The market acceptance of corporate social responsibility: a comparison across six countries/regions

Abstract

Information on the link between market performance and corporate social responsibility (CSR) activities provides an indication of the extent of acceptance by investors of these types of activities. The nature of this relationship is of critical importance for management trying to reconcile the demands of the company’s shareholders with those of a much wider group of stakeholders and for investors pursuing a socially responsible investing strategy. Using an international database we investigate the extent to which expenditures on CSR activities are valued across market in six countries/regions. We find that CSR activities are highly valued by the investors in the European markets, where our findings clearly indicate that such activities lead to higher market valuations. In the US, Japan and Australia expenditures on CSR activities have a neutral impact on company valuation, which is still a good outcome for management who wish to incorporate into their decision process the objectives of a wide spectrum of stakeholders and for investors wishing to tilt their investments towards the more socially responsible companies.

Keywords

International market reaction responsible management socially responsible investing stakeholders

1. Introduction

Corporate social responsibility (CSR) is becoming an increasingly more important issue for companies with a growing emphasis on it by governments, investors and, indeed, within the companies themselves.¹ A key question that is of interest to management and investors alike is the extent to which the market reacts to expenditures on CSR activities. Management is torn between satisfying the interest of shareholders and those of other stakeholders. Investors are concerned with the stock market performance of companies that have strong CSR records. Both will be satisfied as long as there is not a negative relationship between the extent of CSR activities and future stock market performance, as this means that companies can be good corporate citizens without compromising their shareholders and investors can support these same companies without sacrificing investment returns.

In this paper we utilize an extensive data set provided by Sustainable Asset Management (SAM)² to examine the market response to CSR activities in six countries/regions: the US, the UK, Central and Southern Europe (henceforth C&S Europe),³ the Nordic Countries (excluding Iceland), Japan and Australia. We find evidence to suggest that CSR activities are much more highly valued within the European markets than they are outside of these markets.

In Section 2 of the paper, we set out the motivation for the paper in more detail and canvass the literature relevant to the market impact of CSR activities, especially where it is available on a cross-border basis. We proceed in Section 3 to describe our method and the data that we utilize. The findings of our research and their implications are presented in Section 4, with a summary of our major findings being provided in Section 5.

2. Background

The focus of this paper is on the market’s perception of company’s CSR activities as indicated by how the company’s market performance responds to changes in the extent of these activities. The metric that we use to measure market response is the returns realized by shareholders with a positive (negative) relationship between these returns and the changes in CSR activities, providing evidence that the market recognizes such expenditures are beneficial (detrimental) to the value of the firm. In this study we investigate this relationship across six countries/regions, which will enable us to see whether the receptiveness of markets to these activities varies across national borders. The diversity of these countries/regions is quite wide, as it includes the US, three countries/regions in Europe and two in Asia. However, the fact that a common method for scoring CSR activities is employed in all cases is important, as it enables us to provide superior insights into the cross-border market acceptance of these activities to what has previously been possible.

One question that one must ask is why the markets in the various countries/regions should react in different ways to changes in CSR activities. One possibility that we would suggest, and the one on which we concentrate in this paper, is that these differences would reflect variations in the attitudes held with respect to CSR in the various countries/regions. We would argue that the (positive) impact that CSR activities have on market returns would be greatest in those countries/regions where there is a strong community acceptance of the need for such activities, supported by a strong regulatory framework that then encourages companies to be more conscious of their corporate social responsibilities. In the remainder of this section of the paper, we first survey the evidence from comparative studies on the attitudes towards CSR across countries and regions, before turning our attention to a brief survey of the available evidence on the market reaction to a change in CSR activity.

2.1 Comparative studies

Unfortunately, as pointed out by Williams and Aguilera (2008), there is a paucity of comparative studies of CSR across national borders, which they put down to the fact that CSR is only an emerging area and comparative studies have been hampered by a lack of consistency in the definition and measurement issues relating to CSR. Perhaps the paper that comes closest to the geographical coverage in this paper was written by Welford (2005), who surveyed the policies of large companies in many of the countries/regions included in our work. The relevant countries that he included were the US, the UK, Norway (the sole Nordic country), Germany, France, Italy and Spain (all being part of C&S Europe) and Japan. The survey asked for each company’s policy with respect to 20 elements of CSR covering a wide spectrum of areas, albeit with less focus on environmental issues than one might expect.

In almost all of the areas covered, the European firms would appear to have the most developed policies, typically slightly ahead of the North American companies with the Asian companies lagging someway behind. Within Europe, Norway is standout consistent, with evidence pointing towards this country having the highest commitment to CSR.⁴ The picture across the other European countries is mixed, with companies in some countries commonly having policies on some elements of CSR but not on others (e.g. Italian companies commonly have written policies on ethics but very rarely on human rights). Welford’s evidence would suggest that: (i) the Nordic companies have the most developed policies on CSR; (ii) it would be difficult to differentiate between the level of policy development for companies within the UK and the C&S European countries; but (iii) the European countries are ahead of the US, which in turn is superior to companies in the Asian region.

Doh and Guay (2006) examined how differences in the institutional structures and political legacies influence the support for CSR, particularly from non-government organizations, in Europe relative to the US. They conclude that the social democratic traditions in Europe, as compared to libertarian and individualistic strands of political thought in the US, have resulted in more external pressure on European firms to be supportive of CSR. The culture of the Japanese companies towards CSR is somewhat unique, with there being strong support for CSR activities relating to ‘internal’ stakeholders, such as employees, but little support for CSR activities relating to external stakeholders, such as the general community. The many contradictions in the practice of CSR in Japan confuse the outside observers, generating mixed reactions to ethical policies adopted by the corporations (Wokutch, 1992) and contributing to the relatively low level of commitment to CSR activities observed for Japanese companies.⁵ Anderson and Landau (2006) provide a survey of the level of CSR activities by Australian companies. They conclude that the Australian approach to CSR is somewhat tentative, being largely restricted to short-term initiatives of a philanthropic nature. They found, with few exceptions, that most Australian companies had not taken significant steps towards integrating CSR into their strategic approach or corporate culture.

2.2 CSR and corporate performance

Although there have been well in excess of a 100 studies that have evaluated the relationship between CSR activities and corporate performance, none of them have provided comparative evidence across the spectrum of countries/regions covered in this study. The findings in these studies have been mixed, with some finding a negative relationship between CSR and corporate performance, some no relationship and others a positive relationship. Griffin and Mahon (1997) reviewed 51 papers on this topic and reported that 33 reported a positive relationship, nine reported no relationship and 20 reported a negative relationship. Roman et al. (1999) reworked and extended the Griffin and Mahon paper and categorized 33 studies as demonstrating a positive relationship between CSR and corporate performance, 14 that failed to find any relationship and only five that found a negative relationship.

More recent US studies are still getting mixed results, with Rogers et al. (2008) finding a positive relationship only for firms that also devote significant resources to research and development (R&D) and other intangible assets, whereas Becchitti and Ciriretti (2006) find a negative relationship after controlling for industry effects. In studies using other than US data, no relationship was found between CSR and corporate performance for Australian firms (Brine et al., 2009) and Brazilian firms (Fiori et al., 2007), while a negative relationship was found for the UK (Brammer et al., 2006) and Brazilian companies (Crisostomo et al., 2010). On balance, the majority of studies find a positive relationship, suggesting that markets embrace companies that devote resources to CSR activities but to date there is no empirical closure on this matter. The diversity of findings that continue to this day reflect the following: (i) the lack of a strong conceptual foundation; (ii) the lack of appropriate measures of the extent of a company’s CSR activities; (iii) the lack of a sufficient sample size; and (iv) the lack of methodological rigour (Bird et al., 2007). We have access to an extensive and international database on the CSR activities of companies prepared on a consistent basis, which provides us with the opportunity to provide useful insights into how the market appreciates CSR activities in various countries/regions.

2.3 Competiveness index

An alternative comparative measure of how CSR activities impact on firm valuation across nations is the Responsibile Competiveness Index, which was first calculated in 2003.⁶ The most recent index released in 2007 covered 108 countries, including almost all of those contained in this study. The definition of responsible competiveness provided by the promoters of the index is set out below:⁷

Responsible Competiveness means markets where businesses are systematically and comprehensively rewarded for more responsible practices, and penalized for the converse. Strategies for realising Responsible Competiveness aim to enhance productivity for shaping business strategies and practices, and the context in which they operate, to take account of their social economic and environmental impacts.

The index is clearly aimed at ranking nations on the basis of the extent to which the valuations of their companies should benefit from their CSR activities. However, they do recognize the possibility that these activities may not be valued by markets to the appropriate extent, which in turn will reduce the incentives of companies in these markets to pursue these activities. In Table 1 we report the ranking in the Responsible Competiveness Index for each of the countries included in our study. When we average the rankings for the various countries across the three periods for which the index has been calculated in order to calculate a single score for each of our countries/regions, we obtain the following clear ordering:

Nordic countries (first);

UK/Australia (equal second);

USA (fourth);

C&S Europe/Japan (equal fifth).

Table 1.

The competiveness index, by countries. Rankings for relevant countries drawn from the Responsible Competiveness Index prepared by AccountAbility. For a discussion of the calculation of the Responsible Competitiveness Index, see Begley et al. (2007).

Country	Ranking
	2003	2005	2007
USA	3	6	18
UK	8	7	5
Austria	14	10	16
France	23	18	17
Germany	13	12	11
Greece	28	31	33
Ireland	15	17	8
Italy	26	37	32
Netherlands	11	8	12
Portugal	20	25	21
Spain	18	19	26
Switzerland	2	5	13
Denmark	5	2	2
Finland	1	1	3
Norway	4	4	6
Sweden	6	3	1
Japan	19	14	19
Australia	3	9	9

Source: Various publications prepared by AccountAbility, London.

Note: No scores are available for Belgium or Luxembourg

The evidence from the comparative CSR studies and the comparative index rankings confirms that the companies in the Nordic countries are well ahead of the field, while consistent laggards include several European countries (Italy, Greece, Spain, France and Portugal) and Japan. However, there are inconsistencies between the two with (i) Australia having a much higher ranking in the competiveness index than a reading of the CSR literature would suggest, and (ii) the poor competitive index ranking of C&S Europe, which Williams and Aguilera (2008) would rank above the US, although still behind the Nordic countries and the UK.

Of course, neither the comparative studies nor the competitive index are perfectly aligned, but they do provide a useful reference point against which to compare the results of our analysis reported in Section 4 of this paper on the relationship between CSR activities and corporate valuation.

3. Method and data

Our concern in this study is on how the market responds to CSR activities by companies. Both management and socially responsible investors would like to see a positive response, as this would facilitate management taking a wider stakeholder perspective and mean that socially responsible investors would have a comparative advantage over other investors.⁸ The dependent variables that we use in this study are market returns, rather than any accounting measure of performance, as it is market returns that directly impact on owner wealth.⁹ Market returns have been the focus in less than 20% of studies that have evaluated the relationship between CSR and corporate performance (Orlitzky et al., 2003).

3.1 Data

3.1.1 The CSR scores and the dependent variables

The CSR data used in this study was obtained from SAM, which provides social issue ratings that reflect a broad range of CSR activities. SAM has been providing this data since 1999 and it is the data source for the preparation of the Dow Jones Sustainability Indexes.

SAM actually prepares CSR scores at three levels of aggregation. At the lowest level it prepares a score for each company on a wide spectrum of criteria, which vary across industries. These criteria are then aggregated to provide a score in each of three important areas, which they refer to as environmental, economic and social. Finally, these three component scores are combined to provide an aggregate score. The nature of these component scores are described below:

economic, which is a combination of Corporate Governance, Risk and Crisis Management, Codes of Conduct, Compliance, Corruption and Bribery plus a range of Industry-specific criteria;

environmental, which is a combination of Environmental Performance, Environmental Reporting plus a range of Industry-specific criteria;

social, which is a combination of Human Capital Development, Talent Attraction and Retention, Labour Practice Indicators, Social Reporting plus a range of Industry-specific factors.

We conduct our analysis on the aggregate score and the three component scores. SAM has provided us with eight years of CSR scores (1999–2006) for companies operating in approximately 40 countries. However, the sample size available in the majority of these countries is very small and so we have restricted our analysis to six countries/regions: the US, UK, C&S Europe, the Nordic countries, Japan and Australia. We pool the data for each country/regions across the eight years and our total sample size for each country/region is reported in Table 2.

Table 2.

Sample size by country/region. The total number of company-year observations in the final database for each country/region. This represents all data available in the Sustainable Asset Management database for our sample period, for which we have a full set of data.

Country/region	Observations
US	1119
UK	472
C&S Europe	965
Nordic	217
Japan	334
Australia	468
Aggregate	3575

The main source of information on which the scores are based is a questionnaire that is distributed at the beginning of the year, with the responses being received towards the end of the second quarter. Subsequently, the responses are verified by cross-checking them against other supplied documentation and also media and stakeholder reports and through direct contact with the company. Finally, the scores are subject to both an internal audit and also an external audit by PriceWaterhouse Coopers.

For the purposes of this study, it is extremely important for us to provide a precise interpretation of the significance of the SAM data in terms of the impact that it will have on investor expectations and so market prices. There are two things that we need to establish in order to be able to determine how best to use this data in our analysis: (i) the time to which this information relates; and (ii) the nature of the information that it provides to the market.

The SAM CSR scores are actually released to the public during the fourth quarter of each calendar year, but the long process involved in their preparation raises the question as to the actual date to which they relate. Discussion with high-level SAM staff would suggest that the information is best be regarded as almost a year ‘old’ at the time of its release and so we have chosen to ‘date’ the information as pertaining to the beginning of January in the year in which the SAM ratings are released.

It is important to note that we do not regard the release of the scores by SAM to contain significantly new information, as the ratings are almost entirely based on public information relating to events that have already occurred. As such, the SAM scores represent a convenient way of providing confirmatory information as to how active a company is in a particular area of CSR activity and to changes in its activities in these areas. Most importantly, the ratings allow us to categorize a company’s performance on the basis of various types of CSR activities, with the proposition being that changes in the level of these activities will have long-term implications for the performance of a company’s stock. Approximately 70% of the changes included in our sample are positive, with the average change being around two for the environmental and socials scores and around three for the economic scores.

In order to allow for the possibility that it may take an extended period for changes in the level of CSR activities to be reflected in the company’s valuation, we evaluate two dependent variables: the return over the calendar year when the CSR score is released (one-year returns), and the return over that, and the subsequent, year (two-year returns).

Each of the component scores are reported in a range from 0 to 100. A total CSR score is also reported, which is the aggregate of the three component scores. The descriptive statistics for each of these scores is presented in Table 3.

Table 3.

Descriptive statistics of corporate social responsibility (CSR) scores by country/region. Various statistics are reported for the entire sample and certain subsets of the sample. The aggregate CSR scores have a possible maximum score of 100. The environmental, economic and social components are scored out of a maximum of 100.

	Aggregate	US	UK	C&S Europe	Nordic	Japan	Australia
Aggregate CSR score
Mean	81.4	78.8	93.1	80.8	95.1	94.4	65.8
Median	61.5	63.2	68.8	59.0	69.1	72.0	56.3
Stan. dev.	62.3	58.6	68.4	65.0	67.2	65.3	49.0
Skewness	0.89	0.95	0.73	0.86	0.68	0.45	1.32
Environmental CSR score
Mean	23.3	20.5	28.1	23.0	30.1	34.3	15.0
Median	14.6	12.5	18.9	14.1	22.6	25.4	10.0
Stan. dev.	23.2	21.2	25.5	23.4	25.2	25.9	16.5
Skewness	1.11	1.34	0.86	1.07	0.73	0.48	1.73
Economic CSR score
Mean	32.8	33.3	35.9	32.6	36.6	32.6	29.3
Median	27.1	28.8	28.4	26.1	30.4	27.7	25.9
Stan. dev.	21.9	21.4	22.9	23.0	23.3	21.8	18.0
Skewness	0.61	0.49	0.60	0.63	0.44	0.35	1.02
Social CSR score
Mean	25.4	25.0	29.2	25.3	28.3	27.6	21.6
Median	19.3	20.4	22.7	18.0	20.4	20.8	17.1
Stan. dev.	20.8	20.0	22.9	21.7	22.0	20.7	17.4
Skewness	0.90	0.94	0.78	0.87	0.68	0.54	1.26

Based upon the aggregate score, the countries/regions that devote most resources to CSR activities would appear to be the Nordic states, the UK and Japan, who are clearly ahead of both the US and C&S Europe, with Australia bringing up the rear. Both Scandinavia and the UK are above average across the entire sample in each of the three component areas. In the case of Japan, the high overall score is largely attributable to this country being by far the most active in the environmental area, while it is at or slightly above average in the other two areas. C&S Europe is right on average across all three component parts of the aggregate score, whereas the slightly below aggregate performance of the US is largely attributable to the low environmental score. Finally, we have Australia, which scores below average for all three components, with the environmental score being particularly low.

3.1.2 Control variables

We include in our analysis a number of control variables that have been found to explain cross-sectional differences in stock returns. The actual control variables that we include are set out below:

the log of the company’s market capitalization as measured by the market value of the company’s shares (Banz, 1981);

the company’s market-to-book ratio as measured by dividing the market value of the company’s shares by the book value of the company’s ordinary shares (Lakonishok et al., 1994);

the momentum in the company’s stock as measured by the return over the previous six-months (Jegadeesh and Titman, 2001);

the company’s financial leverage as measured by its debt-to-total assets ratio (Fama and French, 1993).

These control variables were calculated as at the beginning of January each year using data obtained from DataStream.

3.2 Method

We use regression analysis in order to establish the nature of the relationship between the extent of changes in a company’s CSR activities (the independent variable) and the wealth of its owners (the dependent variable), while using several control variables. In total we have two dependent variables (one-year returns and two-year returns) and four separate CSR scores (the aggregate score, the environmental score, the economic score and the social score). As a consequence we have eight regression equations in all, each of which includes one of the dependent variables, one of the CSR scores plus each of the four control variables.

The issue on which we focus is the extent to which each CSR activity is valued by the investors in each of the countries/regions. We examine this by pooling all of the observations and including a dummy for the intercept and the CSR variable for each country/region. The actual equation of the regression is set out below:

r_{i, j, t} = \sum_{j = 1}^{6} β_{j} D_{i, j, t} + \sum_{s = 1}^{4} ψ_{s} C V_{s, i, j, t} + \sum_{j = 1}^{6} γ_{j} D_{i, j, t} Δ C S R_{i, j, t} + ε_{i, j, t}

(1)

where r_i,j,t is the return on stock i from country/region j in period t (i.e. from t to t+1 in the case of the one-year returns and t+2 in the case of the two-year returns), D_i,j,t is a dummy variable that takes on a value of 1 when stock i is from country/region j and 0 otherwise, CV_s,I,j,t is the control variable s for stock i from country/region j at the beginning of period t and ΔCSR_i,j,t is the change in the CSR score for stock i from country j from the beginning of period t – 1 to the beginning of period t (i.e. over the previous year).

The sign of the relationship between each of the changes in CSR scores provides us with a measure of the impact that changes in this score have on a company’s market valuation. A positive (negative) coefficient for a change in CSR score indicates that increased activity in this area is likely to have a favourable (unfavourable) impact on shareholder wealth.

We also examine whether there are any differences in the way that changes in CSR activities are valued by investors across the markets. To examine this issue we again have eight equations with the same dependent and independent variables as previously. However, the regression equation differs slightly to the extent that we now include not only a constant term and a CSR variable in all equations, but also include a dummy for the intercept and the CSR variable for each country/region other than the US. The actual equation for the regression is set out below:

r_{i, j, t} = β_{0} + \sum_{j = 1}^{5} β_{j} D_{i, j, t} + \sum_{s = 1}^{4} ψ_{s} C V_{s, i, j, t} + γ_{0} Δ C S R_{i, j, t} + \sum_{j = 1}^{5} γ_{i} D_{i, j, t} Δ C S R_{i, j, t} + ε_{i, j, t}

(2)

where r_i,j,t is the return on stock i from country/region j in period t (i.e. from t to t+1), D_ijt is a dummy variable that takes on a value of 1 when stock i is from country j and 0 otherwise, CV_s,i,j,t is the control variable s for stock i from country/region j at the beginning of period t and ΔCSR_i,j,t is the change in the CSR score for stock i from country j in period t from the beginning of period t – 1 to the beginning of period t (i.e. over the previous year).

The coefficient on each of the dummy variables for each of the CSR variables measures how the relationship between that variable and the dependent variable differs for a particular country/region as compared with our benchmark, the US. For instance, a significant positive coefficient will indicate that a change in the CSR score has a much larger impact on the dependent variable in a particular country/region than it had in the US.

4. Findings

4.1 Differences in returns to CSR practices across countries/regions

As is evident from the discussion in the previous section, we run two sets of regressions. The first set of regressions is aimed at identifying the relationship for each country/region between each of the four CSR variables and each of the two dependent variables across each of the countries/regions. In Table 4 we report the results of the regressions as specified in Equation (1); specifically, we report the coefficient on each of our CSR variables for each of the dependent variables.

Table 4.

The market impact of changes in corporate social responsibility (CSR) activities by country/region. The coefficients pertain to the relationship between the returns over a one- and two-year periods and changes in each of the CSR scores and the aggregate scores. These coefficients are obtained by applying Equation (1) to our data set. A positive (negative) coefficient is indicative that an increase in a CSR score will have a positive (negative) impact on the realized market return. Significant tests are reported for each of the coefficients with *** indicating the coefficient is significant at the 1% confidence level, ** at the 5% level and * at the 10% level

Dep. var.	US	UK	C&S Europe	Nordic	Japan	Australia
CSR variable: environmental score
One-yr. ret.	0.00233***	0.00370***	0.00377***	0.00412***	0.00097	0.00182*
Two-yr. ret.	0.00004	0.00357***	0.00297***	0.00392***	0.00158	0.00280
CSR variable: economic score
One-yr. ret.	0.00150***	0.00439***	0.00339***	0.00426***	0.00123	0.00165
Two-yr. ret.	–0.00082	0.00335***	0.00161*	0.00374**	0.00100	0.00256
CSR variable: social score
One-yr. ret.	0.00237***	0.00335***	0.00245***	0.00334***	0.00030	0.00146
Two-yr. ret.	–0.00043	0.00362**	0.00089	0.00325**	0.00179	0.00221
CSR variable: total score
One-yr. ret.	0.00079***	0.00148***	0.00123***	0.00144***	0.00033	0.00066*
Two-yr. ret.	–0.00017	0.00125***	0.00070**	0.00133**	0.00057	0.00102*

We will discuss our findings first based for each of the component scores and then the aggregate score.

4.1.1 Environmental CSR score

There is a strong positive relationship between changes in the environmental score and market returns over a one-year holding period in the Nordic countries, the C&S European countries, the UK and the US and a slightly weaker positive relationship Australia. Similar, albeit slightly weaker, findings apply over a two-year holding period for all of the European countries/regions, but the results for both the US and Australia are no longer significant, while those for Japan remain insignificant. These results suggest that the European markets embrace attempts by companies to be environmentally conscious and that this will be reflected in the market valuations of these companies. There is a weaker and shorter-term positive impact in both the US and Australia. In Japan it is clear that such activities have no impact on market valuations, which is a somewhat surprising given that the Japanese companies in our sample have by far the highest average environmental score.

4.1.2 Economic CSR activities

First considering a one-year holding period, it is evident that economic CSR activities are welcomed by investors in all of the European markets and in the US, but that they have no significant impact on markets in Japan and Australia. This same finding extends to a two-year holding period for the Nordic countries and the UK, while it weakens but is still significantly positive for C&S Europe, and it disappears for the US. It is the UK and the Nordic countries that have the highest economic scores, which is consistent with investors in these countries being the strongest supporters of companies that take their social responsibilities seriously.

4.1.3 Social CSR scores

The findings for the social CSR activities play out similarly to those previously discussed for environmental and economic CSR activities. The same strong significant relationship exists between one-year returns and changes in social CSR activities in the US, UK, Nordic countries and C&S Europe. This significant positive relationship is confirmed for both the UK and the Nordic countries over a two-year holding period when it becomes insignificant for both the US and C&S Europe and remains insignificant for both Japan and Australia. The same pattern that has emerged for the environmental and economic activities applies to the social activities and this will be discussed in more detail in the following section, which addresses the findings for the aggregate CSR activities.

4.1.4 Aggregate CSR score

The findings for the aggregate CSR score reflects, as one might expect, the findings for the three sub-categories with the investors in the UK and Nordic markets clearly supporting companies who expend resources on CSR activities. The UK and the Nordic companies (along with Japanese companies) have the highest CSR scores and it is clear that they are rewarded for so doing. The investors in the C&S Europe markets are also supportive of companies devoting resources to CSR activities, with the strength of the relationship being weaker and less sustained than is the case for both the UK and the Nordic countries. The US is an interesting case, as the aggregate CSR activities have a positive impact on returns and so market valuation in the first year, but this impact largely reverses itself over the subsequent 12 months, resulting in no significant relationship between returns and the aggregate CSR activities over a two-year holding period. For Australia, there is small significant positive relationship over both a one- and two-year holding period. However, despite the fact that Japanese companies rank with the UK and the Nordic region in terms of the extent of their CSE endeavours, they certainly seem to be the companies included in our sample that benefit least from doing so.

4.2 Comparing each country/region with the US

The second set of regressions is aimed at facilitating a more direct comparison of the market impact of the CSR activities across the six countries/regions. We use the US market as the benchmark, both because it is the largest market and because the analysis to date would suggest that it has a near-average ranking in terms of the market impact of its CSR activities. The coefficients reported in Table 5 for the various countries/regions are derived by applying Equation (2). The coefficients reported in the US column are identical to those reported previously in Table 4, while those for the other countries/regions reflect the difference between the coefficient for each country/region and that for the US, and provide a measure of the statistical significance of these differences.

Table 5.

The impact relative to the US of corporate social responsibility (CSR) activities on shareholder wealth by country/region. The coefficients pertain to the relationship between the returns realized over one- and two- year periods and changes in each of the CSR scores and the aggregate scores for each country/region relative to the US. These coefficients are obtained by applying Equation (2) to our data set. The column repeats the coefficients for the US previously reported in Table 4, while the coefficients reported in the other columns for each of the other countries/regions indicate the extent to which a change in a CSR score for that country/region has a different effect on market-adjusted returns than is the case in the US. A positive (negative) coefficient is indicative that an increase in a CSR score will lead to a greater (lesser) return than is the case in the US. Significant tests are reported for each of the coefficients with *** indicating the coefficient is significant at the 1% confidence level, ** at the 5% level and * at the 10% level

Dep. var.	US	UK	C&S Europe	Nordic	Japan	Australia
CSR variable: environmental score
One-yr. ret.	0.00233***	0.00136*	0.00144**	0.00179*	–0.00136	–0.00051
Two-yr. ret.	0.00004	0.00354***	0.00294**	0.00389**	0.00155	0.00277
CSR variable: economic score
One-yr. ret.	0.00150***	0.00289***	0.00189***	0.00276***	–0.00026	000015
Two-yr. ret.	–0.00082	0.00417***	0.00243**	0.00456***	0.00182	0.00338*
CSR variable: social score
One-yr. ret.	0.00237***	0.00098	0.00008	0.00107	–0.02077**	–0.00091
Two-yr. ret.	–0.00043	0.00305**	0.00132	0.00368**	0.00221	0.00264
CSR variable: total score
One-yr. ret.	0.00079***	0.00068**	0.00044*	0.00065*	–0.00047**	–0.00013
Two-yr. ret.	–0.00017	0.00141**	0.00086**	0.00150**	0.00074	0.00118*

We will use the information contained in Table 5 not only to compare each of the countries/regions to the US experience, but also to compare them to each other.

4.2.1 Environmental CSR score

We have previously identified that CSR activities by US companies relating to the environment are valued by market participants over a one-year holding period, but that this largely dissipates during the second year. It is clear that the investors in all of the European markets place a much higher value on environmental CSR activities than do the US investor over both one- and two-year holding periods. However, it difficult to differentiate between the level of acceptance of these activities in the European markets, with investors in the UK, the Nordic markets and the C&S Europe markets all embracing the environmental CSR activities fairly equally. The reaction to CSR environmental activities in the Australian and Japanese markets is similar to that in the US, suggesting that all three countries should rank fairly equally behind the European markets with respect to the market reaction to their environmental activities.

4.2.2 Economic CSR score

The same clear ranking occurs with respect to market reaction to the CSR economic score that occurs for the environmental score. One difference is that the positive reaction in the European markets is even stronger relative to the US than was the case with the environmental score. Another difference is that the longer term reaction in Australia would seem to be greater than that experienced in either the US or Japan.

4.2.3 Social CSR scores

The US market over a one-year time investment horizon would seem to rank on a par with the European markets in terms of the extent to which it reacts to CSR social activities. Over a two-year time horizon, there is a stronger positive reaction to social CSR activities in both the UK and the Nordic markets than is the case in the US, the C&S European and the Australian markets. However, the Japanese market would appear to be a laggard, clearly being the market with the lowest response to CSR social activities.

4.2.4 Aggregate CSR score

Based on the evidence provided in Tables 4 and 5 and the previous discussion of the components of the aggregate score, it is clear that the six countries/regions fall into two groups. Investors in the UK and the countries within the Nordic States and C&S Europe regions display a stronger positive link between aggregate CSR activities and corporate valuation than is the case in the US, Japan and Australia. Of this first group, it is impossible to differentiate between the Nordic states and the UK, but both enjoy a greater market reaction to their CSR activities than is the case in the C&S European markets. The one-year returns suggests that the US market strongly supports companies that invest in CSR activities, but the fact that the coefficient on the two-year returns is insignificantly different to zero questions whether this support is maintained beyond one year. There is some weak evidence that Australian support is stronger than that in the US over a two-year holding period with similar weak evidence of the Japanese market lagging further behind. Overall the evidence is stronger that there is positive relationship between CSR activities and stock returns in the US than in Japan and Australia, where one could reasonably conclude that there is no relationship.

4.3 Discussion of these rankings

In addition to the rankings from our analysis on the relationship between the CSR activities of a company and its future returns, we have two additional looser sets of rankings: one based on the available comparative evidence in the literature and the other on the competiveness index, which ranks countries in terms of the impact of CSR on their international competiveness.

We have attempted to summarize these three sets of rankings in Table 6. It must be remembered that each of these rankings is based on analysis designed for different purposes: the comparative studies attempting to reflect the CSR behaviour of companies in the various countries/regions; the competiveness index attempting to align the international competiveness of countries/regions on the basis of their CSR activities; this study, which attempts to measure the impact of a company’s CSR activities on its market returns, and so its market valuation. Given these differences, the similarities between the overall findings is quite pleasing, especially because of the similarities between the rankings from this study and that suggested by the comparative studies, which suggest an alignment between the willingness of the corporate sector to embrace their social responsibilities and the willingness of investors to reward them for so doing.

Table 6.

A comparison of the rankings. The rankings under ‘comparative studies’ are based on the literature surveyed in Williams and Aguilera (2008); the ‘competitive index’ entries are based on the information provided in Table 1 in this paper; the ‘our study’ rankings are based on the information provided in Tables 4 and 5.

	Comparative studies	Competitive index	Our study
First	Nordic	Nordic	Nordic/UK(eq.)
Second	UK/C&S Europe (eq.)	UK/Australia (eq.)	Nordic/UK(eq.)
Third	UK/C&S Europe (eq.)	UK/Australia (eq.)	C&S Europe
Fourth	US	US	US
Fifth	Japan/Australia (eq.)	C&S Europe/Japan (eq.)	Japan/Australia (eq.)
Sixth	Japan/Australia (eq.)	C&S Europe/Japan (eq.)	Japan/Australia (eq.)

This finding confirms the proposition of Cannon (1992), who argued that Nordic companies proactively incorporate certain non-shareholder interests in the corporate policy. This overall commitment in CSR activities, especially in areas of environmental stewardship and social issues, created a strong trust of the market participants, who consequently translated the credible social investments into the market price by rewarding (penalizing) the proactive (reactive) involvement in CSR. Indeed, there is a culture in the Nordic countries that values CSR activities of the corporate sector that goes beyond that to be found in other countries.

The consistently high ranking of the UK, which makes it almost the equal of the Nordic countries, reflects attitudes in UK companies that have been found to be a composite of this European culture and also the more philanthropic corporate posture of the US (see Maignan and Ralston, 2002). It particularly reflects the greater emphasis placed on the social responsibilities of companies during the era of Prime Minister Tony Blair (see Williams and Conley, 2005). The fact that the C&S Europe region lags behind the Nordic countries and the UK is reflective of the mixture of attitudes across the 12 countries. This result should come as no surprise considering the more lax legislation adopted by many European countries, where typically governments leave companies relatively free to decide the level of their general commitment to CSR, with the result that these investments, particularly in the area of environmental protection, do not soar to the very top of the corporate agenda of business leaders in C&S Europe, even though they are still more accentuated than in the US.

The neutral relationship between CSR and market performance of the Japanese firms highlights what Tanimoto (2002) defined as a kind of strategic integration of various stakeholders into a system where the interests of the firm have the priority. In the last 60 years, with the corporate ownership under a cross-shareholding control, Japanese stakeholders were put within the scope of corporate control and ended up identifying the corporate interest as their own as long as the integration guaranteed economic prosperity to society. This alignment of views, combined with the ‘micro moral unity paradigm’ suggested by Wokutch and Shepard (1999) combine to explain the relatively poor showing of Japanese companies documented in this study. In the case of Australia, Lucas (2004) argued that Australian corporate leaders were ill prepared to meet the new challenge of corporate citizenship and needed to clearly understand the expectation of a larger set of stakeholders. Glazebrook (1999) used annual reports and other information released by firms to analyse the corporate vision, business objectives and overall performance measures of the top 500 Australian companies. He found that only 37 companies (7% of his sample) actually viewed CSR as central to the strategic direction of their business. Of these 37, there were only 15 instances of the company’s policies actually being consistent with its aspired strategic social posture. In more recent times there are signs of a greater commitment by Australian companies and the Australian Government to CSR, especially in the environmental area.

5. Summary

Our focus is on the relationship between a company’s CSR activities and its market performance, which is important both for managers trying to balance the objectives of shareholders with those of a wider set of stakeholders and for investors contemplating socially responsible investing. There have been many US studies of this subject, few in other countries and even less that have attempted a comparative study across countries/regions. In this paper we seek to redress this situation by undertaking a cross-border comparison of the attitude of investors to corporate CSR activities.

We find evidence to suggest that investors in European markets react favourably to CSR activities, whereas these activities have little or no impact on market prices in the US, Australia and Japan. These findings are consistent with our expectations based on the attitudes displayed to CSR in the various countries/regions. When we conduct the same study at the level of the three sub-sectors, we find very similar results to those found at the aggregate levels for both environmental and economic CSR activities. However, the results differ slightly for the social CSR activities, where the superiority of the European companies is weaker and the reaction of the Japanese investors is at variance to those for the other countries/regions.

The findings from our analysis of the cross-border market impact of CSR activities are fairly consistent with the scant evidence from previous studies conducted at the level of the individual countries/regions. We have been able to align our rankings of the markets attitude to CSR activities across countries/regions with community attitudes towards CSR and to an independent index that relates the competiveness of countries to their CSR activities. Our overall findings are supportive of management taking a corporate responsible stance, as there is little evidence to suggest that this is to the detriment of their current and prospective shareholders in any of our countries/regions.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Notes

References

Anderson

Landau

(2006) Corporate Social Responsibility in Australia: A Review, Monash University Department of Business Law & Taxation Research Paper No. 4. Available at SSRN: http://ssrn.com/abstract=1027845

Banz

(1981) The relationship between return and market value of common stock. Journal of Financial Economics 9: 3–18.

Becchitti

Ciriretti

(2006) Corporate social responsibility and stock market performance. Applied Financial Economics 19: 1283–1293.

Begley

do Nascimento

MacGillivray

Boechat

(2007) The Responsible Competiveness Index. London: AccountAbility.

Bird

Hall

Momenté

Reggianni

(2007) What corporate social responsibilities are valued by the market? Journal of Business Ethics 76: 189–206.

Brammer

Brooks

Pavelin

(2006) Corporate social performance and stock returns: UK evidence from disaggregate measures. Financial Management 35: 97–116.

Brine

Brown

Hackett

(2009) Corporate Social Responsibility and Financial Performance in the Australian Context. Australian Treasury Working Paper.

Cannon

(1992) Corporate Responsibility. London: Pitman Publishing.

Crisostomo

Freire

de Vasconcellos

(2010) Corporate Social Responsibility, Firm Value and Performance in Brazil. University of Brazil Working Paper.

10.

Doh

Guay

(2006) Corporate Social Responsibility, Public Policy, and NGO Activities in

11.

Economic Intelligence Unit (2008) A special report on corporate responsibility. The Economist 19 January.

12.

Fama

French

(1993) Common risk factors in the returns of stocks and bonds. Journal of Financial Economics 33: 3–56.

13.

Fiori

di Donato

Izzo

(2007) Corporate Social Responsibility and Firms Performance. An Analysis on Italian Listed Companies. Luis Guido Carli University Working Paper.

14.

Glazebrook

(1999) Corporate citizenship and action research: an Australian perspective. In: Proceedings of the International Association for Business and Society Conference, 10th Anniversary, Paris, France.

15.

Griffin

Mahon

(1997) The corporate social performance and corporate financial performance debate. Business and Society 36: 5–31.

16.

Jegadeesh

Titman

(2001) Profitability of momentum strategies: an evaluation of alternative explanations. Journal of Finance 56: 699–720.

17.

Lakonishok

Shleifer

Vishny

(1994) Contrarian investment, extrapolation, and risk. Journal of Finance 49: 1541–1578.

18.

Lucas

(2004) The emerging practice of corporate citizenship in Australia. Journal of Corporate Citizenship 13: 28-30.

19.

Maignan

Ralston

(2002) Corporate social responsibility in Europe and the US: Insights from business self-presentations. Journal of International Business Studies 33: 497–514.

20.

Orlitzky

Schmidt

Rynes

(2003) Corporate social and financial performance: a meta-analysis. Organization Studies 24: 403-441.

21.

Rogers

Choy

Guiral

(2008) Do Investors Value a Firm’s Commitment to Social Activities? The Moderating Role of Intangibles and the Impact of the Sarbanes-Oxley Act. UCL Riverside Working Paper.

22.

Roman

Hayibor

Agle

(1999) The relationship between social and financial performance: repainting a portrait. Business and Society 38: 109–125.

23.

Tanimoto

(2002) Reconstructuring the Corporate Society. Working paper, Chikura Shobo, Tokyo.

24.

Tanimoto

Suzuki

(2005) Corporate Social Responsibility in Japan: Analyzing the participating companies in global reporting initiative. Stockholm School of Economics, EIJS, Working Paper 208.

25.

Vogel

(1992) The globalization of business ethics: why America remains distinctive. California Management Review Fall: 30–49.

26.

Welford

(2005) Corporate social responsibility in Europe, North America and Asia. Journal of Corporate Citizenship Spring: 33–52.

27.

Williams

Aguilera

(2008) Corporate social responsibility in a comparative perspective. In: Crane

McWilliams

Matten

Moon

Siegel

(eds) The Oxford Handbook of Corporate Social Responsibility. Oxford: Oxford University Press.

28.

Williams

Conley

(2005) An emerging third way?: The erosion of the Anglo-American shareholder value construct. Cornell International Law Journal 38: 493-551.

29.

Wokutch

(1992) Myths of the Japanese factory. Journal of Commerce 26: 10A.

30.

Wokutch

Shepard

(1999) The maturing of the Japanese economy: corporate social responsibility implications. Business Ethics Quarterly 9: 527–540.