Sustainability reporting quality and firm value : does industry matter?

(1)

Amsterdam Business School

Master Thesis

Sustainability reporting quality and firm value:

does industry matter?

Student: Aikebaijiang Shawuti

Student Number: 10605029

Supervisor: Brendan O'Dwyer

Date: August 15, 2014

Study Program: MSc Accountancy& Control, Control track Field of Research: Social Accounting

(2)

(3)

Abstract

Previous empirical evidence provides mixed results on the relationship between corporate social performance and its financial performance. This paper revisits this relationship by testing social reporting quality’s impact on future firm value based on the predictions driven from voluntary disclosure using more comprehensive research design. This paper uses a ranking score from KLD database, and using independent third party resource is less likely to be biased rather than conducting a content analysis.

This paper documents that cost of capital negatively associates with the social reporting quality, indicating that increasing social reporting quality can lower the cost of capital. However, this paper also provides evidence that future cash flow is also negatively related to the social reporting quality and this might result from low public visibility.

This paper also examines whether social reporting quality’s impact on future performance varies from sector to sector. However, sustainability reporting quality neither significantly associates with the cost of capital not the future cash flow in the more environmentally sensitive sectors (i.e. energy, utilities and materials).

(5)

1 Introduction

1.1 Background

There is an increasing demand for sustainability information disclosure, mainly voluntary disclosure to broader stakeholders of a firm, and there are some regulations towards disclosure of a firm’s environmental impact in some countries. Furthermore, international efforts to develop common principles guiding the production of sustainability reports, which demonstrate the relation of a firm’s social and environmental behavior and its organizational strategy, to help firms to make more sustainability reports as well as help stakeholders to fully understand how a firm performs. All of those facts mentioned above highlighted the importance of our understanding about suitability performance of a firm.

Recent years, there are numbers of researches undertaken on sustainability area. They mainly focus on following specific topics: i) investigation of the motivation of making sustainability reports from firm’s perspective, e.g. to increase corporate transparency, to legitimize the firm for the long-term and to address stakeholders’ demand; ii) assessment of sustainability, including quantity, such as how many pages or words sustainability reports consists annual reports of a firm, and quality by designing models to evaluate the quality of sustainability reports, for example, Clarkson Index; iii) examination of the relations between firm value and sustainability. Those limited existing literature tried to find out if increasing sustainability reports quality would increase firm value correspondingly, and they basically examined its impact on cost of capital and operational performance.

1.2 Research question and motivation

There has been an increasing demand towards quality and quantity of environmental and social disclosure from investors, market regulators as well as the broader stakeholders. Correspondingly, 95% of world’s largest companies issued sustainability reports by 2011, while there were only 35% of those companies making such reports by 1999 (KPMG, 2002; KPMG, 2011). While a trend towards more frequent and higher-quality sustainability disclosure have been found in many countries, relatively little evidence exists as to the impact of such disclosures have on financial markets (Gamble et al.1996). Therefore, the main purpose of this research is to reexamine the impact of sustainability reporting quality on firms’ financial

(6)

performance. More importantly, many researches, mostly examined such relationship in more environmentally and socially sensitive industries (Pulp and paper, Chemicals, Metals and Mining, Oil and gas, and Utilities) documented there is no significant relationship or a negative relationship between financial performance of a firm and its sustainability disclosure quality (Pattern et al.2002, Clarkson et al. 2010). These findings are inconsistent with the results that have a broader scope of sample in terms of the environmental sensitivity. Therefore, the sensitivity of industries would also be an essential concern of this paper.

Hence the main question of this research is What impact do sustainability

reporting quality have on firms’ financial performance and does such impact differ in different industries?

Many researchers also highlighted the importance of studying the possible impacts of sustainability disclosure on firms’ financial performance. Establishing the usefulness of transparent voluntary environmental disclosure and the source of this usefulness is of fundamental importance for corporate social responsibility practitioners. Their task is to convince top management that such transparency is worthwhile in terms of enhancing firm value (Clarkson et al. 2011). Murray et al. (2006) also stated that ‘if further evidence could be gathered to suggest that markets can be persuaded to start to see the social and environmental implications of their financial decisions then a practical case can be added to the moral case that substantive environmental disclosure needs to become a regular, significant and regulated part of corporate disclosure’ in their research.

There has been some research on how environmental disclosure will affect on firm’s environmental performance (Clarkson et al., 2008). Some other research investigated how environmental disclosure would affect firm’s financial performance (Plumlee et al., 2010). However, there is very little empirical study on how overall sustainability performance would directly impact a firm’s operational behavior, eventually its value (Wilson et al., 2013). Moreover, most of the existing literature on this specific topic showed mixed results. For example, Plumlee et al. (2010) studied the relationship between cost of equity and the quality of firms’ environmental disclosures and the results showed that there is a positive relationship between them. Welker et al. (2001) also reported a positive relationship between social disclosure and cost of capital. In contrast to those findings, many results

(7)

supported that increasing voluntary social and environmental disclosure can significantly lower the cost of capital (Aerts et al. 2008). These mixed results necessitate the further research on this topic to provide more evidence.

Given the fact of increasing demand on more empirical evidence of how the higher-quality sustainability reporting would affect firm performance and mixed findings from the prior literature, I believe it is necessary and interesting to reinvestigate such relationship.

2 Literature review and hypotheses

2.1 Definition of sustainability reporting

Sustainability is a company’s capacity to prosper in a hypercompetitive and changing global business environment. Companies that anticipate and manage current and future economic, environmental and social opportunities and risks by focusing on quality, innovation and productivity will emerge as leaders that are more likely to create a competitive advantage and long-term stakeholder value (Dow Jones Sustainability World Index Guide, Version12.3, 2013). Global Reporting Initiative (GRI 2006) defined sustainability reporting [the practice of measuring, disclosing and being accountable to internal and external stakeholders for organizational performance towards the goal of sustainability development. “Sustainability reporting” is a broad term synonymous with others used to describe reporting on economic, environmental and social impacts.] While this definition encompasses social and environmental reporting, Wilson et al. (2013) expected environmental aspects to dominate any firm value effects. They argued that the majority of large firms’ future cash flows are more obviously sensitive to environmental outcomes, through fines, restoration costs and consumer sentiment. However, their primary measure of sustainability reporting quality potentially reflected all aspects of sustainability reporting while their secondary proxies focused on environmental reporting only. In this research, I will be following the definition from GRI, and will take all aspects of sustainability reporting into account to measure reporting quality.

2.2 Measurement of sustainability reporting quality

Al-Tuwaijri et al. (2004) classified the environmental disclosure measurement techniques into two general groups. The first group includes measures that quantify

(8)

the level of environmental disclosure in the annual report, such as the number of pages, sentences, and words. They also pointed out that such measures have their limitations. For example, while pages may include pictures that have no information on environmental or social activities, sentences and words may ignore necessary graphics and tables. Moreover, they argued that these types of measurement is susceptible to “greenwashing”, in which management outs its best “spin” on what otherwise might be a lackluster environmental performance. The second measurement technique uses a disclosure-scoring measure derived from content analysis. In their research they also developed their own disclosure-scoring methodology based on content analysis that incorporates four key environmental indicators:

• The total amount of toxic waste generated and transferred or recycled • Financial penalties resulting from violations of federal environmental laws • Potential Responsible Party (PRP) designation for the cleanup

responsibility of hazardous-waste sites

• The occurrence of reported oil and chemical spills

Similarly, Clarkson et al. (2008) developed another comprehensive index of measuring sustainability reporting quality. In their methodology, Clarkson et al. (2008) distinguished hard disclosure from soft disclosure. Hard disclosure items, such as quantitative measures of performance, are weighted more heavily than soft disclosures, such as the existence of environmental policy, because the latter is easily mimicked by the firms regardless of disclosure quality. Wilson et al. (2013) argued that Clarkson Index has the advantage of distinguishing between genuine firms that have a good performance and disclosure from those that perform badly and yet disclosure positively.

However, methodologies of measuring sustainability reporting quality are not limited to these two measures as Al-Tuwaijri et al. (2004) suggested. Wilson et al.(2013) argued that disclosure scores necessarily require subjective weighting of items which, however robustly determined, require the researcher to impute the beliefs of a heterogeneous group of stakeholders. Therefore, they did not restrict their analysis to scholar-defined measures of sustainability reporting quality. They used propriety classifications of reporting quality provided by an independent, non-profit Australian research firm: Corporate Analysis Enhanced Responsibility (CAER). They measurement criteria towards reporting quality includes: 1) details of firms’

(9)

sustainability policy are provided; 2) the main impacts/issues in all key area such as energy, emissions, waste, water are described; 3) quantitative data in all key areas, as graphics and tables are provided; 4) measures of performance against targets in all key areas are provided. Moreover, in determining whether these criteria were satisfied, CAER researchers considered the substance of the information reported by companies, and both the quality and quantity of information disclosed (Wilson et al. 2013).

There are also some other popular sustainability rating indexes provided by independent third parties, such as Dow Jones Sustainability Indices (DJSI). DJSI, which launched in 1999, are a family of indexes evaluating the sustainability performance of the largest 2,500 companies listed on the Dow Jones Global Total Stock Market Index. They comprise global and regional benchmarks, and subsets of these indices allow investors to exclude certain industries from performance measurement. DJSI assesses corporate sustainability based on the annual RobecoSAM Questionnaire, which consists of an in-depth analysis featuring approximately 80-120 questions on financially relevant economic, environmental and social factors/dimensions with a focus on companies’ long-term value creation. 1

Table 1 Assessment Criteria of DJSI2

Economic Dimensions Environmental Dimensions Social Dimensions -Corporate governance

-Code of Conduct -Risk& Crisis management

-Consumer relationship management -Innovation management

-Environmental management system -Environmental performance -Climate Strategy

-Product stewardship -Biodiversity

-Human capital development -Talent attraction and retention -Occupational health and safety -Stakeholders Engagement -Social reporting

In this research, the total KLD score of social and environmental strengths of a firm serves as a proxy for sustainability reporting quality, because firms with better social performance have a greater incentive to disclosure (Dye 1985). KLD ranks firm’s CSR performance in seven main categories: (1) community (2) corporate governance (3) diversity (4) employee relations (5) environment (6) human rights and (7) product in a scale of zero to four for each category. KLD database is widely used in the corporate social responsibility research (Servaes and Tomayo 2013; Dhaliwal et al. 2010; Jo et al. 2011).

1_{See: http://www.sustainability-indices.com/sustainability-assessment/corporate-sustainability-assessment.jsp} 2_{See the same webpage above.}

(10)

2.3 Sustainability reporting quality and financial performance

Clarkson et al. (2008) classified existing literature in environmental accounting research into three broad categories: studies that examine the valuation relevance of corporate environmental performance information and has found that such information is valuable to investors seeking to assess corporate environmental liabilities; studies that examine factors affecting managerial decisions towards environmental and social disclosure; literature that explores the relationship between environmental and social disclosure and firm performance, both financially and environmentally. This research contributes directly to the third category.

2.3.1Cost of capital

The voluntary disclosure literature predicts a negative association between the quality of sustainability disclosure and cost of equity capital. While this types pf literature typically uses voluntary financial disclosure as examples, and they believe that the same logic applies to other information relevant to the future financial performance of the firm (Wilson et al. 2013). Clarkson et al. (2011) documented that firms with marked improvement in environmental performance experience significant improvement in financial performance in the subsequent period, and this findings is consistent with the prediction of voluntary disclosure literature. However, Dhaliwal et al. (2010) argued that a straightforward generalization of the cost of capital effect from financial disclosure to non-financial CSR disclosure is not always obvious and there is a common concern about the usefulness of this type of disclosure because of non-comparability and potential credibility issues.

Increasing voluntary disclosure quality basically can influence the financial performance via decreasing information asymmetry and estimated risk. Diamond and Verrecchia (1991) argued that disclosure affects information asymmetry between informed insiders and uninformed investors, and that this information asymmetry is priced. Their research also showed that failure to disclose performance information results in the market assuming a firm to be of low quality because the undisclosed information is likely to be ‘bad news’. Figge (2005) also believes that environmental disclosure might lead to the creation of corporate value as a result of synergies between economic and environmental performance, and explained his view by signalling theory. High-quality firms therefore have an incentive to signal their type by disclosing their performance, which in turn is reflected in a higher price and a

(11)

lower cost of equity capital. Coles et al. (1995) suggested that as uncertainty regarding a firm’s future performance increases, the assumed correlation between that firm’s returns and that of the market increases, and if this risk is not diversifiable, the cost of equity necessarily increases. Connors and Silva-Gao (2009) also suggested that relative environmental performance captures a dimension of firm risk that matters to investors and ultimately affects the cost of equity. Wilson et al. ( 2013) also predicted such mechanism through slightly different perspective: greater disclosure is argued to reduce uncertainty regarding future performance, and thus the cost of equity is reduced.

Empirical research provided substantial support for this theoretical predictions. Wilson et al. (2013) conducted on this specific topic based on Australian companies. They found that quality of sustainability repots has a negative impact on cost of capital while it positively associates with future performance of a firm. Dhaliwal et al. (2010) examined whether the increasing voluntary disclosure of corporate social responsibility (CSR) activities is associated with a reduction in firm’s cost of equity capital. And they found out that firms with a high cost of equity capital in the previous year tend to initiate disclosure of CSR activities in the current year and that initiating firms with superior social responsibility performance enjoy a subsequent reduction in the cost of equity capital.

However, Welker et al. (2001) documented a statistically significant positive relation between the level of social disclosure and the cost of capital, which is consistent with more social disclosure raising the cost of capital, by employing a measure of firms social reporting quality, in which environmental reporting represents about 18% of the total measure. And they argued that many aspects of a firm’s socially responsible behavior may be perceived by the market as negative NPV investments, and that the disclosure of this behavior may increase the perceived risk. Wilson et al. (2013), nevertheless, argued that firms with an inherently low cost of equity capital have less incentive to make high-quality sustainability disclosure, and that this selection bias may have influenced Welker et al.’s (2001) results. After conducting a research for a sample of US firms across five environmentally more sensitive industries, Plumlee et al. (2010) documented a significantly positive relationship between voluntary environmental overall disclosure quality and firm value through both the cash flow and the cost of equity while increasing level of “soft”

(12)

environmental disclosure might decrease the cost of equity. More complex and mixed results tend to be found in Magnan et al.’s (2011) research while studying such relationship in different countries in Europe and North America. Al-Tuwaijri et al. (2004) conjecture that the mixed findings found in prior research may be attributable to the fact that researchers have not jointly considered environmental disclosure, environmental performance, and economic performance. To avoid this issue, this research would take overall impact of sustainability reporting, both social and environmental, into account.

Clarkson et al. (2011) reported that firm’s cost of equity capital has a strong positive association with the level of its relative toxic emissions (TRI) while it is unrelated to the level of voluntary environmental disclosure. They concluded that the investors use actual emissions data such as TRI to assess a firm’s environmental risk and that voluntary environmental disclosure provides no incremental information beyond that contained in the TRI data.

In summation, prior research provides mixed results for the association between environmental disclosure and cost of equity, however, I will be following the findings of Wilson et al. (2013) and Dhaliwal et al. (2010) since these findings are consistent with the prediction of voluntary disclosure theories. Hence, the first hypothesis is:

H1: Sustainability reporting quality is negatively related to the cost of equity.

2.3.2Future cash flow

Most of the research findings in environmental accounting literature focus only on dominator (i.e. cost of equity) rather than numerator (i.e. cash flow) of firm value (Clarkson et al. 2011, Wilson et al. 2013, Plumlee et al. 2009, Dhaliwal et al. 2010). In this research, I will consider both the numerator and dominator component of firm value to extent the prior literature.

Higher-quality sustainability reporting decreases information asymmetry between firm insiders and other stakeholders, and may reduce investors’ uncertainty regarding the firm’s future environmental and social performance, and the cash flows associated with these (Wilson et al., 2013). To examine whether these social and environmental disclosure have real informational contribution to the investors, Magnan et al. (2011) conducted a research in Canada. Their findings suggested that

(13)

social disclosure reinforces the informativeness of environmental disclosure for stock markets, even substituting for it under certain conditions. Stakeholders must assess and retain an increasing flow of information: a more efficient disclosure strategy becomes critical if firms want to convey the right picture of their Corporate Social Responsibility performance.

Voluntary environmental disclosure not only associates with future cash flow through perceived riskiness. Al-Tuwaijri et al. (2004) suggested that environmental disclosure quality serves as signal of firms’ environmental practices that affect financial performance, ultimately, firm value. Clarkson et al. (2010) also argued that quality environmental performers are able to signal their type by releasing high-quality environmental reports, and that this positively affects the market’s expectation of future net cash flows, and therefore, firm value. Furthermore, Lev et al. (2008) also predicted that disclosure related to these issues inform stakeholders beyond investors who opt to patronize or work for environmentally responsible firms, which leads to superior sales, thereby influence the future cash flow. Thus, second hypothesis of this research is:

H2: Higher sustainability reporting quality positively associates with firm value.

2.4 Does the industry matter?

Some research findings suggest that there is a negative association between sustainability disclosure and firms’ sustainability or financial performance. Oddly, almost all of those findings are suggested in the researches that specifically examined such relationship in more environmentally sensitive industries (Pulp and paper, Chemicals, Metals and Mining, Oil and gas, and Utilities). Patten (2002) found a negative relation between environmental disclosure and a toxics release inventory (TRI) based environmental performance indicator. However, these findings are inconsistent with Clarkson (2008) et al.’s findings, which suggested that there is a positive association between environmental performance and the level of discretionary environmental disclosures. After there years, in a different setting, nevertheless, Clarkson et al. (2011) reported a negative relationship between firms’ environmental performance and the quality of voluntary environmental disclosure in more environmentally sensitive industries. Therefore, I believe that it is important to reexamine this relationship in a different setting.

(14)

H3: Higher sustainability reporting quality negatively associates with firm value in more environmentally sensitive industries.

3 Research Design

3.1Measurement of sustainability reporting quality

The difficulty in accessing sustainability performance is well documented in the literature (see, Ilinitch et al., 1998), and many different approaches have been used to measure sustainability reporting quality. I classified them into three broad categories:

The first one is scholar-defined measuring technics. These measures capture variation in voluntary sustainability disclosure quality across firms based on specific, line-by-line, voluntary disclosure (Plumlee et al., 2010). Many early papers focusing specifically on environmental disclosures adopt a variant of the Wiseman (1982) index (Patten, 2002; Bewley and Li, 2000;), which allocates a score to various disclosures according to specificity of disclosure and whether quantitative data is reported. Cormier et al. (2009) contend that the Wiseman Index is a good proxy for environmental disclosure because it is a comprehensive measure that allows information of various types to be integrated into a single comparable figure. To seek a stronger measure, Clarkson et al. (2008) developed another comprehensive index of measuring sustainability reporting quality. In their methodology, Clarkson et al. (2008) distinguished hard disclosure from soft disclosure. Hard disclosure items, such as quantitative measures of performance, are weighted more heavily than soft disclosures, such as the existence of environmental policy, because the latter is easily mimicked by the firms regardless of disclosure quality. Wilson et al. (2013) argued that Clarkson Index has the advantage of distinguishing between genuine firms that have a good performance and disclosure from those that perform badly and yet disclosure positively, and many recent papers adopt Clarkson Index (Clarkson et al. 2010; Plumlee et al. 2010;). However, I believe that disclosure scores necessarily require subjective weighting of items, which might raise doubt regarding the reliability of one’s research. Moreover, while developing such measuring technics, different researchers take different aspects of sustainability disclosure into account, meaning that each method is different, and this might increase the comparability of different outcomes on this field.

(15)

The second one is evaluation of reporting quality by an independent third party, usually non-profit research or consulting firms. For example, Wilson et al. (2013) use propriety data obtained from a specialist responsible investment research firm, Corporate Analysis Enhanced Responsibility (CAER), an organization analyses corporations’ statutory annual reports, websites and stand-alone disclosures to advise subscribers interested in applying environmental, social and governance criteria to their investment activity. However, such organizations are not common and hard to get access into.

The last one is using popular rating indexes. For example, the KLD database provides detailed corporate social responsibility performance rating for individual firms and such ratings are adopted in many previous researches (Servaes and Tomayo 2013; Dhaliwal et al. 2010; Jo et al. 2011). Another popular rating index is DJSI, which assesses corporate sustainability based on the annual RobecoSAM Questionnaire, which consists of an in-depth analysis featuring approximately 80-120 questions on financially relevant economic, environmental and social factors/dimensions with a focus on companies’ long-term value creation. Wilson et al. (2013) argued that recent researches heavily relied on author-constructed measures of quality (Clarkson et al. 2008, 2010; Plumlee et al. 2007, 2010), and they believed that using index enable researchers better examine the impact of reporting quality as it is observed.

After considering pros and cons of each approach, I decided to use total strengths of each company in KLD database as a proxy for social reporting quality because I assume that firms with better social performance have greater incentive to disclosure (Dye 1985; Dhaliwal 2010).

3.2 Cost of Capital

As elaborated in the hypothesis development, prior theoretical and empirical research provides mixed evidence regarding the association between voluntary sustainability disclosure and cost of capital. Traditional economic theory, e.g., voluntary disclosure theory, predicts that increased voluntary disclosure will be associated with a decrease in cost of capital through a reduction in information asymmetry or estimated risk (Verrecchia 1991; Wilson et al. 2013;).

To test H1, my research focuses on the relationship between the denominator firm value, cost of capital, and voluntary sustainability disclosure quality.

(16)

3.2.1 Measurement of cost of capital

There are many different methods to calculate cost of capital such as Dividend Discount Model (DDM), CAPM. However, I have noticed that most of the well cited literature used more advanced methods and I will briefly explain the most common methods to calculate cost of capital in social accounting literatures below:

Target price method. This method utilize Value Line forecasts of target price

(the average of the minimum and maximum forecasted firm stock price 3-5 years in the future) as the primary component of future cash flows along with forecasts of dividend payouts and current stock price to derive an implied cost of capital.

Plumlee et al. (2010) calculate cost of capital using the value line forecasts of target prices made in the third calendar quarter after the voluntary environmental disclosure data is collected. Botosan and Plumlee (2002) also used such method in their research by the equation given below:

𝑃𝑡 = ∑ (1 + 𝑟)4𝜏=1 −𝜏𝐸𝑡[𝑑𝑡+𝜏] + (1 + 𝑟)−4𝑃4 (1)

Where: 𝑃_𝑡= 𝑝𝑟𝑖𝑐𝑒 𝑎𝑡 𝑑𝑎𝑡𝑒 𝑡

𝑟 = 𝑡ℎ𝑒 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑐𝑜𝑠𝑡 𝑜𝑓 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 𝐸𝑡(𝑜) = 𝑡ℎ𝑒 𝑒𝑥𝑝𝑒𝑐𝑡𝑎𝑡𝑖𝑜𝑛 𝑜𝑝𝑒𝑟𝑎𝑡𝑜𝑟

𝑑𝑡 = 𝑑𝑖𝑣𝑖𝑑𝑒𝑛𝑡𝑠 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒 𝑓𝑜𝑟 𝑦𝑒𝑎𝑟 𝑡

They collected forecasts of dividends for fiscal year t+1 and t+2 and the long-range dividend forecast and maximum and minimum future price estimates by Value Line during the third quarter of the AIMR Report Publication year. Furthermore, they used t+2 and long-range dividend forecasts to obtain an estimated forecast of dividends for t+3, and by very similar logic they also computed the forecast of price in year T=4 (P4).

Residual income valuation models. Residual income valuation (RIV) models

address the difficulties in estimating a long-term growth rate by utilizing accounting information (Claus and Thomas 2001; Witmer and Zorn 2007). These models equate the current share price into the sum of two components, namely, the present value of expected dividends per share over a short horizon and a discounted terminal value (Witmer and Zorn 2007). Claus and Thomas (2001) implemented the RIV model using a four-year forecasting horizon (T=4) and set the growth rate (gL) equal to the

(17)

𝑃𝑡 = ∑ _(1+𝑟)𝑑𝑡 𝑡+ 𝑑𝑡 (𝑟−𝑃𝑒)(1+𝑟)4 4 𝑡=1 (2) Where: 𝑃_𝑡= 𝑝𝑟𝑖𝑐𝑒 𝑎𝑡 𝑑𝑎𝑡𝑒 𝑡 𝑟 = 𝑡ℎ𝑒 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑐𝑜𝑠𝑡 𝑜𝑓 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 Pe = 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑖𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 𝑑𝑡 = 𝑑𝑖𝑣𝑖𝑑𝑒𝑛𝑡𝑠 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒 𝑓𝑜𝑟 𝑦𝑒𝑎𝑟 𝑡

If dividends are all positive and the cost of equity is greater than the expected inflation rate, there is only one value of r that will solve this equation.

Price Earnings Growth (PEG) ratio models. A key element of the model is

recognition of the central role of short-term forecasts of earnings in valuation. Easton (2004) used a model of the PEG ratio to show how the difference between accounting earnings and economic earnings characterizes the role of accounting earning in valuation. Dhaliwal et al. (2010) used three different models, including Easton’s (2004) PEG ratio model, to calculate cost of capital and the mean of the three measures served as a proxy for the cost of equity capital.

𝑟2_{− 𝑟 �𝑑𝑝𝑠}1 𝑃0 � � − (𝑒𝑝𝑠2− 𝑒𝑝𝑠1) 𝑃0 � = 0 (3) While 𝑑𝑝𝑠₁ = 0, 𝑟𝑃𝐸𝐺 = �𝑒𝑝𝑠2− 𝑒𝑝𝑠1⁄ 𝑃0 (4) Where: 𝑃₀ = 𝑝𝑟𝑖𝑐𝑒 𝑎𝑡 𝑓𝑖𝑠𝑐𝑎𝑙 𝑦𝑒𝑎𝑟 𝑇 = 0 𝑟 = 𝑡ℎ𝑒 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑐𝑜𝑠𝑡 𝑜𝑓 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 𝑒𝑝𝑠1 = 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑖𝑎𝑐𝑐𝑜𝑢𝑛𝑡𝑖𝑛𝑔 𝑒𝑎𝑟𝑛𝑖𝑛𝑔 𝑎𝑟𝑒 𝑓𝑖𝑠𝑐𝑎𝑙 𝑦𝑒𝑎𝑟 𝑇 = 1 𝑒𝑝𝑠2 = 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑖𝑎𝑐𝑐𝑜𝑢𝑛𝑡𝑖𝑛𝑔 𝑒𝑎𝑟𝑛𝑖𝑛𝑔 𝑎𝑟𝑒 𝑓𝑖𝑠𝑐𝑎𝑙 𝑦𝑒𝑎𝑟 𝑇 = 2 𝑑𝑝𝑠₁ = 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑑𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑠 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒 𝑎𝑡 𝑓𝑖𝑠𝑐𝑎𝑙 𝑦𝑒𝑎𝑟 𝑇 = 1

Equation 3 is the original PEG model, while most of the previous literature adopted equation 4 as they assumed that expected dividends per share at the fiscal year T=1 is zero. This model is so widely used in the prior literature (Easton 2004; Dhaliwal et al. 2010; Clarkson et al. 2011; Wilson et al. 2013). Additionally, Botosan and Plumlee (2005) compare the validity of four different proxies and concluded that the Value Line Cost of Equity estimate and Easton’s (2004) PEG estimate outperform other estimates. The PEG estimation in Easton (2004) is less onerous in terms of data requirements because it only requires price and earnings growth to estimate the cost

(18)

of capital (Clarkson et al. 2010). I therefore will adopt equation 4 to compute cost of capital.

3.3 Firm Value

Agency theory does not identify the precise procedures for measuring the relevant accounting and security market performance indexes. Therefore, researchers make this choice on the basis of observed institutional contractual arrangements and ease in interpreting the empirical results (Lambert and Larcker 1987). In the social accounting area, however, most of the prior research only used security market measures as a proxy for financial performance. Specifically, stock price at the date of estimating cost of capital is the most common proxy used in the prior literature for firm value (Plumlee 2010; Clarkson et al. 2004; Clarkson et al. 2010; Wilson et al. 2013). I will employ the stock price as of the date I estimate the implied cost of capital as a proxy for overall firm value.

3.4 Empirical Models

To test the first research question regarding the relationship between sustainability reporting quality and cost of equity capital, I employ the following model:

𝐶𝑜𝐶 = 𝛼 + 𝛽1𝑅𝑒𝑄𝑢𝑎 + 𝛽2𝑆𝐼𝑍𝐸 + 𝛽3𝐵𝑡𝑜𝑀 + 𝛽4𝐵𝐸𝑇𝐴 + 𝜀 (5)

Where

CoC = cost of equity capital

ReQua = total KLD score

SIZE = log of total assets

BtoM = book-to-market ratio

BETA = firms equity beta estimated using previous 36 monthly stock returns

Size is included because larger firms have been found to have lower cost of equity capital, presumably because of lower perceived risk (Botosan and Plumlee, 2005).

Firms’ book-to-market ratio (BtoM) consistently exhibits a positive relationship with cost of equity capital in prior studies (Plumlee et al. 2010). This relationship arises because the market has less confidence in the economic value of high book-to-market firms’ reported assets and earnings, and thus discount future

(19)

expected growth in reported earnings more severely (Wilson et al. 2013). Firm’s equity beta (BETA) is included to control the impact of systematic risk on the cost of capital (Plumlee et al. 2010; Wilson et al. 2013).

To test the overall affect of sustainability reporting quality on expected future financial performance of a firm, I use the standard Ohlson (1995) valuation model:

𝑃 = 𝛼 + 𝛽1𝐵𝑉 + 𝛽2𝐸𝐴 + 𝛽3𝑅𝑒𝑄𝑢𝑎 + 𝜀 (6)

Where

P = stock price as of the date in which cost of capital estimated

BV = book value per share

EA = earnings per share ReQua = total KLD score

To test the impact of sustainability reporting quality on firm value in environmentally more sensitive industries, I divide my research sample into three categories: overall, environmentally more sensitive (Pulp and paper, Chemicals, Metals and Mining, Oil and gas, and Utilities), environmentally less sensitive (the rest). By doing so, I am able to set a benchmark (overall), and get a clear comparison from the empirical results of more/less sensitive industries.

4 Empirical Evidence

4.1 Sample

The sample consists of firms that are included in the S&P 500 from 2009 to 2011. The potential sample of approximately 500 firms each year was reduced for several reasons. In order to estimate firms’ cost of equity capital, model 4 is only defined if analysts’ two-year ahead earnings forecast (EPS2) is greater than forecast earnings one year ahead (EPS1). I have collected my related data from COMPUSTAT, IBES, KLD databases and excluded samples because of data availability. At the end, the sample consists of 846 firm years.

I classify firms based on Global Industry Classification Standard, while sensitive industries are categorized based on Clarkson et al.’s (2010) research, which includes Pulp and paper, Chemicals, Metals and Mining, Oil and gas, and Utilities as environmentally more sensitive industries. These industries fall into the industrial groups of Materials, Energy and Utilities according to the GICS. Samples from sensitive industries represented about one fifth of the total observations. Specifically,

(20)

sensitive industries contribute 20%, 14% and 18% to the sample from 2009 to 2011 respectively as shown in Table 2.

Table 2 Industrial Backgrounds of the Observations

Industries 2009 2010 2011 Sensitive Industries: Energy Utilities Materials 61 25 14 22 35 11 9 15 52 19 14 19 Others 236 222 240 Total 297 257 292 Sensitive / Total 20.5% 13.6% 17.8%

4.2 Descriptive statistics

Descriptive statistics regarding the distribution of main variables are presented in Table 3. The mean of the reporting quality is 5.4, which is about one fourth of the total score. This indicates that most of the sample firms have relatively lower social strengths. The mean value of cost of capital is about 11.8%, which is consistant with the results from Wilson et al. (2013) and Dhaliwal et al. (2010) while beta has mean score of 0.33, which is lower than the results from Wilson et al.(2013). I believe such difference is reasonable because the samples are from different capital makes and each market faces its unique systematic risk.

Table 3 Descriptive Statistics

Variable Mean Std. Dev. Min Max

P 49.93995 50.02554 1.39 578.65 CoC .1181249 .0897564 .0175412 1.094532 SIZE 4.151486 0.5776206 2.829547 6.355051 BtoM 2.386722 10.78877 .027658 281.6205 BETA .3332959 .218106 -.5753148 1.359012 EA 2.536511 2.415458 -14.49 19.47 BV 17.34997 24.37938 -31.266 607.41 ReQua 5.355792 4.443985 0 22

CoC= cost of equity capital, calculated based on �𝑒𝑝𝑠2− 𝑒𝑝𝑠1⁄ (eps𝑃0 2>eps1); BETA= firm’s

equity beta, covariance of previous 36 monthly stock return from November 1st of each fiscal year and S&P 500 monthly return during the same period; SIZE= log of total asset; BtoM= total asset divided by total market value of a firm; EA= earnings per share; BV= book value per share; ReQua= sum of KLD social strengths score (Max=24, Min=0); P= stock price at the date of November 1st of each fiscal year.

(21)

As expected, the correlaition cofficient in table 4 indicates that social reporting quality negatively and significantly correlated with cost of equity capital. BtoM ratio is positively correlated to cost of capital at 99% confidence level. This relationship arises because the market has less confidence in the economic value of high book-to-market firms’ reported assets and earnings, and thus discount future expected growth in reported earnings more severely (Wilson et al. 2013). I surprisingly notice that cost of capital negatively associates with the firm size and this is contradictory to the result from Plumlee and Botosan (2005). Last but not least, BETA also positively correlates with cost of capital at 90% confidence level. Overall, except firm size, predicted relationships between cost of capital and other variables exist.

However, the correlation between social reporting quality and stock price is negative, and this is contradictory to my hypothesis. Moreover, earnings per share and book value per share are positively and significantly correlated with stock price, and this is consistent with results from Wilson et al. (2013).

Table 4 Correlation Matrix (P Value)

CoC= cost of equity capital, calculated based on �eps₂− eps₁⁄ (eps2>eps1); BETA= firm’s equity beta, P₀ covariance of previous 36 monthly stock return from November 1st of each fiscal year and S&P 500 monthly return during the same period; SIZE= log of total asset; BtoM= total asset divided by total market value of a firm; EA= earnings per share; BV= book value per share; ReQua= sum of KLD social strengths score (Max=24, Min=0);

P= stock price at the date of November 1st of each fiscal year.

* Significant at 90% confidence level; ** Significant at 95% confidence level; *** Significant at 99% confidence level;

P CoC SIZE BtoM BETA EA BV ReQua P 1.0000 CoC -0.1586*** 1.0000 (0.0000) SIZE -0.1003*** 0.0730** 1.0000 (0.0035) (0.0339) BtoM -0.0617* 0.0893*** 0.3125*** 1.0000 (0.0730) (0.0094) (0.0000) BETA -0.0382 0.0661* 0.0061 0.0998*** 1.0000 (0.2666) (0.0548) (0.8585) (0.0037) EA 0.5800*** -0.2277*** 0.0507 -0.0486 -0.0789** 1.0000 (0.0000) (0.0000) (0.1410) (0.1577) (0.0217) BV 0.1379*** 0.0248 0.2793*** 0.1690*** -0.0469 0.0040 1.0000 (0.0001) (0.4704) (0.0000) (0.0000) (0.1725) (0.9086) ReQua -0.0490 -0.1343*** 0.4740*** 0.0795** -0.0695** 0.1081*** -0.0163 1.0000 (0.1546) (0.0001) (0.0000) (0.0208) (0.0433) (0.0016) (0.6365)

(22)

To summarize the correlation matrix, firms with higher social reporting quality have lower cost of equity capital and stock price. Furthermore, independent variable and all the control variables in Model 5 are significantly correlated to the dependent variable, cost of capital. And the control variables of Ohlson Model (EA, BV) are significantly correlated to the stock price. In addition, it is reasonable to conclude that there is no presence of significant multicolinearity among control variables in each model.

4.3 Results

Table 5 reports the result of the OLS to test the overall relationship between cost of capital and sustainability reporting. Column I of this table reports significantly negative relationship between cost of capital and sustainability reporting quality without any control variables. Column II, however, reports the test result of equation 5. The model is reasonably well fitted, with an F value of 10.39. Social reporting quality is significant and the coefficient on cost of capital is negative, supporting my hypothesis that sustainability reporting quality is negatively related to cost of capital (coeff.=-0.0042, p<0.01). Thus, it appears that market pays a slight premium for the equity of firms with higher social strength. Book to market ratio and BETA are not significantly related to the cost of capital.

Table 5 Regression of cost of capital against sustainability reporting quality(P Value)

CoC= cost of equity capital, calculated based on �eps₂− eps₁⁄ (eps2>eps1); BETA= firm’s equity beta, P₀ covariance of previous 36 monthly stock return from November 1st of each fiscal year and S&P 500 monthly return during the same period; SIZE= log of total asset; BtoM= total asset divided by total market value of a firm; ReQua= sum of KLD social strengths score (Max=24, Min=0).

Variables I II III ReQua -.0027135*** -.0042204*** -0.0037212 (0.000) (0.000) (0.135) SIZE .0241141*** -0.0094305 (0.000) (0.651) BtoM 0.0004401 0.0098349 (0.139) (0.194) BETA 0.0186414 -0.0426013 (0.182) (0.317) Adjusted R-Square 0.018 0.0426 0.04081 F Value 15.51 10.39 0.52

(23)

Column III reports the impact of reporting quality on cost of capital in sensitive industries. This model is not well fitted as evidenced from the low Adjusted R-Square and F value. However, I notice that firm size and equity beta as well as social reporting quality are negatively related to cost of capital. Lower cost of equity capital has been found in larger firms, presumably because of the perceived risk (Botosan and Plumlee, 2005). However, equity beta is also negatively related to the cost of capital, and this might partially reflect the fact that sensitive industries are only contribute to a small portion of the total sample and thus the estimation of the regression is noisier.

Table 6 Regression of future cash flow against sustainability reporting quality (P Value)

Variables I II III ReQua -0.55139 -1.24684*** -0.58919 (-0.155) (0.000) (0.200) BV .2745026*** 0.123196 (0.000) (0.407) EA 12.24864*** 4.966584*** (0.000) (0.000) Adjusted R-Square 0.0012 0.3669 0.2749 F Value 2.03 162.64 19.57

P= stock price (closing) at November 1st of each fiscal year; BV= book value per share; EA= earnings per share; ReQua= sum of KLD social strengths score (Max=24, Min=0).

Table 6 displays the results of Ohlson regression, which I use to examine whether there is evidence of an association between future cash flow and social reporting quality. Column I tests the relation between future cash flow and social reporting quality without any control variables and it reports that there is a not significant negative relationship between these two factors. In the Column II, however, I included control variables and the model has an Adjusted-R2 of 36.7% and F value of 162.64. Besides, the coefficient of sustainability reporting quality is negative and significant (-1.24684, P=0.000), inconsistent with the hypothesis. Book value per share and earnings per share are positively and significantly related to the future cash flow of a firm. However, the coefficient of book value per share (0.2745, p=0.000) is lower than the results from Wilson et al. (2013) and Clarkson et al. (2010).

In Column III, I repeat the Ohlson regression on the environmentally sensitive sectors (energy, materials and utilities). This model has a lower Adjusted-R2 and F

(24)

Value compared to Column II. Moreover, independent variable is not significantly associated with the future cash flow. In this test, only earnings per share is significantly correlated to the future cash flow, with a coefficient of 4.97, which is significant at 99% confidence level.

5 Conclusion

This paper has provided further evidence of the association between S&P 500 firms’ sustainability reporting practices and firm value, as measured by ex-ante cost of equity capital and expected future performance. I report a negative significant association between cost of capital and social reporting quality of a firm, and this result is in line of scholarly community theoretically and empirically. I also document that social reporting quality has negative effect on future market value of a firm. This is contradictory with the H2, however, consistent with some previous research. Iqbal et al. (2012) also documents that CSR has negative effect on the market value of the share, even though many other previous researches predicts and documents positive relationship between social reporting quality or social activity and future cash flow. Servaes et al. (2013), however, find out that corporate social responsibility and firm value are positively related in the firms with high customer awareness, while for the firms with low consumer awareness, the relation is either negative or insignificant. Their findings emphasize the role of public visibility of a firm on such mechanism between firms’ social activities and financial performance. However, I fail to document such relations in more environmentally sensitive sectors due to the limited sample.

This paper provides more insights to this field of research where mixed results exist and examines the relationship between social activities and firm value by using two different models, while majority of the previous studies only focus on either cost of equity capital or future cash flow of a firm. Moreover, this paper uses extensive amount of hand-collected and manually calculated data sets, and this increases the reliability of the evidence. However, as discussed earlier, this paper fails to report whether the impact of social reporting quality varies from sector to sector as predicted in H3.

(25)

References

Aerts, W., Cormier, D. and Magnan, M. 2008, Corporate environmental disclosure, financial markets and the media: An international perspective. Ecological Economics, 64.

Al-Tuwaijri, S., Christensen, T.E. and Hughes, K.E. II, 2004, The relations among environmental disclosure, environmental performance, and economic performance: a simultaneous equations approach, Accounting, Organizations and Society, 29.

Artiach, T. and Clarkson, P. 2011, Disclosure, Conservatism and the Cost of Equity Capital: A Review of the Foundation Literature, Accounting and Finance, 51, 1: 2–49.

Bachoo, K., Rebecca, T., Wilson, M. 2013, Firm Value and the Quality of Sustainability Reporting in Australia, Australian Accounting Review, 23: 67-87.

Botosan, C., Plumlee, M. 2002, A Re-examination of Disclosure Level and the Expected Cost of Equity Capital. Journal of Accounting Research, 40.

Botosan, C., Plumlee, M. 2005, Assessing Alternative Proxies for the Expected Risk Premium, Accounting Review, 80: 21-53.

Clarkson, P., Fang, X.H., Li, Y. and Richardson, G. 2010, The Relevance of Environmental Disclosures for Investors and Other Stakeholder Groups: Are Such Disclosures Incrementally Informative? Working Paper, Social Science Research

Network.

Clarkson, P., Richardson, G. and Vasvari, F. 2008, Revisiting the Relation Between Environmental Performance and Environmental Disclosure: An Empirical Analysis,

Accounting, Organizations and Society, 33.

Claus, J., Thomas, J. 2001, Equity Premia as Low as Three Percent? Evidence from Analysts’ Earnings Forecasts for Domestic and International Stock Markets. The

Journal of Finance, 5: 1629-1666.

Coles, J.L., Loewenstein, U. and Suay, J. 1995, On Equilibrium Pricing Under Parameter Uncertainty, Journal of Financial and Quantitative Analysis, 30.

Connors, E. and L. Silva-Gao, 2008, The Impact of Environmental Risk on the Cost of Equity Capital: Evidence From the Toxic Release Inventory. Working paper.

Dhaliwal, D., Li, O., Tsang, O. and Yang, Y. 2010, Voluntary Nonfinancial Disclosure and the Cost of Equity Capital: The Initiation of Corporate Social Responsibility Reporting, The Accounting Review, 86 (1).

Diamond, D. and Verrecchia, R. 1991, Disclosure, Liquidity, and the Cost of Capital,

(26)

Dye, R.A., 1985, Disclosure of Nonproprietary Information, Journal of Accounting

Research, Vol. 23, pp. 123-145.

Easton, P. 2004, PE Ratios, PEG Ratios, and Estimating the Implied Expected Rate of Return on Equity Capital. The Accounting Review, 79: 73-95.

Figge, F. 2005, Value-based Environmental Management: From Environmental Shareholder Value to Environmental Option Value, Corporate Social Responsibility

and Environmental Management, 12: 19–30.

Gamble, G.O., Hsu, K., Jackson, C. and Tollerson, C.D. 1996, Environmental Disclosure in Annual Reports: An International Perspective, International Journal of

Accounting, 31.

Global Reporting Initiative (GRI), 2006, Sustainability Reporting Guidelines 3.0, Available at: https://www.globalreporting.org/resourcelibrary/G3-Sustainability-Reporting-Guidelines.pdf.

Hughes, J., Liu, J., and Liu, J. 2009, On the Relation Between Expected Returns and Implied Cost of Capital, Review of Accounting Studies, 14:246–259.

Ilinitch, A., Soderstrom, N., &Thomas, T. 1998, Measuring Corporate Environmental Performance. Journal of Accounting and Public Policy, 17, 387-408.

Iqbal, N., Ahmad, N., Basheer, N.A., Nadeem, M. 2012, Impact of Corporate Social Responsibility on Financial Performance of Corporations: Evidence from Pakistan.

International Journal of Learning & Development, Vol 2, No.6.

Jo, H., Harjoto, M. 2011, Corporate Governance and Firm Value: the Impact of Corporate Social Responsibility. Journal of Business Ethics, 103:351-388.

KPMG, 2002, KPMG International Survey of Corporate Sustainability Reporting 2002, Available at: http://www.gppi.net/fileadmin/gppi/KPMG2002.pdf.

KPMG, 2011, KPMG International Survey of Corporate Sustainability Reporting 2011, http://www.kpmg.com/PT/pt/IssuesAndInsights/Documents/corporate-responsibility2011.pdf.

Lambert, R., Larcker, D. (1987). An Analysis of the Use of Accounting and Market Measures of Performance in Executive Compensation Contracts. Journal of

Accounting Research, 25: 85-125.

Lev, B., C. Petrovits, and S. Radhakrishnan. 2008, Is Doing Good Good for You? How Corporate Charitable Contributions Enhance Revenue Growth. Working paper.

Magnan, M., Cormier, D., Ledoux, M-J. 2011, The informational contribution of social and environmental disclosures for investors, Management Decision, 49.

Murray, A., Sinclair, D., Power, D. and Gray, R. 2006, Do Financial Markets Care about Social and Environmental Disclosure? Further Evidence and Exploration from the UK, Accounting, Auditing and Accountability Journal, 19, 2: 228–55.

(27)

Ohlson, J. 1995, Earnings, Book Values, and Dividends in Equity Valuation.

Contemporary Accounting Research, 661-687.

Patten, D. 2002, The Relation Between Environmental Performance and Environmental Disclosure: A Research Note. Accounting, Organizations, and Society, 27.

Plumlee, M., Brown, D., Hayes, R.M. and Marshall, R.S. 2010, Voluntary Environmental Disclosure Quality and Firm Value: Further Evidence, Working Paper,

Social Science Research Network.

Richardson, A.J. and Welker, M. 2001, Social Disclosure, Financial Disclosure and the Cost of Equity Capital, Accounting, Organizations and Society, 26.

Servaes, H., Tamayo, A. 2013, The Impact of Corporate Social Responsibility on Firm Value: the Role of Customer Awareness. Management Scienece, 59, 1045-1061.

S&P Dow Jones Indices, 2013, Dow Jones Sustainability World Index Guide 12.3, Available at:

http://www.sustainability-indices.com/images/djsi-world-guidebook_tcm1071-337244.pdf.

Witmer, J., Zorn, L. 2007, Estimating and Comparing the implied cost of equity for Canadian and U.S. firms. Working Paper, Bank of Canada, 48.

Sustainability reporting quality and firm value : does industry matter?

Amsterdam Business School

Master Thesis