responsibility (Inoue & Lee, 2011; Kohers & Simpson, 2002; Lee & Park, 2010). Makni, Francoeur, and Bellavance (2009) also use an independent organization’s CSR performance scores. They use the corporate social performance scores of Michael Jantzi Reaearch Associates Inc. (MJRA), a business partner of the KLD database. On the contrary, Brammer and Millington (2008) use only one dimension to measure CSR: corporate charitable giving.
Furthermore, differences in the results can be explained by different control variables that are included into the model. Size is extensively used as a control variable (Brammer & Millington, 2008; Inoue & Lee, 2011; Kohers & Simpson, 2002; Lee & Park, 2010; Makni, Francoeur, & Bellavance, 2009) and leverage is also extensively used as a control variable (Brammer & Millinton, 2008; Inoue & Lee, 2011; Lee & Park, 2010; Makni, Francoeur, & Bellavance, 2009). Furthermore, some researchers use industry dummies (Brammer & Millington, 2008; Makni, Francoeur, & Bellavance, 2009) or year dummies (Brammer & Millington, 2008; Inoue & Lee, 2011; Lee & Park, 2010).
Finally, differences in the results can be explained by differences in research samples and time periods used for the analysis. For example, the relevance of corporate social responsibility to the corporate financial performance might depend on the sector that companies belong to (Sassen, Hinze, & Hardeck, 2016). Sassen, Hinze, and Hardeck (2016) find that the link between environmental performance and the systematic risk of a company is significantly more negative in environmental sensitive industries. Thus, differences in sectors that are examined in the studies might cause inconsistencies in results. Furthermore, inconsistencies can be explained by the different home countries of the companies included in the analysis. For instance, Sassen, Hinze, and Hardeck (2016) argue that corporate social responsibility is particularly important in Europe. According to Ho, Wang, and Vitell (2012), prior research reveals that European companies are prominent in corporate social performance in comparison to companies in countries outside of Europe. Thus, for European companies the link between CSR and corporate financial performance
12
might be more substantial than for non-European companies. In addition, differences in time periods used for conducting the research might cause differences in results. Changing economic conditions over time might have an impact on the financial performance of firms (Brammer & Millington, 2008; Inoue & Lee, 2011; Sassen, Hinze, & Hardeck, 2016). For instance, the financial crisis in 2008 and 2009 might affect the link between corporate social responsibility and corporate financial performance.
3. Methodology and data
In this section the methodology that is used to analyze the relationship between corporate social responsibility and the financial performance is presented. This section is divided into two subsections. The first subsection includes the hypotheses that are tested. Furthermore, the model that is used to test the hypotheses is given. The dependent-, main independent-, and control variables are explained in the end of this subsection. In the second subsection the used data for this research is described and the descriptive statistics are presented.
3.1 Methodology
Consumers around the world are increasingly involved in environmental, social, and governance issues. Therefore, they are considered to spend more on airlines that consider these CSR factors in their business plan. On the other hand, the implementation of corporate social responsibility activities into the business plans of airlines comes with extra costs. The question is whether benefits outweigh the costs of CSR or not. Besides, the results on the impact of CSR on the financial performance in previous literature is very inconsistent. Because of inconsistencies in previous literature, both positive and negative impacts of corporate social responsibility on the financial performance are examined in this research. For each of the three corporate social responsibility pillars, and for the overall CSR scores, the impact on the financial performance of airline companies are tested. Therefore, the following hypotheses are tested in this research:
Hypothesis 1: Corporate social responsibility overall has a positive or negative impact on
13
Hypothesis 2: Environmental-, social-, and/or governance performance have a positive or
negative impact on corporate financial performance of airline companies.
To test for the influence of corporate social responsibility on the corporate financial performance, a multiple regression analysis is performed. The effects of the environmental-, social-, and governance performance and of the overall CSR performance on the return on assets are analyzed. Control variables are added to control for their effects on the corporate financial performance. In line with the research of Sassen, Hinze, and Hardeck (2016) supplemented with components from the model used by Inoue and Lee (2011) and extended with a dummy variable for whether an airline is a low cost carrier or not, the model reads as follows:
ROA (ROE) = β0 + β1CSR + β2ENV + β3SOC+ β4GOV+ β5SIZE + β6LEV + β7BUD + β8- 17YEAR05-14
This model is used to test whether the hypotheses are true. The first hypothesis is true if β1 is not
equal to zero. To test for this hypothesis, the variables for environmental- (ENV), social- (SOC), and governance performance (GOV) are left out of the model because of multicollinearity issues. The second hypothesis is true if β2 and/or β3 and/or β4 are not equal to zero. To test for the second
hypothesis, the variable for the overall CSR performance (CSR) is left out because of multicollinearity issues.
As section 2 points out, most studies measure corporate financial performance by using the return on assets (ROA) (Berman, Wicks, & Jones, 1999; Clemens, 2006; Dooley & Lerner, 1994; Inoue & Lee, 2011; Kang et al., 2010; Kohers & Simpson, 2002; Lee & Park, 2010; Makni, Francoeur, & Bellavance, 2009; Waddock & Graves, 1997). The return on assets is an accounting based measure that captures short-term profitability (Inoue & Lee, 2011). In this paper return on assets is used as accounting based measure for the financial performance. The return on assets is collected from the COMPUSTAT database. This database contains data on the total turnover and on the total assets of companies. The return on assets is computed by the total turnover divided by the total assets of a company.
14
An alternative accounting based measure that is used in previous researches, is the return on equity (ROE) (Choi & Jung, 2008; Dunn & Sainty, 2009; Lee & Park, 2010; Makni,
Francoeur, and Bellavance, 2009). In the robustness check in subsection 4.2 the return on equity is used as measure for the financial performance. Data on the return on equity is collected from the COMPUSTAT database. The return on equity is calculated by the total turnover divided by the common equity of a company.
In line with the research of Ferrel, Liang, and Renneboog (2016) and Sassen, Hinze, and Hardeck (2016), corporate social responsibility is measured using the ASSET4 data from Thomson Reuters. According to Ioannou and Serafeim (2012), Thomson Reuters is an independent rating organization specialized in providing objective and relevant information concerning corporate social responsibility. Specialized researchers from Thomson Reuters use publicly available information from company CSR reports, company websites, annual reports, and news providers to compute CSR scores (Sassen, Hinze, & Hardeck, 2016). ASSET4 was founded in 2003, and acquired by Thomson Reuters in 2009 (Hassel & Semenova, 2015).
ASSET4 classifies the CSR data into different categories within three major pillars: environmental, social, and governance (ESG) (Thomson Reuters Corporate Responsibility Ratings Methodology, 2013). The environmental pillar (ENV) consists of three categories: emission reduction, product innovation, and resource reduction. The social pillar (SOC) has seven categories: community, diversity, employment quality, health and safety, human rights, product responsibility, and training and development. Finally, the governance pillar (GOV) consists of five categories: board functions, board structure, compensation policy, shareholders policy, and vision and strategy. Based upon data in these categories, ASSET4 ratings are derived by comparing companies for 226 performance indicators in total (Thomson Reuters Corporate Responsibility Methodology, 2013). For each of the three ESG pillars, a rating from 0 (bad performance) to 100 (good performance) is composed.
Ioannou and Serafeim (2012) state that it is estimated that investors representing over 2.5 trillion euro of assets, use the ASSET4 data while analyzing their investments. Furthermore, the ASSET4 ratings are used as a measure for corporate social responsibility in several researches (Endrikat, et al., 2015; Ferrel, Liang, & Renneboog, 2016; Halbritter & Dorfleitner, 2015). Therefore, in this study corporate social responsibility is also measured by using ASSET4 ratings.
15
For each of the three pillars the impact on the corporate financial performance is measured. Furthermore, in line with the research of Sassen, Hinze, and Hardeck (2016) and the research of Halbritter and Dorfleitner (2015), an overall score for corporate social responsibility (CSR) is composed of the equally weighted pillars. The influence of this overall score on the corporate financial performance is also measured.
Based on previous literature, four control variables have been included in the regression analysis: the logarithm of size (SIZE), leverage (LEV), a dummy variable for whether an airline is a low cost carrier or not (BUD), and dummy variables for the years 2005 to 2014 (YEAR05-14). These control
variables are included to control for their possible effects on the relationship between corporate social responsibility and corporate financial performance.
First of all, according to Luo and Bhattacharya (2006), the size of the company might have a significant effect on the link between corporate social responsibility and the financial performance of companies. Inoue and Lee (2011) state that large firms usually implement more CSR initiatives compared to small firms. Furthermore, several studies indicate that the size of the firm has a significant impact on the corporate financial performance (Hillman & Keim, 2001; Waddock & Graves, 1997). Consistent with previous research (McWilliams & Siegel, 2000; Sassen, Hinze, & Hardeck, 2016; Simpson & Kohers, 2002), the company size (SIZE) is calculated as the natural logarithm of total assets in US dollars. The data on total assets are retrieved from COMPUSTAT.
Second, leverage is included in the regression analysis. Leverage is included to control for the effect that the capital structure of a company has on the relationship between CSR and the financial performance. Waddock and Graves (1997) argue that companies which are highly risk tolerant (firms with high leverage) might act differently concerning corporate social responsibility than companies which are not as risk tolerant. This difference arises from the different levels of risk that are associated with corporate social responsibility. Furthermore, Waddock and Graves (1997) demonstrate that a high level of leverage has a negative effect on the corporate financial performance. Coherent to previous research, leverage (LEV) is measured by the total debt of the company divided by its total assets (Inoue & Lee, 2011; McWilliams & Siegel, 2000; Waddock & Graves, 1997). The data on total debt and total assets are retrieved from the COMPUSTAT database.
16
Furthermore, a dummy variable for whether an airline is a low cost carrier or not (BUD), is included into the regression. For this variable, an observation is equal to 1 for low cost carrier airlines. An observation is equal to 0 for full service airlines. Data on whether an airline is a low cost carrier or not is retrieved from the airline companies’ website.
Finally, Brammer and Millington (2008) find that the intensity of the relationship between corporate social responsibility and the financial performance might fluctuate over time. To correct for year-specific impacts, previous literature included a set of year dummy variables (YEAR) into the regression (Brammer & Millington, 2008; Inoue & Lee, 2011). Dummy variables are included for the years 2005 to 2014. An observation for this variables is equal to 1 for a focal year, and 0 for the alternative years.
3.2 Data and descriptive statistics
Financial data (total turnover, total assets, common equity, and total debt) has been retrieved from COMPUSTAT database. The information on CSR is obtained from the ASSET4 database. This database includes 41 airline companies. For these airlines, data for the years 2005 until 2015 has been retrieved. The final sample consists of 233 observations. The sample exists of an unbalanced data set, as for some years data for some airline companies was missing.
To correct for outliers, observations outside of the 99.7% confidence interval were deleted. This method for omitting outliers is consistent with the research methods of previous researches (Inoue & Lee, 2011; Tabachnick & Fidell, 2006).
The summary of descriptive statistics for the airlines included in the analysis are presented in table 1. This table includes the number of observations (N), the mean for every variable, the standard deviation (SD), and the minimum and maximum value.
17
Table 1 Summary of descriptive statistics
ROA is measured as total turnover divided by total assets; ROE is measured as total turnover divided by common equity; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
Variable N Mean SD Min Max
ROA 233 0.742 0.213 0.209 1.338 ROE 214 3.381 4.293 -19.418 35.401 ENV 233 61.117 24.438 9.690 94.740 SOC 233 61.712 24.459 7.990 97.030 GOV 233 52.231 28.664 2.790 95.710 CSR 233 58.353 19.138 15.347 91.060 SIZE 233 9.372 0.875 7.017 10.899 LEV 233 0.730 0.186 0.090 1.416 BUD 233 0.262 0.441 0.000 1.000
Source: COMPUSTAT, ASSET4, airline company websites
In table 2, the results of the correlation analysis are presented. The correlation analysis reveals that return on assets and return on equity are significant positively correlated. Thus, companies with a higher return on assets usually have a higher return on equity. Furthermore, the correlation analysis results indicate that environmental, social, and governance performance all have a significant positive correlation with return on assets. Moreover, the overall CSR score also has a significant positive correlation with the return on assets. Thus, firms that implement more CSR activities into their business policies, usually perform better than firms with less consideration of CSR in their business policies. Leverage also has a significant positive correlation with the return on assets. This implies that the companies that are more risk tolerant, usually have a higher return on assets. On the contrary, size and budget don’t have any significant correlation with the return on assets.
18
Table 2 Summary of correlation coefficients
ROA is measured as total turnover divided by total assets; ROE is measured as total turnover divided by common equity; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
ROA ROE ENV SOC GOV CSR SIZE LEV BUD
ROA 1.000 ROE 0.271*** 1.000 ENV 0.358*** 0.173** 1.000 SOC 0.267*** 0.092 0.774*** 1.000 GOV 0.279*** 0.131* 0.068 0.185*** 1.000 CSR 0.405*** 0.180*** 0.789*** 0.848*** 0.607*** 1.000 SIZE 0.082 0.299*** 0.564*** 0.472*** 0.093 0.488*** 1.000 LEV 0.409*** 0.472*** 0.295*** 0.307*** 0.191*** 0.352*** 0.328*** 1.000 BUD -0.100 -0.199*** -0.344*** -0.265*** 0.275*** -0.122* -0.427*** -0.240*** 1.000
*, ** and *** represent significance levels of 0.10, 0.05 and 0.01 respectively. Source: COMPUSTAT, ASSET4, airline company websites
However, size does have a significant positive correlation with the environmental- and social performance, and with the overall CSR score. This indicates that larges companies usually implement more CSR (environmental and social) activities than smaller companies do. Furthermore, leverage has a significant positive relation with the environmental-, social-, and governance pillars and with the overall CSR score. This is consistent with previous findings from Waddock and Graves (1997). They mention that companies with a higher leverage ratio usually have a better corporate social responsibility performance than companies with a lower leverage ratio. Finally, budget has a significant negative correlation with the environmental and social pillar, and the overall CSR score. This implies that low cost carrier airlines implement less environmentally and socially responsible strategies than full service airlines do. However, budget
19
is significantly positive correlated with the governance pillar. Thus, low cost carriers are more governance responsible compared to full service airlines.
4. Analysis
In this section, the data is analyzed through an ordinary least squares (OLS) regression. This section is divided into two subsections. In the first subsection, the main results of the regression analysis are presented and discussed. The second section contains a robustness check. In this robustness check the return on equity is used as an alternative measure for the financial performance.
4.1 Empirical results
In table 3 the OLS regression results with robust standard errors are presented. Note that in this table the year dummy variables are not included, while they were included in the regression analysis of models 1 and 2. The extended version of table 3, including year dummy variables, is included in the appendix 1. Tabel 3 includes six different models that have been used to determine the impact of corporate social responsibility on the financial performance of airlines.
In the first model each of the three CSR pillars (social, environmental, and governance) is included. Also, the control variables size, leverage, the low cost carrier dummy, and year dummies are included. This analysis reveals that the independent variables in this model explain the variation in the return on assets significantly (R2 = 0.368). The environmental- and governance
performance both have a significant positive effect on the return on assets. Thus, the more environmental- and governance responsible airlines are, the higher their return on assets is. In contrast, social responsibility has insignificant effect on the return on assets. Among the control variables, size has a significant negative effect on the return on assets. Thus, airlines with a higher amount of assets tend to have a lower return on assets. This result is in line with the results of previous research (Inoue & Lee, 2011). Leverage has a significant positive effect on the return on assets. This indicates that airlines with a higher total debt to total asset ratio usually have a higher return on assets. This result is in contrast with previous research (Capon, Farley, & Hoening, 1990; Waddock & Graves, 1997). Budget doesn’t have a significant impact on the return on assets. Thus, the return on assets is not significantly influenced by whether an airline is a low cost carrier or not.
20
Furthermore, all of the year dummy variables don’t have a significant effect on the return on assets. Therefore, in model 3 to 6, the year dummy variables are omitted.
In the second model the impact of the overall CSR on the financial performance is tested. The independent variables in this model explain the variation in return on assets significantly (R2
= 0.327). In line with previous research (Kohers & Simpson, 2002; Lee & Park, 2010), this analysis reveals that CSR has a significant positive effect on return on assets. This means that if the corporate social responsibility of an airline increases, the return on assets increases as well. Furthermore, in line with research of Inoue and Lee (2011), size has a significant negative impact on the return on equity. Thus, larger airline companies tend to have a lower return on assets than smaller airline companies. In contrast with previous research (Capon, Farley, & Hoening, 1990; Waddock & Graves, 1997) leverage has a significant positive impact on the return on assets. This indicates that airline companies with higher total debt to total asset ratios, usually have a higher return on assets. Whether an airline is a low cost carrier or not does not make a difference for the return on assets, since budget has an insignificant impact on the return on assets. Consistent with model 1, all of the year dummy variables don’t have a significant effect on the return on assets. Therefore, as stated before, the year dummy variables are omitted in models 3 to 6.
In the third- and fourth model, the year dummy variables are omitted. The independent variables in this model still explain the variation in the return on assets significantly (R2 = 0.332
and R2 = 0.287 respectively). Furthermore, these models indicate the same significant impacts of
the CSR pillars and overall CSR on the financial performance as models 1 and 2 do.
The budget dummy variable isn’t included into the models of previous written literature. Therefore, in the fifth and sixth model, the budget dummy variable is omitted. The variation in the return on assets is still significantly explained by the independent variables (R2 = 0.327 and R2 =
0.282). These models indicate the same significant impacts of the CSR pillars and overall CSR on the financial performance as the previous models (1 to 4).
21
Table 3 Summary of the OLS regression analysis, ROA as independent variable
ROA is measured as total turnover divided by total assets; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
(1) (2) (3) (4) (5) (6) Intercept 0.922 (5.00)*** 0.867 (4.13)*** 0.871 (5.01)*** 0.805 (4.01)*** 0.787 (5.50)*** 0.709 (4.62)*** ENV 0.004 (4.27)*** 0.004 (4.45)*** 0.004 (4.68)*** SOC -0.001 (-1.34) -0.001 (-1.56) -0.001 (-1.50) GOV 0.002 (4.39)*** 0.002 (4.16)*** 0.002 (4.27)*** CSR 0.005 (6.53)*** 0.005 (6.19)*** 0.004 (6.36)*** SIZE -0.081 (-3.98)*** -0.073 (-3.05)*** -0.073 (-3.56)*** -0.063 (-2.69)*** -0.066 (-3.67)*** -0.054 (-2.86) *** LEV 0.402 (6.63)*** 0.402 (6.29)*** 0.384 (6.27)*** 0.380 (5.79)*** 0.400 (6.44)*** 0.394 (5.97)*** BUD -0.049 (-1.28) -0.045 (-1.28) -0.042 (-1.16) -0.039 (-1.15) R2 0.368 0.327 0.332 0.287 0.327 0.282
*, ** and *** represent significance levels of 0.10, 0.05 and 0.01 respectively.
The t-values are provided in parenthesis underneath the coefficient estimates. The t-values are based on robust standard errors. Source: COMPUSTAT, ASSET4, airline company websites
22
4.2 Robustness check
In order to test for robustness of the results in this research, an additional analysis is conducted. Although measuring the financial performance of companies using the return on assets is consistent with previous literature (Berman, Wicks, & Jones, 1999; Clemens, 2006; Dooley & Lerner, 1994; Inoue & Lee, 2011; Kang et al., 2010; Waddock & Graves, 1997), some researchers measure financial performance using the return on equity (ROE) instead (Choi & Jung, 2008; Dunn & Sainty, 2009; Lee & Park, 2010; Makni, Francoeur, and Bellavance, 2009). As stated before, ROE is calculated as total net turnover divided by common equity.
In table 2, the results of the correlation analysis are presented. The correlation analysis reveals that return on assets and return on equity are significant positively correlated. Thus, companies with a higher return on assets usually have a higher return on equity. This makes sense, since both ROA and ROE are short-term accounting based measures for the financial performance. Furthermore, the results indicate that return on equity is positively correlated with the environmental pillar on the 5 percent significance level, and also with the governance pillar on the 10 percent level. Moreover, the return on equity is positively correlated with the overall CSR score on the 1 percent significance level. Thus, airline companies with a better CSR performance usually have a higher return on equity. Size and leverage are both positively correlated with the return on equity. Thus, larger companies and companies with higher leverage usually have a higher return on equity. Finally, budget is negatively correlated with the return on equity. This indicates that full service airlines usually have a higher return on equity than low cost carriers.
In table 4 the OLS regression results for the robustness check are presented. Note that in this table the year dummy variables are not included, while they were included in the regression analysis of models 1 and 2. The extended version of table 4, including year dummy variables, is included in appendix 2. Table 4 includes six different models that have been used to compare the impact of CSR on return on assets with the impact of CSR on return on equity. The return on equity (the dependent variable) is explained for a significant part by the independent variables in these six models (R2 equals 0.343, 0.267, 0.329, 0.253, 0.306, and 0.250 respectively for each model).
The analysis indicates that the environmental pillar of CSR has no significant impact on the return on equity. This result is in contrast with the results of the main analysis with ROA as
23
dependent variable. In the main analysis the environmental performance positively influences the corporate financial performance (ROA).
Table 4 Summary of the OLS regression analysis, ROE as dependent variable
ROE is measured as total turnover divided by common equity; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
(1) (2) (3) (4) (5) (6) Intercept -13.589 (-2.80)*** -12.201 (-2.54)** -13.064 (-2.78)*** -11.908 (-2.57)** -16.145 (-3.12)*** -13.434 (-3.10)*** ENV 0.021 (0.79) 0.019 (0.73) 0.023 (0.82) SOC -0.056 (-2.13)** -0.055 (-2.16)** -0.053 (-2.04)** GOV 0.038 (3.03)*** 0.038 (2.98)*** 0.027 (2.76)*** CSR -0.002 (-0.14) -0.002 (-0.20) -0.005 (-0.39) SIZE 0.733 (1.78)* 0.696 (1.61) 0.807 (1.91)* 0.743 (1.75)* 1.102 (2.26)** 0.891 (1.99)** LEV 14.662 (3.77)*** 13.058 (3.33)*** 14.040 (3.66)*** 12.541 (3.19)*** 14.090 (3.49)*** 12.691 (3.28)*** BUD -1.748 (-2.48)** -0.610 (-1.09) -1.734 (-2.43)** -0.615 (-1.09) R2 0.343 0.267 0.329 0.253 0.306 0.250
*, ** and *** represent significance levels of 0.10, 0.05 and 0.01 respectively.
The t-values are provided in parenthesis underneath the coefficient estimates. The t-values are based on robust standard errors. Source: COMPUSTAT, ASSET4, airline company websites
24
Furthermore, this analysis reveals that the social performance has a significant negative impact on the return on equity. This result is also in contrast with the results of the main analysis, where the social performance has no significant impact on the financial performance. The governance performance has a significant positive impact on the return on equity. This result is in line with the results of the main analysis. Thus, governance performance positively influences both the return on assets and the return on equity. In contradiction to the main analysis results, the overall CSR performance doesn’t significantly affect the return on equity. Among the control variables, size has a significant positive impact on the return on equity in models 1, 3, 4, 5, and 6. This is in contrast with the results of the main analysis, since size has a significant negative impact on return on assets. Corresponding with the results of the main analysis, leverage has a significant positive impact on the return on equity. Thus, firms with a higher total debt to total assets ratio tend to have a higher financial performance. Finally, in the models 1 and 3, the budget dummy variable has a significant negative impact on the return on equity. This indicates that full service airlines usually have a high return on equity compared to low cost carriers.
This robustness check validates the result of the main analysis that higher governance performance has a positive impact on corporate financial performance. Furthermore, the robustness check confirms the positive impact of leverage on the corporate financial performance of airline companies. However, several results of this robustness check differ from the main analysis with the return on assets as dependent variable. Based on these differences, it can be concluded that the relationship between corporate social responsibility and the corporate financial performance may vary even among accounting based measures of financial performance. Despite the fact that both ROA and ROE are short-term accounting based measures, they capture different financial features of profitability. The return on assets measures the profitability of companies in terms of asset usage, while the return on equity measures the profitability regarding the efficiency of using shareholders’ capital (Palepu & Healy, 2008).
5. Conclusion
This study examines the effects of corporate social responsibility on the financial performance of airline companies. This study is based on 41 airline companies for the years 2005 to 2015, resulting in 233 observations since for some years for some airline companies data was missing. In this study, corporate social responsibility is disaggregated in three categories: environmental issues,
25
social issues, and governance issues. The effects of each of the three categories, and of the overall corporate social responsibility on the financial performance is analyzed. The corporate financial performance of airlines was measured by the return on assets and by the return on equity.
The evidence from the analysis in this paper indicates that higher return on assets is associated with adoption of environmental- and governance responsible activities into the business strategy. In addition, the overall corporate social responsibility performance has a positive impact on the return on assets. Thus, airline companies that implement corporate social responsibility activities, in particular environmental- and governance responsible activities, usually have a better financial performance. These acquired results are in line with the results of Kohers and Simpson (2002), and with the results of Lee and Park (2010). Moreover, the results of this study confirm the second hypothesis: environmental-, social-, and/or governance performance have an impact on the corporate financial performance of airline companies. Furthermore, the acquired results confirm the first hypothesis, which states that corporate social responsibility overall has an impact on the corporate financial performance of airline companies.
The analysis with the return on equity as dependent variable confirms that airline companies with a better governance performance, usually have a better financial performance. However, this analysis reveals that the environmental performance has no impact on the return on equity and the social performance has a negative impact on the return on equity. Moreover, the overall CSR performance isn’t significantly related to the return on equity. Thus, it can be concluded that the relationship between corporate social responsibility and corporate financial performance may vary even among accounting based measures of financial performance.
This research entails several limitations. First of all, the sample size of the current study is relatively small because the ASSET4 database contains only 41 airline companies worldwide. A replication study with a larger sample size might provide different findings. Furthermore, for this study airline companies from different countries from all over the world are included in the analysis. However, the link between corporate social responsibility and corporate financial performance may differ by home country of the airline companies (Ho, Wang, Vitell, 2012). Therefore, a suggestion for further research is to implement dummy variables to control for differences between different countries or different economic regions. Another limitation of this study is the fact that only the impact of corporate social responsibility itself on the financial
26
performance is examined. However, the customers’ perception of CSR implementation by airline companies might be an equally substantial factor influencing the financial performance of airline companies. If an airline company implements many CSR activities but doesn’t communicate this to shareholders and customers, then the benefits of a good CSR performance might decrease. Therefore, in future research the customers’ perception of CSR implementation by airline companies could be measured through questionnaires. The results of these questionnaires could be included in the regression analysis.
Despite the limitations of this study, the implications of the findings in this study might be valuable for airline companies. The main findings of this research suggest that airline executives should consider implementing (more) corporate social responsibility activities to increase the financial performance. Especially increasing the environmental- and governance performance could have a positive impact on the return on assets. Furthermore, by providing practical knowledge which could help compose more profitable airline industry investment portfolios, the findings of this study could be valuable for financial analysts and investors. Especially airlines that are improving their governance performance could be interesting for investors, since improvement of governance performance usually has a positive impact on the return on equity.
27
References
Air France-KLM. (2016). Corporate Social Responsibility Report 2015. Retrieved from http://www.airfranceklm.com/en/sustainable-development-2015-report
Arvidsson, S. (2010). Communication of corporate social responsibility: a study of the views of management teams in large companies. Journal of Business Ethics, 96(3), 339-354 Becken, S. (2007). Tourists’ perception of international air travel’s impact on the global climate
and potential climate change. Journal of Sustainable Tourism, 15(4), 351-368
Berman, S. L., Wicks, A. C., & Jones, T. M. (1999). Does stakeholder orientation matter? The relationship between stakeholder management models and firm financial performance. Academy of Management Journal, 42(5), 488-506
Bowen, H. R. (1953). Social responsibilities of the businessman. New York: Harper & Row Brammer, S., & Millington, A. (2008). Does it pay to be different? An analysis of the
relationship between corporate social and financial performance. Strategic Management Journal, 29(12), 1325-1343
Capon, N., Farley, J. U., & Hoening, S. (1990). Determinants of financial performance: a meta- analysis. Management Science, 36(10), 1143-1159
Carroll, A. B. (1979). A three-dimensional conceptual model of corporate social performance. Academy of Management Review, 4(4), 497-505
Choi, T. H., & Jung, J. (2008). Ethical commitment, financial performance, and valuation: an empirical investigation of Korean companies. Journal of Business Ethics, 81(2), 447-463 Clemens, B. (2006). Economic incentives and small firms: does it pay to be green? Journal of
Business Research, 59(4), 492-500
Cowper-Smith, A. & Grosbois, D. (2011). The adoption of corporate social responsibility practices in the airline industry. Journal of Sustainable Tourism, 19(1), 59-77
Dando, N., & Swift, T. (2003). Transparency and assurance minding the credibility gap. Journal of Business Ethics, 44(2), 195-200
Dooley, R. S., & Lerner, L. D. (1994). Pollution, profits, and stake-holders: the constraining effect of economic performance on CEO concern with stakeholder expectations. Journal of Business Ethics, 13(9), 701-711
Dowell, G., Hart, S., & Yeung, B. (2000). Do Corporate Global Environmental Standards Create or Destroy Market Value? Management Science, 46(8), 1059-1074
28
Dunn, P., & Sainty, B. (2009). The relationship among board of director characteristics,
corporate social performance and corporate financial performance. International Journal of Managerial Finance, 5(4), 407-423
Endrikat, J., Guenther, E., Trumpp, C., & Zopf, C. (2015). Definition, conceptualization, and measurement of Corporate Environmental Performance: a critical examination of a multidimensional construct. Journal of Business Ethics, 126(2), 185-204
Ferrel, A., Liang, H., & Renneboog, L. (2016). Socially responsible firms. Journal of Financial Economics, 122(3), 585-606
Habisch, A., Jonker, J., Wegner, M., & Schmidpeter, R. (2005). Corporate social responsibility across Europe. Berlin: Springer
Halbritter, G., & Dorfleitner, G. (2015). The wages of social responsibility – where are they? A critical review of ESG investing. Review of Financial Economics, 26(11), 25-35
Hassel, L. G., & Semenova, N. (2015). On the validity of environmental performance metrics. Journal of Business Ethics, 132(2), 249-258
Hillman, A. J., & Keim, G. D. (2001). Shareholder value, stakeholder management, and social issues: what’s the bottom line? Strategic Management Journal, 22(2), 125-139
Ho, F. N., Wang, H. M. D., & Vitell, S. J. (2012). A global analysis of corporate social performance: the effects of cultural and geographic environments. Journal of Business Ethics, 107(4), 423-433
Idowu, S. O., & Louche, C. (2011). Theory and Practice of Corporate Social Responsibility. Berlin: Springer-Verlag Berlin Heidelberg
Igalens, J., & Gond, J. (2005). Measuring corporate social performance in France: a critical and empirical analysis of ARESE data. Journal of Business Ethics, 56(2), 131-148
Inoue, Y., & Lee, S. (2011). Effects of different dimensions of corporate social responsibility on corporate financial performance in tourism-related industries. Tourism Management, 32(4), 790-804
Ioannou, I. & Serafeim, G. (2012). What drives corporate social performance? The role of national-level institutions. Journal of International Business Studies, 43(9), 834-864 Kang, K. H., Lee. S., Huh, C. (2010). Impacts of positive and negative corporate social
responsibility activities on company performance in the tourism industry. International Journal of Hospitality Management, 29(1), 72-82
29 Kohers, T. & Simpson, W. (2002). The link between Corporate Social and Financial
Performance: evidence from the banking industry. Journal of Business Ethics, 35(2), 97-109
Kuo, T. C., Kremer, G. E. O., Phuong, N. T., & Hsu, C. (2016). Motivations and barriers for corporate social responsibility reporting: evidence from the airline industry. Journal of Air Transport Management, 57, 184-195
Lee, S., & Park, S. (2010). Financial impacts of socially responsible activities on airline companies. Journal of Hospitality & Tourism Research, 34(2), 185-203
Luo, X., & Bhattacharya, C. B. (2006). Corporate Social Responsibility, customer satisfaction, and market value. Journal of Marketing, 70(4), 1-18
Makni, R., Francoeur, C., & Bellavance, F. (2009). Causality between Corporate Social Performance and Financial Performance: evidence from Canadian firms. Journal of Business Ethics, 89(3), 409-422
Matten, D., & Moon, J. (2008). “Implicit” and “explicit” CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review, 33(2), 404-424
McGuire, J. B., Sundgren, A., & Schneeweis, T. (1988). Corporate Social Responsibility and Firm Financial Performance. Academy of Management Journal, 31(4), 854-872 McWilliams, A., & Siegel, D. (2001). Corporate social responsibility: a theory of the firm
perspective. Academy of Management Review, 26(1), 117-127
Palepu, K. G., & Healy, P. M. (2008). Business analysis & valuation: Using financial statements. Mason, OH: Thompson Higher Education
Sassen, R., Hinze, A., & Hardeck, I. (2016). Impact of ESG factors on firm risk in Europe. Journal of Business Economics, 86(8), 867-904
Soloman, R., & Hansen, K. (1985). It’s good business. New York: Atheneum
Tabachnick, B. G., & Fidell, L. S. (2006). Using multivariate statistics. Boston: Allyn & Bacon Thomson Reuters Corporate Responsibility Ratings Methodology. (2013). Retrieved January 8,
2017, from http://financial.thomsonreuters.com/content/dam/openweb/documents/pdf/tr-com-financial/methodology/corporate-responsibility-ratings.pdf
Tsoutsoura, M. (2004). Corporate Social Responsibility and Financial Performance (Working Paper Series, Paper 7). Berkeley, CA: Center for Responsible Business.
30
Vance, S. (1975). Are socially responsible firms good investment risks? Management Review, 64(8), 18-24
Waddock, S. A., & Graves, S. B. (1997). The corporate social performance-financial performance link. Strategic Management Journal, 18(4), 303-310
31
Appendices
Appendix 1 Summary of the OLS regression analysis; ROA as independent variable (including year dummy variables)
ROA is measured as total turnover divided by total assets; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
(1) (2) (3) (4) (5) (6) Intercept 0.922 (5.00)*** 0.867 (4.13)*** 0.871 (5.01)*** 0.805 (4.01)*** 0.787 (5.50)*** 0.709 (4.62)*** ENV 0.004 (4.27)*** 0.004 (4.45)*** 0.004 (4.68)*** SOC -0.001 (-1.34) -0.001 (-1.56) -0.001 (-1.50) GOV 0.002 (4.39)*** 0.002 (4.16)*** 0.002 (4.27)*** CSR 0.005 (6.53)*** 0.005 (6.19)*** 0.004 (6.36)*** SIZE -0.081 (-3.98)*** -0.073 (-3.05)*** -0.073 (-3.56)*** -0.063 (-2.69)*** -0.066 (-3.67)*** -0.054 (-2.86) *** LEV 0.402 (6.63)*** 0.402 (6.29)*** 0.384 (6.27)*** 0.380 (5.79)*** 0.400 (6.44)*** 0.394 (5.97) BUD -0.049 (-1.28) -0.045 (-1.28) -0.042 (-1.16) -0.039 (-1.15) YEAR05 0.022 (0.34) 0.029 (0.47) YEAR06 0.118 (1.38) 0.110 (1.49)
32 Appendix 1 (continued) (1) (2) (3) (4) (5) (6) YEAR07 0.069 (1.15) 0.070 (1.10) YEAR08 0.084 (1.45) 0.071 (1.18) YEAR09 -0.046 (-0.93) -0.068 (-1.33) YEAR10 -0.038 (-0.75) -0.048 (-0.92) YEAR11 -0.008 (-0.15) -0.021 (-0.42) YEAR12 0.006 (0.10) -0.000 (-0.01) YEAR13 0.007 (0.12) -0.003 (-0.05) YEAR14 0.012 (0.20) -0.003 (-0.04) R2 0.368 0.327 0.332 0.287 0.327 0.282
*, ** and *** represent significance levels of 0.10, 0.05 and 0.01 respectively.
The t-values are provided in parenthesis underneath the coefficient estimates. The t-values are based on robust standard errors. Source: COMPUSTAT, ASSET4, airline company websites
33
Appendix 2 Summary of the OLS regression analysis; ROE as independent variable (including year dummy variables)
ROE is measured as total turnover divided by common equity; ENV represents the environmental pillar of CSR, which is measured by the environmental score in the ASSET4 database; SOC represents the social pillar of CSR, which is measured by the environmental score in the ASSET4 database; GOV represents the governance pillar of CSR, which is measured by the governance score in the ASSET4 database; CSR represents the overall CSR score, which is measured by an equally-weighted average of the three CSR pillars from the ASSET4 database; SIZE represents the size of a company and is measured by the natural logarithm of the total assets; LEV represents the capital structure of a company and is measured by total debt divided by total assets; BUD is equal to 1 for low cost carriers, and equal to 0 for traditional full service airlines.
(1) (2) (3) (4) (5) (6) Intercept -13.589 (-2.80)*** -12.201 (-2.54)** -13.064 (-2.78)*** -11.908 (-2.57)** -16.145 (-3.12)*** -13.434 (-3.10)*** ENV 0.021 (0.79) 0.019 (0.73) 0.023 (0.82) SOC -0.056 (-2.13)** -0.055 (-2.16)** -0.053 (-2.04)** GOV 0.038 (3.03)*** 0.038 (2.98)*** 0.027 (2.76)*** CSR -0.002 (-0.14) -0.002 (-0.20) -0.005 (-0.39) SIZE 0.733 (1.78)* 0.696 (1.61) 0.807 (1.91)* 0.743 (1.75)* 1.102 (2.26)** 0.891 (1.99)** LEV 14.662 (3.77)*** 13.058 (3.33)*** 14.040 (3.66)*** 12.541 (3.19)*** 14.090 (3.49)*** 12.691 (3.28)*** BUD -1.748 (-2.48)** -0.610 (-1.09) -1.734 (-2.43)** -0.615 (-1.09) YEAR05 0.925 (1.06) 0.839 (1.28) YEAR06 1.446 (1.14) 0.988 (0.97) YEAR07 0.627 (0.92) 0.149 (0.24)
