Corporate Social Performance of Latin American companies

(1)

MASTER’S THESIS

Corporate Social Performance of Latin American

companies

An Event Study

August 2009

Rick Johannes Boerkamp

(2)

Corporate Social Performance of Latin

American companies: An Event Study

Abstract:

Literature examining the Corporate Social Performance (CSP) – Corporate Financial Performance (CFP) link is extensive, but little is known about this relationship outside Western markets. A statistically significantly negative CSP-CFP link is found for a sample of 50 events of stock listed Latin American companies. We argue that the relatively smaller portion of consumers in Latin America (compared to Western countries) that is able to buy responsibly, suggests less need for companies that are merely or mostly active in Latin America to act socially responsible. Consequently, if unsolicited, Corporate Social Responsibility (CSR) is spending of shareholders’ money.

Keywords:

Corporate Social Responsibility, Corporate Financial Performance, Corporate Social Performance, Event study, Latin America, Latin American investor, Western investor.

Publication: Master’s Thesis (August 7, 2009) Author: Rick J. Boerkamp

(3)

2

1 Introduction

In the literature examining the link between Corporate Social Performance (CSP) and Corporate Financial Performance (CFP), little attention is paid to the differences between geographical regions. Most studies focus on the USA. Meta-study analyses showed that overall, literature emphasizes a positive CSP-CFP link. (Orlitzlky et al., 2003; Wu, 2006). However, one could question whether the conclusions drawn from the above mentioned studies are generalizable on a global level.

Quazi and O’Brien (2000) and Jamali et al. (2009) provide some evidence that differences exist between orientations toward Corporate Social Responsibility (CSR) of managers from different countries. We may therefore expect to see differences in orientation toward CSR between consumers and investors as well.

Investors from different countries could have different thoughts on what is good for society1, caused by differences in culture between countries. However, we will focus on economic arguments to provide a possible explanation for the existence of differences in investors’ behavior of investors of different countries. We argue that even if investors’ values and/or preferences are identical around the globe, differences in consumers’ attitude may induce differences in investor behavior.

Corporate reputation among consumers is an important driver to conduct CSR activities. Non Governmental Organizations (NGOs), like Amnesty International, Greenpeace and the WWF can prompt consumers to boycott a certain company if it acts irresponsibly, which could lead to a dramatic drop in sales. However, as Valor (2008) points out, to buy responsibly is still time consuming, stressful and economically disadvantageous. The latter suggest that rich consumers

1_{For example, investors from Muslim countries are likely to oppose investing in banks that charge interest}

(5)

4 are probably better able to buy responsibly than poor consumers (assuming that the rich consumers still have some time left for responsible buying). In other words it is likely that the price–argument weighs more heavily in a purchase decision of the majority of the consumers in a developing country than of consumers in a rich country. In addition to Valor’s findings, the Environmental Kuznets Curve2 and Maslow’s theory of human needs3 (Maslow, 1943) also imply that consumers’ ability to buy responsible products is constrained by their budget.

Table 1.1

GDP per capita, poverty and income inequality (Gini coefficient) for some Latin American countries, Canada, China, the UK and the USA.

Presents figures of GDP per capita at Purchasing Power Parity (PPP) in international dollars as estimated by the IMF, country’s global rank based on the latter, percentage of the population living below the national poverty line according to CIA’s World Factbook and Gini coefficients, a measure for income inequality, calculated by the United Nations. A Gini coefficient value of 0 represents absolute equality, and a value of 100 absolute inequality.

Country

GDP per capita at PPP*

Rank (GDP per capita at PPP)*

Population living below national poverty line in %

(year data refers to)** UN Gini***

Argentina 14,413 58 23.4 (2007) 51.3 Brazil 10,326 77 31 (2005) 57 Canada 39,183 13 10.8 (2005) 32.6 Chile 14,510 56 18.2 (2005) 54.9 China 5,963 100 - 46.9 Mexico 14,560 54 13.81(2006) 46.1 United Kingdom 36,523 18 14 (2006) 36 USA 46,859 6 12 (2004) 40.8

* World Economic Outlook April 2009 (IMF), data refers to 2008; ** The World Factbook (CIA), accessed April 10, 2009; *** UNDP Human Development Report 2007/2008; 1 Food-based poverty. Asset based poverty amounted to more than 40%

2_{The Environmental Kuznets Curve (EKC) is the inverted U relationship between income per capita and a}

pollutant factor per capita (Scholtens, 2006). The EKC describes the process of environmental quality deterioration at early stages of economic development and the improvement at later stages. Dinda (2004) states that people with higher income tend to have a higher preference for environmental quality.

3

(6)

5 As can be seen in Table 1.1., GDP per capita in Latin American countries is lower than in the USA, the UK, and Canada. Furthermore, income inequality (Gini coefficient) and the portion of the population that lives below the national poverty line is greater in the Latin American countries than in the USA, the UK or Canada.

The preceding gives economic reasons to assume that the attitude toward CSR of society/stakeholders and therefore investors’ attitude in Latin-America is different from the Western investor. In an event study we examined the CSP-CFP link for stock listed companies in Argentina, Brazil, Chile and Mexico for the period August 19, 2002 till December 31, 2008. The results of our study provide some evidence that the CSP-CFP relationship is negative for Latin America.

The structure of the paper is as follows; in the next chapter relevant literature will be discussed, in chapter three our hypotheses and in chapter four performance measurements and event study methodology. Data descriptions will be given in chapter five, the results in chapter six and a robustness check in chapter seven. Finally, concluding remarks will follow in chapter eight, limitations and recommendations in chapter nine.

2 Literature

In this literature chapter the following will be discussed; in brief the contemporary notion of CSR, literature examining the CSP-CFP relationship, the way literature measured CSP and CFP, and event studies examining the CSP-CFP link. Information about measurement of CSP and CFP used in our event study is given in chapter four.

2.1 The contemporary notion of CSR

(7)

6 the CSR concept transitioned significantly to alternative themes such as stakeholder theory, business ethics theory, CSP, and corporate citizenship (Carroll, 1999). According to Montiel (2008) earlier research did not define CSR well and the range of definitions found in literature reflects the ambiguous nature of perspectives on CSR. Montiel (2008) found that “The most

often cited definition is Carroll’s (1979) statement that - the social responsibility of business encompasses the economic, legal, ethical, and discretionary expectations that society has of organizations at a given point in time -”. A fairly recent definition that in our opinion captures

the concept of CSR well is McWilliams and Siegel’s (2001) definition of CSR:

“actions that appear to further some social good, beyond the interests of the firm and that which is required by law.”

According to McWilliams and Siegel (2001) CSR means “going beyond obeying the law”. We will use their definition as point of departure in our study.

2.2 The CSP-CFP link

(8)

7 intervening variables, the CSP-CFP relationship could even well not exist, except perhaps by chance.

Literature examining the CSP-CFP relationship is extensive and earlier literature was called inconclusive by Wu (2006). Orlitzky et al. (2003) and Wu (2006), both recent literature surveys, tried to end the inconclusiveness by making use of meta-analysis technique4 and emphasize a positive CSP – CFP relationship.

An overview of the CSP-CFP literature is given in table A.1, found in the appendix. The list in table A.1 is a compilation of the studies used in the literature reviews of Margolis and Walsh (2003), Orlitzky et al. (2003), Wu (2006), and Beurden and Gössling (2008). Note that 38 studies were excluded, of which one used a subjective measure of CFP, the others were not available to the author (mainly Business and Society Review publications). In the literature overview of 119 studies just two focus on a region outside the Western world. He et al. (2007) focused on China, while Peinado-Vara (2006) focused on Venezuela and Colombia. As can be seen in Table A.1, Western focus dominates in CSP-CFP literature. As said before, the conclusions drawn from previous literature do not necessarily apply to Latin America or other geographical regions in the world.

Studies that found a negative CSP-CFP relationship are clearly a minority in the literature review. Most studies emphasize a positive CSP-CFP relationship.

2.3 Differences in measurement

The use of different measurement techniques yields different results. Therefore the literature is categorized based on a classification proposed by Beurden and Gössling (2008). They categorized CSP into three categories; measurement based on disclosure (CSP 1), on corporate actions (CSP 2), and on corporate reputation ratings (CSP 3). CFP measurement was divided in

4

(9)

8 two categories; market-based (CFP 1), and accounting-based (CFP 2). We will add a third category for other measurement (CFP 3), like organizational competence, market share, and cost advantages. All categories can be found in table 2.1.

Table 2.1

Corporate Social and Financial Performance measurement categorization

Presents descriptions of different Corporate Social Performance (CSP) and Corporate Financial Performance (CFP) measurement categories. Adapted version of Beurden and Gösslings’ (2008) classification.

CSP 1 Disclosure CFP 1 Market-based measures The extent and content analysis of

social disclosure

Market to book value, price per share, market return, etcetera

CSP 2 Corporate action CFP 2 Accounting-based measures Concrete observable CSR processes

and outcomes

_{Profitability and asset utilization} CSP 3 Corporate reputation ratings CFP 3 Other measures

KLD, Fortune, EIRIS, socially

responsible mutual funds

Organizational competences, cost advantages, market share, etcetera

Market- and accounting-based measures both have their merits to reflect social performance (CSP measures will be discussed later in this chapter). Wu (2006) found that market-based measures are weaker predictors of CSP than other financial measures, and that CSP measures based on perception (like Fortune ratings) provide the strongest positive link. The latter can be attributed to a halo-effect, because Fortune’s annual ratings are strongly influenced by past financial performance (Brown and Perry, 1994; Wu. 2006). This example illustrates that it is often hard to isolate CSR activities and especially with accounting-based measures it is not clear whether higher CFP follows from higher CSP or the other way around. The latter (higher CSP following from higher CFP) is supported by the slack resource theory (Waddock and Graves, 1997).

(10)

9 measures (74 to 65). It is beyond the scope of this study to go into details about all the different measures used in the CFP categories, but an overview of event studies will follow below.

Table 2.2

Occurrence of Corporate Social and Financial Performance measurement combinations found in studies of our literature overview

Presents the occurrence of Corporate Social and Financial Performance measurement combinations found in studies of our literature overview. CSP 1: based on disclosure; CSP 2: corporate actions and CSP 3: corporate reputation ratings. CFP measurement is divided in market-based (CFP 1), accounting-based (CFP 2) and other (CFP 3). Between parentheses respectively the number of studies that found a positive, non-significant, negative, or mixed relationship. Note that some studies used several categories.

Performance categories CFP 1: Market CFP 2: Accounting CFP 3: Other CSP 1: Disclosure 14 (10/2/0/2) 10 (5/4/0/1) 0 (0/0/0/0) CSP 2: Corporate Action 12 (5/4/3/0) 22 (16/4/1/1) 4 (4/0/0/0) CSP 3: Corporate reputation 48 (21/14/2/11) 33 (17/9/0/7) 1 (1/0/0/0)

Total 74 65 5

2.4 Event studies

Of the fourteen studies that fall into the categories CSP 1 and CFP 1, six are event studies. The event studies in the category CSP 1 used disclosures of the following as measure of CSP; pollution expenditures (Belkaoui, 1976; Blacconiere and Northcut, 1997; Blacconiere and Patten, 1994), the impact of a forthcoming labor standard (Freedman and Stagliano, 1991), various disclosure categories (Ingram, 1978), and TRI (toxic release inventory) emissions (Konar and Cohen, 1997). All the six event studies found a positive relationship. Although self-disclosure could be used to measure CSP, self-disclosure is not always consistent with objective measures of CSP (Ullmann, 1985).

(11)

10 event studies five studied announcement of withdrawal from South-Africa, reporting conflicting results. McWilliams et al. (1999) point out that differences in defining the event, difficulty in identifying the correct event date, differences in the length of the event windows, and differences in controlling for confounding events and industry led to the conflicting results.

Table 2.3

Event studies with corporate actions as measurement of Corporate Social Performance

Presents what corporate actions were used as measurement for Corporate Social Performance (measurement category CSP 2) by event studies of our literature overview. The event studies found positive, non-significant or negative abnormal returns and CSP-CFP relationships. No event study in the CSP 2 category found a mixed CSP-CFP relationship. Abbreviations: DII: Defense Industries Initiative; SA: South Africa.

Findings CSP 2 measurement Abnormal returns

Positive relationship

Davidson and Worrell (1992) Product recalls Negative Kumar et al. (2002) End of boycott of SA Positive Posnikoff (1997) Divestment or withdrawal from SA Positive

Non-significant relationship

Patten (1990) Signing / Non signing Sulivan Principles Not significant Teoh et al. (1999) Withdrawal from SA Not significant

Negative relationship

Boyle et al. (1997) Signing DII(ethical code) Negative

Meznar et al. (1994) Withdrawal from SA Negative

Wright and Ferris (1997) Withdrawal from SA Negative

(12)

11 The five event studies mentioned by McWilliams et al. (1999) that studied withdrawal from South Africa left aside, most event studies in our literature review emphasize a positive CSP-CFP link.

3 Hypotheses

As mentioned in chapter two, previous literature, which predominantly had a Western focus, points at a positive CSP-CFP relationship. However, compared to the USA, the UK, or Canada, Latin American countries face lower GDP per capita, higher Gini – coefficients, and a greater share of the population lives below the (international) poverty line. We argue that it is likely that it does not pay for Latin-American firms to be socially responsible, because the majority of the consumers does not have the necessary means to buy socially responsible products, which are in general products of expensive premium brands. Therefore, we expect a negative CSP-CFP relationship for Latin American countries.

As we are interested in the investors’ reaction after CSR - announcements (with positive social output aim; explained in detail in chapter four) made by Latin-American companies and want to know whether this reaction is negative the following is hypothesized.

Hypothesis 1:

H0: Corporate Financial Performance (measured as abnormal returns) of stock

listed Latin American companies has no significant relationship to CSR – news of those companies.

H1: Corporate Financial Performance (measured as abnormal returns) of stock

listed Latin American companies is significantly negatively related to CSR – news of those companies.

(13)

12 investors will probably make up a larger part of the investor public of Latin American stocks with a cross- listing (ADR or GDR) than of Latin American stocks without a cross-listing. Levy Yeyati et al. (2009) examined the cross-market premium (the ratio between the domestic and the international market price of cross-listed firms) for nine countries; among Argentina, Brazil, Chile and Mexico. Levy Yeyati et al. (2009) found that price deviations across markets are rapidly arbitraged away and bands are narrow. They found that (at least for Argentina, Brazil, Chile and Mexico) the biggest price deviations occurred during periods of government control on cross-country capital movement. None of the used events in our study fell in a period of capital control. As such we can assume that, given some fluctuations in the cross-market premium, in general the arbitrage mechanism works well and the law of one price holds for the examined countries in our study (Levy Yeyati et al., 2009). Thus, while acknowledging that a price change of the domestic stock and the DR (depositary receipt) after a news event may not be the same or coincide constantly, the prices tend to converge. Hence, the domestic stock price and the price of the DR highly influence each other. As we are interested in whether the stock market’s reaction to CSR news of Latin American companies with a cross-listing (which are likely under more scrutiny of Western investors than Latin American companies without a cross-listing) differs from the stock market’s reaction to CSR news of Latin American companies without a cross-listing , the second hypothesis is as follows:

Hypothesis 2:

H0: The Corporate Financial Performance (measured as abnormal returns) –

Corporate Social Performance relationship of companies with a cross-listing on a Western stock market is not statistically significantly different from companies with no cross-listing on a Western stock market.

H1: The Corporate Financial performance (measured as abnormal returns) –

(14)

13 these companies, it is not preferable to differentiate the sample with a limited number of firms among industries. Besides, after examining the studies in our literature review, this is not uncommon. Further, other studies argue that size is of influence on CSR activity as well (e.g. Stanwick and Stanwick, 1998), however Wu (2006) found no support for this claim. Therefore, firms in our sample will not be differentiated on industry nor on size.

4 Methodology

Our study is based on event study methodology, which is explained in this chapter. Furthermore, the performance measures and the statistical tests used are explained.

Our measure of CSP is voluntary corporate CSR action with the aim to create a positive social output5 that is not forced upon by a regulatory body; signing of ethical treaties, environmental investments (for example ISO 14000), investments in occupational health (OHSAS 18000), substantial charitable contributions, corporate governance improvements, health and education improvements. Actions that fit our measure of CSP are CSR actions that could result in higher financial returns if the investor expects that the gains (such as reputation improvement) will be greater than the costs of the particular event. Excluded are actions of which a priori CSR-related financial profits can be expected, as in the case of environmental investment projects which will be registered as an United Nations Clean Development Mechanism (CDM) project.6

Trough the LexisNexis news portal archives of daily newspapers in Argentina, Brazil, Chile and Mexico were searched for headwords that would detect the firm news of the topics that we are

5

If a firm made charitable contributions it expects a positive social output. When a firm causes environmental contamination, like an oil spill by a sunk oil tanker, the social output is clearly a negative one. See Maron’s (2006) unified theory on the CSP-CFP link.

6_{Projects that could generate substantially positively future cash flows for the firm at which the project is}

(15)

14 interested in. In order to reduce time spent on unnecessary searching, Latin-American countries with relatively small stock exchanges are not included in this study. The headwords of the topics are formulated in a way that the topics can be detected in English, Spanish and Portuguese. Boolean searches were manually performed in the following construction: headwords(name firm) AND headwords(topic). The headwords for the topics are specified in table 4.1. A list of the headwords used for the name of the firm is available on request from the author.

Table 4.1

Headwords used to find relevant news articles in the LexisNexis database

Presents the headwords used to find relevant news articles in the LexisNexis database based on Boolean search. The exclamation mark is used to replace an infinite number of letters following a word root.

Topic Headwords (OR relation)

Code of good practice (good practice!) (best practice!) (buena! practica!) (boa! pratica!)

Corporate Governance (corporate

governance)

(gobierno

corporativo) (governança corporativa)

Corporate Social Responsibility (social responsibility) (responsabilidad social) (responsabilidade social) (responsabilidade corporativa) Discrimination discrimina!

Donations dona! doa!

Education educa!

Environment environment! medioambient! (medio ambient!) (meio ambient!)

ISO 14000 14000 14001 14004

OHSAS 18000 18000 18001 18002

The event date must be determined to correctly assess the impact of the event on the financial return of a firm. News articles are only helpful if they reflect new information about the event. News that reports on a charitable contribution made by a foundation that carries the company’s name is therefore not identified as an event, because it is unclear whether this charitable contribution coincides with a negative cash flow for the firm at that moment. The funds could be allocated to the foundation earlier instead, in which case the charitable contribution is made by the company foundation and not by the company itself.

(16)

15 the news item occurred. In order to really capture the news effect, the event dates are verified on the company’s website or websites of third parties (NGOs for example).

To be able to assess the impact of the news on the market value of the firm we conduct an event study. Our measure of CFP is therefore abnormal return, which will be explained below. For each event the impact of the news on the financial return will be examined. After the exact event date is determined, the historical relationship of the companies’ return with the return of a market index will be determined within an estimation window. During the event window, which consists of several days surrounding the event, actual returns will be compared to the returns that normally would occur (i.e. that were predicted based on its historical behavior). Predictions can be made to calculate the abnormal returns by constructing a model; the market model is commonly used (MacKinlay, 1997). The rate of returns on the share price of firm i on day t is expressed as:

( Equation 1)

where is the rate of returns on the share price of firm i on day t, is the rate of returns on a market portfolio of stocks on day t. The parameters and are the constant and the systematic risk of stock i, respectively, and is the error term. The parameters of this simple market factor model are estimated by Ordinary Least Squares (OLS) regression.

Calculation of the abnormal returns in the event window is as follows: (Equation 2)

where and are the estimates obtained from the regression of on over an estimation period preceding the event window. In addition to the simple market factor model, the model will be enhanced by the return of a sector index (called model 2) a two-factor model. The third model (model 3), a three-factor model, will consist of the return of a market index, the return of a sector index, and the return of a interest rate.

(17)

16 that could arise when insiders were already aware of the news prior to the event date, or contrary when markets respond later due to inefficiency.

Lagrange Multiplier tests for Auto Regressive Conditional Heteroskedasticity (ARCH) showed that ARCH effects were present in the residuals after OLS regression under at least one of the three factor models mentioned earlier. For 38% of the events the assumption that no ARCH effects are present was rejected at a 15% significance level. The three factor models described above will therefore be adjusted to ARCH models as well, so that the conditional variance can be modeled. We admit that Generalized ARCH (GARCH) models could reduce the prediction error term even further. Nonetheless, Hwang and Valls Pareira (2006) propose at least 500 observations for GARCH(1,1) models and 250 for ARCH(1) models (the 1 between parenthesis refers to the number of lags). Hwang and Valls Pareira (2006) found that for these numbers of observation, size of biases, and convergence errors are acceptable. Consequently in this event study an ARCH(1) model is preferred, otherwise the estimation window would be too large and would dramatically reduce the number of useable events.

Returns defined by means of an ARCH(1) model are obtained by the expression: (Equation 3)

Where

(Equation 4) and note that is independent and identically distributed (i.i.d)

The conditional variance is represented as:

(Equation 5)

(18)

17 Once the parameters are estimated, in the event window the error ( ) will coincide with ( ), since we assume that the market factor returns have a zero mean. See Resti and Sironi (2007). The conditional variance in the event window is represented as:

(Equation 6)

The abnormal returns will be tested by the Student’s t test to see whether they differ significantly from zero for each day in the event window. Nevertheless this test requires that the abnormal returns are normally distributed. For this matter a Corrado Rank test is conducted as well (McWilliams and Siegel, 1997). Furthermore, some stocks under study do not show perfect liquidity (see table A.2 in the appendix). Bartholdy et al. (2007) recommend that in such cases results of non-parametrical tests should be emphasized. Cumulative abnormal returns (CAR) will be tested, whether they differ significantly from zero on different time intervals in the event window, by Student’s t test only. Since the Corrado Rank test was originally constructed for an event window of one day (Corrado, 1989), the suitability of the use of the Corrado Rank test for testing CAR of large time intervals can be questioned.7

5 Data

Daily coverage of news in the LexisNexis database for all the countries under study started no earlier than August 19, 2002. A list of included news sources can be found in table A.3 in the appendix. Financial return data is obtained from Thomson DataStream for the period August 19, 2002 till December 31, 2008. Firms need to be a constituent of either one of the following indices: the Argentine Bolsa-G, Brazilian Bovespa, Chilean IPSA or the Mexican IPC. The interest rate and country indices used can be found in table A.4 of the appendix. Days when the particular stock exchange was closed were eliminated from the sample.

7

(19)

18 Descriptive statistics of the daily returns of the included companies can be found in table A.5 in the appendix. The mean average daily returns vary between 0.02% and 0.214%. The standard deviations of 1.29% to 3.24% reflect common daily fluctuations. The median is for a large number of companies 0% (especially Argentine and Chilean companies), caused by days at which no trading in that particular stock occurred and indicates illiquidity. Skewness varies between -0.048 and 0.945. Except for Telecom Argentina and Vale do Rio Doce, the daily returns of the companies are right-skewed. For 29 companies, skewness of the daily returns is greater than 0.155. Given the number of observations (N) and that skewness is greater than 0.155 for most companies, using the Jarque-Bera statistic, it is likely that the daily returns are not normally distributed (Jarque and Bera, 1980).

None of the companies outperformed their respective indices, since Alpha (α) for all companies is zero. The daily returns of the majority of the companies (29 of a total of 38 companies) show a less than perfect correlation with their country index, for just nine companies the daily returns are equally (more) volatile as (than) the daily returns of the country index. Correlations with interest rates are low (around zero), while correlations with world sector indices show a mixed picture and vary between -0.36 to 0.84.

The search for events that suited our definition resulted in 80 events. For four events the first listing day of the firm was during the estimation window and were therefore excluded. Removing events that had other events of the same firm in the estimation or event window led to exclusion of another fourteen events. Three events are excluded because the beginning of the estimation period is before the start of daily newspaper coverage (August 19, 2002) and event clustering cannot be ruled out. Nine events of stocks that were medium or thinly traded during their estimation or event window are removed from the sample as well. Stocks of the resulting events meet the classification criteria of Bartholdy et al. (2007) to be regarded as a thickly traded stock on a small stock exchange, i.e. trading occurs on more than eighty percent of the trading days. Again, see table A.2 for the degree of illiquidity still left. Finally, after the filtering process 50 usable events were left to incorporate in our analysis.

(20)

19

Table 5.1

Events differentiated by country and year

Presents the distribution of events among countries and year. Note that first day of our analysis is August 19, 2002. Country 2002 2003 2004 2005 2006 2007 2008 Total Argentina 0 0 8 0 1 0 1 10 Brazil 0 1 4 2 9 1 6 23 Chile 0 1 0 0 1 1 2 5 Mexico 0 0 2 5 1 2 2 12 Total 0 2 14 7 12 4 11 50 Table 5.2

Events differentiated per CSR category and country

Presents the distribution of events among CSR categories and countries.

CSR Category Argentina Brazil Chile Mexico Total

Education 0 1 1 1 3

Environment 2 9 2 4 17

Social* 1 4 1 2 8

Both Environment and Social ~ 0 8 1 0 9

Broad CSR initiative^ 7 1 0 5 13

Total 10 23 5 12 50

* Includes a good practice code for suppliers and an anti-corruption pact

~ Includes Equator principles

^ UN Global Compact and introduction of a charity foundation

(21)

20

Table 5.3

Events differentiated per form of CSR action and country

Presents the distribution of events among forms of CSR action and countries.

Form of CSR action Argentina Brazil Chile Mexico Total

Donation 2 6 1 5 14

Environmental investment 1 3 1 0 5

Pact, treaty or code 7 12 3 5 27

Product improvement 0 0 0 2 2

Both donation and product* 0 2 0 0 2

Total 10 23 5 12 50

* Includes a 25% discount at Banco Itau for financing sustainable projects

As can be derived from table 5.3 most events used in our analysis are in the form of a pact, treaty or code; 27 events. Donations (fourteen events) also make up a substantial portion of the events. To determine to what form of CSR action an event pertains is relatively easy, to determine in which CSR category the event should fall is harder. Especially pacts and treaties can cover a wide array of CSR subjects and therefore make up almost all events that fall into the categories ‘Both Environment and Social’ and ‘Broad CSR initiative’. For events per category and country see table 5.2. In the appendix tables can be found in which CSR categories and the form of CSR actions are differentiated by year (respectively table A.7 and A.8 ).

Dates of the first and last event per country can be found in table A.9. During our research period only in Argentina capital controls were in place. However, the first event of an Argentine company dates April 1, 2004 and is well after a period of capital control that lasted till May 2003. Stronger restrictions were already lifted in December 2002 and the cross-market premium was largely reduced before (Levy Yeyati et al., 2009).

6 Results

(22)

21 could be that on the particular days confounding news not anticipated by the author resulted in the statistically significant abnormal returns.

The abnormal returns obtained by the models regressed by OLS yielded slightly different results (found in the appendix in table A.10), with more abnormal returns statistically significant.

Table 6.1

Abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for the whole sample. Auto-Regressive Conditional

Heteroskedasticy (ARCH) models

Presents the abnormal returns (AR) for each day around the event date for the ARCH models, for which the parameters were estimated by conditional maximum likelihood. AR values are calculated using factor models: Ri =α + βRm +ε, where Ri is company’s i’s return and Rm a vector of market indices ( the underlying

factors). The conditional variance: hit = ci + ∑λij ε2it-j. Model 1 is a one-factor model using a local index,

model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. T-values of the student’s t-statistic and Corrado Rank t- statistic. Source: Datastream.

***,**,* indicates significance at the 1%, 5% and 10% level respectively. ARCH models. All events (N = 50)

Model 1 Model 2 Model 3

t-value t-value t-value

(23)

22

Table 6.2

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for the whole sample.

Auto-Regressive Conditional Heteroskedasticity (ARCH) models

Presents the cumulative abnormal returns (CAR) in different event windows (in days) for the ARCH models, for which the parameters were estimated by conditional maximum likelihood. AR values are calculated using factor models: Ri =α + βRm +ε, where Ri is company’s i’s return and Rm a vector of market

indices ( the underlying factors). The conditional variance: hit = ci + ∑λij ε2it-j. Model 1 is a one-factor model

using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. T-values of the student’s t-statistic. Source: Datastream. ***,**,* indicates significance at the 1%, 5% and 10% level respectively.

ARCH models. All events (N = 50)

Event window CAR (%) Student CAR (%) Student CAR (%) Student

[0] -0.05 -0.195 0.08 0.357 0.09 0.377 [-1,1] -0.75 *-1.984 -0.60 -1.624 -0.57 -1.549 [-2,2] -1.24 **-2.487 -1.09 **-2.247 -1.07 **-2.193 [-5,5] -0.94 -1.474 -1.00 -1.666 -0.91 -1.501 [-10,10] -1.97 *-1.965 -1.84 *-1.876 -1.74 *-1.748 Table 6.3

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for the whole sample. Linear models,

regression by OLS.

Presents the cumulative abnormal returns (CAR) in different event windows for the 3 models, for which the parameters were estimated by OLS regression.. AR values are calculated using factor models: Ri =α + βRm +ε, where Ri is company’s i’s return and Rm a vector of market indices ( the underlying factors). Model

1 is a one-factor model using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. T-values of the student’s t-statistic. Source: Datastream. ***,**,* indicates significance at the 1%, 5% and 10% level respectively.

Linear models (OLS regression). All events (N = 50)

[0] -0.01 -0.046 0.11 0.468 0.12 0.503

[-1,1] -0.77 **-2.022 -0.62 *-1.679 -0.59 -1.591

[-2,2] -1.34 ***-2.716 -1.17 **-2.445 -1.14 **-2.380

[-5,5] -1.16 *-1.797 -1.21 *-1.981 -1.11 *-1.809

(24)

23 In figure 6.1, a consistently negative trend in CAR, which stays statistically significant from day -1, can be observed. Comparing the results under the two different types of model (tables 6.2 and 6.3); the model types show similar trends, though the results of linear models obtained by OLS regression are statistically even more significant. The null hypothesis of no relationship between CSP and CFP therefore needs to be rejected. Our result of a negative relationship is not in line with the Peinado-Vara (2006) paper, which observed a positive CSP-CFP link. Though remark that she examined the effect on sales for two companies, of which one is not of Latin American origin.

Figure 6.1

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for the whole sample.

Presents the cumulative abnormal returns (CAR) for each day around the event date for the ARCH models, for which the parameters were estimated by conditional maximum likelihood. AR values are calculated using factor models: Ri =α + βRm +ε, where Ri is company’s i’s return and Rm a vector of market

(25)

24

6.1 Cross-listing

In table 6.4 CAR of the events for which the stock was not cross-listed is presented 8. The CAR of the events for which the stock was cross-listed are given in table 6.5 9. Although CAR for both groups show statistically significant results, this does not apply to the same event window.

Table 6.4

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for stocks without cross-listing.

using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. T-values of the student’s t-statistic. Source: Datastream. ***,**,* indicates significance at the 1%, 5% and 10% level

respectively.

ARCH models. Events for which the stock was not cross-listed (N = 10)

[0] -0.31 -0.870 -0.26 -0.753 -0.27 -0.776 [-1,1] -0.95 -1.518 -0.87 -1.399 -0.87 -1.398 [-2,2] -0.99 -1.237 -0.90 -1.117 -0.95 -1.161 [-5,5] -2.46 -1.637 -2.40 -1.540 -2.48 -1.614 [-10,10] -5.51 **-2.327 -5.62 **-2.446 -6.24 **-2.687

8_{The CAR obtained by the models regressed by OLS yielded similar results , except that the CAR in event}

window [-5,5] of Model 1 and 3 are statistically significantly negative at the 10%-level. Available on request from the author.

9

(26)

25 Stock markets’ reaction to CSR news differs between cross-listed stocks and stocks without a cross-listing. However, when the means of the CAR of cross-listed and not cross-listed stocks were compared with a t-test, no statistically significant differences were detected, except for the event window [-10,10], which is statistically significantly different at the 10%-level. Since only one time frame is statistically significantly different, this is no convincing evidence to reject the null hypothesis of the second hypothesis. This difference could easily be created by other news, not anticipated by the author with a disturbing effect on the stock price. Interestingly the CAR of the stocks without cross-listing is, except for event window [-2,2], in all cases lower, which could suggest that investors value (pure) Latin American companies without a cross-listing even lower at CSR news publication than Latin American companies with a listing on a Western stock exchange. Nevertheless the difference is not consistently statistically significant.

Table 6.5

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for stocks with cross-listing.

using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. T-values of the student’s t-statistic. Source: Datastream. ***,**,* indicates significance at the 1%, 5% and 10% level

respectively.

ARCH models. Events for which the stock was cross-listed (N = 40)

[0] 0.02 0.072 0.17 0.605 0.18 0.633

[-1,1] -0.70 -1.557 -0.53 -1.215 -0.50 -1.137

[-2,2] -1.31 **-2.194 -1.14 *-1.975 -1.10 *-1.899

[-5,5] -0.56 -0.800 -0.65 -1.017 -0.51 -0.794

(27)

26

7 Robustness

As a robustness check the CAR of the illiquid stocks (table A.2 in the appendix) and of the liquid stocks are displayed in table 7.110 and 7.211 respectively.

Table 7.1

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for illiquid stocks. Auto-Regressive

Conditional Heteroskedasticity (ARCH) models

using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. For illiquidity, see table A.2 for which events zero return dates occurred. T-values of the student’s t-statistic. Source: Datastream.

***,**,* indicates significance at the 1%, 5% and 10% level respectively. ARCH models. Events of illiquid stocks (N = 19)

[0] -0.57 -1.705 -0.48 -1.507 -0.47 -1.485 [-1,1] -1.04 -1.519 -0.88 -1.425 -0.84 -1.338 [-2,2] -0.38 -0.494 -0.29 -0.448 -0.27 -0.411 [-5,5] 0.13 0.099 -0.12 -0.108 -0.06 -0.055 [-10,10] -0.21 -0.156 -0.29 -0.255 -0.23 -0.194

10_{The CAR obtained by the models regressed by OLS yielded similar results, being no CAR statistically}

significant. Available on request from the author.

11

(28)

27

Table 7.2

Cumulative abnormal returns found by our event study after the effect of Corporate Social Responsibility actions with a positive social output aim, for liquid stocks. Auto-Regressive

Conditional Heteroskedasticity (ARCH) models

using a local index, model 2 is a two-factor model using a local index and global sector index and model 3 is a three-factor model using a local index, a global sector index, and a local interest rate. Estimation window: [-270,-20]. See tables A.4 and A.5 for the indices and interest rates used. For illiquidity, see table A.2 for which events zero return dates occurred. T-values of the student’s t-statistic. Source: Datastream.

***,**,* indicates significance at the 1%, 5% and 10% level respectively. ARCH models. Events of liquid stocks (N = 31)

[0] 0.274 0.881 0.432 1.367 0.435 1.379

[-1,1] -0.567 -1.274 -0.421 -0.916 -0.408 -0.886

[-2,2] -1.773 **-2.748 -1.585 **-2.370 -1.558 **-2.326 [-5,5] -1.593 **-2.444 -1.540 **-2.323 -1.424 **-2.149 [-10,10] -3.057 **-2.217 -2.794 *-1.985 -2.665 *-1.880

(29)

28

8 Conclusion

The results show that investors react negatively to CSR actions of Latin American companies. We found a statistically significantly negative CSP-CFP relationship for stock listed Latin American companies in the period August 19, 2002 till the end of 2008. However, interestingly we found no convincing evidence to reject our second hypothesis that the stock market’s reaction to CSR news differs between companies with a cross-listing on a Western stock market and companies without.

Companies with an ADR or GDR are likely under more scrutiny of Western investors, than companies without a cross-listing. Due to lower transaction costs, better investor protection and higher liquidity, foreign investors will probably make up a larger part of the investor public of Latin American stocks with a cross-listing than of Latin American stocks without a cross-listing. While most recent literature studies point at a positive CSP-CFP relationship in Western markets, our study shows the contrary for the Latin American market.

Since we found no statistically significant difference in investors’ reaction to CSR news of companies with a cross-listing on a Western stock market and companies without, we conclude that the found negative CSP-CFP relationship cannot be explained by different investor groups (Latin American versus Western investor).

(30)

29

9 Limitations and Recommendations

The great portion of illiquid stocks in the Latin American stock market limited the number of companies usable for study. Therefore our research did not control for differences between industries. Since an industry could face specific problems in a given social area, that another industry does not experience at all, industry effects may have blurred the result.

Further research could be undertaken to investigate the CSP-CFP link in other regions of the world. Ideally this would cover well developed financial markets, to make controlling for differences between industries possible. Another limitation of our research is that our measure of CSP does not consist of a homogenous group of CSR actions. Investors could value the different types of CSR actions used in our research differently. Consequently, the found CSP-CFP relationship may not apply to the all the different types of CSR actions used in this study.

References

Abbott, W.F., and Monsen, R. J. (1979) On the measurement of corporate social responsibility: Self-reported disclosures as a method of measuring corporate social involvement.

Academy of Management Journal, 22(3), 501-515

Aggarwal, R., Dahiya, S., and Klapper, L. (2007) ADR holdings of US-based emerging market funds. Journal of Banking & Finance, 31(6), 1649-1667

Agle, B.R., Mitchell, R.K., and Sonnenfeld, J.A. (1999) Who Matters to CEOs? An Investigation of Stakeholder Attributes and Salience, Corporate Performance, and CEO Values. The

Alexander, G.J., and Buchholz, R.A. (1978) Corporate social responsibility and stock market performance. Academy of Management Journal, 21(3), 479-486

(31)

30 Aupperle, K.E., Carroll, A.B. and Hatfield, J.D. (1985) An empirical examination of the relationship between corporate social responsibility and profitability. Academy of

Management Journal, 28(2), 446-463

Austin, J., Refico, E. et al. (2004) Social Partnering in Latin America: Lessons Drawn From

Collaborations of Businesses and Civil Society Organizations (Harvard University Press,

Cambridge, MA), p. 332

Balabanis, G., Philips, H.C., and Lyall, J. (1998) Corporate social responsibility and economic performance in the top British companies: are they linked? European Business Review, 98(1), 25–42

Bartholdy, J., Olson, D., and Peare, P. (2007) Conducting Event Studies on a Small Stock Exchange The European Journal of Finance, 13(3), 227-252

Barnett, M.L., and Salomon, R.M. (2006) Beyond Dichotomy: The Curvilinear Relationship Between Social Responsibility and Financial Performance. Strategic Management

Journal, 27(11), 1101–1156

Baucus, M.S., and Baucus, D.A. (1997) Paying the Piper: An Empirical Examination of Longer-Term Financial Consequences of Illegal Corporate Behavior. The Academy of

Belkaoui, A. (1976) The impact of the disclosure of the environmental effects of organizational behavior on the market. Financial Management, 5(4), 26-31

Belkaoui, A., and Karpik, P.G. (1989) Determinants of the corporate decision to disclose social information. Accounting, Auditing and Accountability Journal, 2(1), 36-51

Berman, S.L., Wicks, A.C., Kotha, S., and 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

(32)

31 Blacconiere, W.G., and Northcut, W.D. (1997) Environmental information and market reactions to environmental legislation. Journal of Accounting, Auditing and Finance, 12(2), 149-178

Blacconiere, W.G., and Patten, D.M. (1994) Environmental disclosures, regulatory costs, and changes in firm value. Journal of Accounting and Economics, 18 (3), 357–377

Bowman, E.H. (1978) Strategy, annual reports, and alchemy. California Management Review, 20(3), 64–71

Bowman, E., and Haire, M. (1975) A Strategic Posture Toward Corporate Social Responsibility.

California Management Review, 18(2), 49-58

Boyle, E.J., Higgins, M.M., and Rhee, S.G. (1997) Stock market reaction to ethical initiatives of defense contractors: theory and evidence. Critical perspectives on accounting, 8(6), 541–561

Bragdon, J.H., Jr., and Karash, R. (2002) Living-asset stewardship: How organizational learning leads to exceptional market returns. Reflections, 4(1), 55-65

Brammer, S., Brooks, C., and Pavelin, S. (2006) Corporate Social Performance and Stock Returns UK Evidence from Disaggregate Measures. Financial Management, 35(3), 97–116

Brown, B. (1997) Stock market valuation of reputation for corporate social performance

Corporate Reputation Review, 1(1-2), 76-80

Brown, B. (1998). Do Stock Market Investors Reward Companies with Reputations for Social Performance? Corporate Reputation Review, 1(3), 271–280

Brown, B., and Perry, S. (1994) Removing the financial performance halo from Fortune's 'most admired' companies. Academy of Management Journal, 37(5), 1347-1359

Buehler, V.M., and Shetty, Y.K. (1976) Managerial response to social responsibility challenge.

Carroll, A. B. (1979) A three-dimensional conceptual model of corporate performance. Academy

(33)

32

Carroll, A.B. (1999) Corporate Social Responsibility: Evolution of a Definitional Construct.

Business and Society, 38(3), 268-295

Carter, C.R., Kale, R., and Grimm, C.M. (2000) Environmental Purchasing and firm performance: an empirical investigation. Transportation Research, 36(1), 219–228

Chen, K.H., and Metcalf, R.W. (1980) The relationship between pollution control record and financial indicators revisited. Accounting Review, 55(1), 168-177

Christmann, P. (2000) Effects of 'best practices' of environmental management on cost advantage: The role of complementary assets. Academy of Management Journal, 43(4), 663-680

Cochran, P. L., and Wood, R.A. (1984) Corporate social responsibility and financial performance.

Academy of Management Journal, 27(1), 42-56.

Corrado, C.J. (1989) A nonparametric test for abnormal security-price performance in event studies Journal of Financial Economics, 23(2), 385-395

Cottrill, M.T. (1990) Corporate social responsibility and the marketplace. Journal of Business

Ethics, 9(9), 723-729

Cowen, S.S., Ferreri, L.B., and Parker, L.D. (1987) The impact of corporate characteristics on social responsibility disclosure: A typology and frequency-based analysis. Accounting,

Organizations and Society, 12(2), 111-222

Davidson, W.N. III, and Worrell, D.L. (1992) Research notes and communications: The effect of product recall announcements on shareholder wealth. Strategic Management Journal, 13(6), 467–473

Diltz, D.J. (1995) The private cost of socially responsible investing. Applied Financial Economics, 5(2), 69-77

(34)

33 Dooley, R.S., and Lerner, L.D. (1994) Pollution, profits, and stakeholders: The constraining effect of economic performance on CEO concern with stakeholder expectations. Journal

of Business Ethics, 13(9), 701-711

Dowell, G., Hart, S., and Yeung, B. (2000) Do corporate global environment standards create or destroy market value? Management Science, 46(8), 1059–1074

Engle, R. (1982) Autoregressive conditional heteroskedasticity with estimates of the variance of U.K. inflation. Econometrica, 50(4), 987-1007

Epstein, M. J., and Schnietz, K. E. (2002) Measuring the cost of environmental and labor protests to globalization: An event study of the failed 1999 Seattle WTO talks. International

Trade Journal, 16(2), 129-160

Fogler, H.R., and Nutt, F. (1975) A note on social responsibility and stock valuation. Academy of

Fombrun, C., and Shanley, M. (1990) What's in a name? Reputation building and corporate strategy. Academy of Management Journal, 33(2), 233-258

Freedman, M., and Jaggi, B. (1982) Pollution disclosures, pollution performance and economic performance. Omega, 10(2), 167-176

Freedman, M. and Stagliano, A.J. (1991) Differences in social-cost disclosures: A market test of investor reactions. Accounting, Auditing and Accountability Journal, 4(1), 68–83

Friedman, M. (13 September 1970). 'Social responsibility of business', New York Times Magazine

Fry, F.L., and Hock, R.J. (1976) Who claims corporate responsibility? The biggest and the worst.

Business and Society Review, (18), 62-65

Fry, L.W., Keim, G.D., and Meiners, R.E. (1982) Corporate contributions: Altruistic or for profit?

(35)

34

Goll, I., and Rasheed, A.A., (2004) The Moderating Environmental Munificence and Dynamism on the Relationship Between Discretionary Social Responsibility and Firm Performance.

Journal of Business Ethics, 49(1), 41–54

Graves, S.B., and Waddock, S.A. (1994) Institutional owners and corporate social performance.

Graves, S.B., and Waddock, S.A. (1999) A look at the Financial-Social Performance Nexus when Quality of Management is Held Constant. International Journal of Value-Based

Management, 12(1), 87–99

Graves, S.B., and Waddock, S.A. (2000) Beyond built to last ... Stakeholder relations in 'built- to-last' companies. Business and Society Review, 105(4), 393-418

Greening, D.W. (1995) Conservation strategies, firm performance, and corporate reputation in the U.S. electric utility industry. Research in Corporate Social Performance and Policy, Supplement 1: 345-368

Gregory, A., Matatko, J., and Luther, R. (1997) Ethical unit trust financial performance: Small company effects and fund size effects. Journal of Business Finance and Accounting, 24(5), 705-725

Grossman, G.E., and Krueger, A.B., (1995) Economic Growth and the Environment, The Quarterly

Journal of Economics, 110(2), 353–377

Guerard, J.B. Jr. (1997) Is there a cost to being socially responsible in investing? Journal of

Investing, 6(2), 11–18

Hamilton, S., Jo, H., and Statman, M. (1993) Doing well while doing good? The investment performance of socially responsible mutual funds. Financial Analysts Journal, 49(6), 62-66

(36)

35

Hart, S.L., and Ahuja, G. (1996) Does it pay to be green? An empirical examination of the relationship between emission reduction and firm performance. Business Strategy and

the Environment, 5(1), 30–37

He, Y., Tian, Z., and Chen, Y. (2007) Performance implications of nonmarket strategy in China.

Asia Pacific Journal of Management, 24(2), 151–169

Herremans, I.M., Akathaporn, P., and McInnes, M. (1993) An Investigation of Corporate Social Responsibility Reputation and Economic Performance. Accounting, Organizations and

Society, 18(7/8), 587–604

Hickman, K.A., Teets, W.R. and Kohls, J.J. (1999) Social investing and modern portfolio theory.

American Business Review, 17(1), 72-78

Hillman, A.J., and Keim, G.D. (2001) Shareholder value, stakeholder management, and social issues: What's the bottom line? Strategic Management Journal, 22(2), 125-139

Hwang, S., and Valls Pereira, P.L. (2006) Small Sample Properties of GARCH Estimates and Persistence. The European Journal of Finance, 12(6-7): 473-494

Hylton, M.O. (1992) Socially responsible' investing: Doing good versus doing well in an efficient market. American University Law Review, 42(1) 1-52.

Ingram, R.W. (1978) An investigation of the information content of (certain) social responsibility disclosures. Journal of Accounting Research, 16(2), 270-285

Jamali, D., Sidani, Y., and El-Asmar, K. (2009) A Three Country Comparative Analysis of Managerial CSR Perspectives: Insights From Lebanon, Syria and Jordan. Journal of

Business Ethics, 85(2), 173-192

Jarque, C.M. and Bera, A.K. (1980) Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Economic Letters, 6(3), 255-259

(37)

36 Jones, R., and Murrell, A.J. (2001) Signaling positive corporate social performance. Business and

Society, 40(1), 59-78

Judge, W.Q. Jr. (1994) Correlates of organizational effectiveness: A multilevel analysis of a multidimensional outcome. Journal of Business Ethics, 13(1), 1–10

Judge, W.Q. Jr., and Douglas, T.J. (1998) Performance implications of incorporating natural environmental issues into the strategic planning process: An empirical assessment.

Journal of Management Studies, 35(2), 241–262

Kedia, B.L., and Kuntz, E.C. (1981) The context of social performance: An empirical study of Texas banks. Research in Corporate Social Performance and Policy, 3, 133-154

Klassen, R.D., and McLaughlin, C.P. (1996). The impact of environmental management on firm performance. Management Science, 42(8), 1199–1214

Klassen, R.D., and Whybark, D.C. (1999) The impact of environmental technologies on manufacturing performance. Academy of Management Journal, 42(6), 599-615.

Kolari, J., and Pynnonen, S., (2008) Generalized Rank Test for Testing Cumulative Abnormal Returns in Event Studies, 21st Australasian Finance and Banking Conference 2008 Paper, draft November 25, 2008 Accessed via: http://ssrn.com/abstract=1254022

Konar, S., and Cohen, M.A. (1997) Information as regulation: The effect of community right to know laws on toxic emissions. Journal of Environmental Economics and Management, 32(1), 109-124

Konar, S., and Cohen, M.A. (2001) Does the market value environmental performance? Review

of Economics and Statistics, 83(2), 281-289

Kumar, R., Lamb, W.B. and Wokutch, R.E. (2002) The End of South African Sanctions, Institutional Ownership, and the Stock Price Performance of Boycotted Firms, Business

and Society, 41(2), 133–165

(38)

37

Levy Yeyati, E., Schmukler, S., and Van Horen, N. (2009) International financial integration through the law of one price: The role of liquidity and capital controls. Journal of

Financial Intermediation, 18(3), 432-463

Luo, X., and Bhattacharya, C.B. (2006) Corporate Social Responsibility, Customer Satisfaction, and Market Value. Journal of Marketing, 70(1), 1–18

Luther, R.G., and Matatko, J. (1994) The performance of ethical unit trusts: Choosing an appropriate benchmark. British Accounting Review, 26(1), 77-89

Luther, R.G., Matatko, J., and Corner, D.C. (1992) The investment performance of U.K. 'ethical' unit trusts. Accounting, Auditing and Accountability Journal, 5(4), 57-70

MacKinlay, A.C. (1997) Event Studies in Economics and Finance. Journal of Economic Literature, 35(1), 13-39

Mahapatra, S. (1984) Investor reaction to a corporate social accounting. Journal of Business

Finance and Accounting, 11(1), 29-40

Mallin, C.A., Saadouni, B. and Briston, R.J. (1995) The financial performance of ethical investment funds. Journal of Business Finance and Accounting, 22(4), 483-496

Margolis, J.D., and Walsh, J.P. (2003) Misery Loves Companies: Rethinking Social Initiatives by Business. Administrative Science Quarterly, 48(2), 268–305

Maron, I.Y. (2006) Toward a Unified Theory of the CSP–CFP Link. Journal of Business Ethics, 67(2), 191–200

Maslow, A.H. (1943) A Theory of Human Motivation. Psychological Review, 50(4), 370-396 McGuire, J.B., Schneeweis, T., and Branch, B. (1990) Perceptions of firm quality: A cause or

result of firm performance. Journal of Management, 16(1), 167-180