Integrated Report

lable at ScienceDirect

Journal of Cleaner Production 136 (2016) 102e110

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Sustainability management and reporting: the role of integrated reporting for communicating corporate sustainability management

Riccardo Stacchezzini a, *, Gaia Melloni b, Alessandro Lai a

a University of Verona, Department of Business Administration, Via Cantarane, 24, 37129 Verona, Italy b Bocconi University, Department of Accounting, Via Roentgen, 1, 20136 Milano, Italy

a r t i c l e i n f o

Article history: Received 8 August 2014 Received in revised form 6 October 2015 Accepted 27 January 2016 Available online 23 February 2016

Keywords: Integrated reporting Sustainability management Sustainability reporting Sustainability accounting Leading indicators Impression management

* Corresponding author. E-mail address: [email protected] (R.

http://dx.doi.org/10.1016/j.jclepro.2016.01.109 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

According to the International Integrated Reporting Council (IIRC), integrated reporting (IR) should disclose the leading indicators that managers use in their sustainability decision processes and thus overcome traditional silo thinking. In this respect, IR could favor the integrative management of sus- tainability. Yet some scholars are more critical and argue that IR cannot contribute to sustainability management.

This article contributes to the debate by analyzing how IR adopters communicate managerial aspects of corporate sustainability. Drawing on impression management studies, this study seeks to detect manipulations in this disclosure practice. A manual content analysis of reports available on the IIRC website and a multivariate statistical analysis reveal that firms offer biased IR disclosures. Firms not only provide limited forward-looking and quantitative disclosure of their actions to achieve sustainability outcomes but also avoid providing information about their sustainability performance when their social and environmental results are poor. The evidence suggests pessimistic conclusions about the capability of this reporting process to encourage the integrative management of corporate sustainability.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Firms encounter strong pressures to behave in sustainable ways and be transparent about their sustainability practices (Lozano and Huisingh, 2011), by communicating about the actions they have taken and results achieved in terms of the economic, social, and environmental dimensions of sustainability (Elkington, 1998). In response, companies may conveniently adopt an integrative man- agement to measure, manage, and report on their sustainability performance (Schaltegger and Wagner, 2006a). Specifically, firms may rely on sustainability accounting, a comprehensive “process of the collection, analysis and communication of sustainability- related information” (Bennett et al., 2013, p. 3) and integrate sus- tainability management accounting (SMA, defined as internal per- formance measurement tools and information that help managers address sustainability issues) with external sustainability reporting (SR), which communicates information about how the company acts to improve its economic, environmental, and social effective- ness and efficiency (Daub, 2007). These efforts should contribute to

Stacchezzini).

corporate sustainability (Lozano and Huisingh, 2011), though for SR to have an impact on sustainability attitudes, practices, and culture (Bebbington, 2007), firms cannot perceive it as a mere tool to advance their image or detract attention from their actual poor engagement in sustainability issues (O'Dwyer, 2003). That is, SR needs to be strictly linked to SMA and provide information about sustainable value creation in a way that enables external stake- holders to realize the actual achievements and potentials of corporate sustainability management (SM). Because information disclosed to external actors is “inherently judgmental” (Bennett et al., 2013, p. 4), both the integration of SR and SMA and the integrative management of corporate sustainability are contested grounds.

Accounting scholars accordingly express concerns about the ability of various SR initiatives to promote such integration. Inte- grated reporting (IR), as defined by the International Integrated Reporting Council (IIRC), seeks to overcome such limitations by integrating information systems and data that support both inter- nal and external reporting. Adopters are required to disclose the information that managers use in their decision making (IIRC, 2013a) and embrace the notion of integrated thinking, to avoid traditional forms of silo thinking in their performance

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110 103

measurement and reporting efforts (Giovannoni and Fabietti, 2013). The IIRC also requires firms to describe how they deal with different dimensions of sustainability, including not just their performance but also the actions taken to achieve it (IIRC, 2013a). Companies should disclose information about leading indicators of sustainability, including the performance drivers with the greatest influence on their achievement of core strategic objectives (Figge et al., 2002; Kaplan and Norton, 1996).

However, even in the face of the emerging debate about whether and how IR creates transparency about firms' sustain- ability commitment (Adams, 2015; Adams et al., 2011; Burritt, 2012; Flower, 2015; Milne and Gray, 2013), we lack empirical evi- dence about whether IR supports the integration of SMA and SR or, more generally, how IR can enhance integrative management for sustainability. To address this research gap, we investigate how IR adopters actually communicate about their SM. Among the different perspectives available for studying corporate discre- tionary disclosure strategies (Merkl-Davies and Brennan, 2007), we choose impression management (IM) as a foundation (Hooghiemstra, 2000; Neu et al., 1998) to identify biases between corporate disclosure and corporate behavior, such that firms might use IR disclosures to distract attention away from their poor sus- tainability achievements. We thus aim to answer a central research question: Does IR favor the integrative management of sustainability by conveying unmanipulated sustainability disclosures? A negative answer would support the IM perspective but also lead us to reject the competing perspective of incremental information (Merkl- Davies and Brennan, 2007), which suggests that discretionary dis- closures help limit reporting asymmetries, because they are un- manipulated (Baginski et al., 2000).

In a first level of analysis, we examine the SM disclosure pro- vided in the IR. With an in-depth, manual content analysis of the single text units of the reports available on the IIRC website, we gather evidence about how IR adopters incorporate information about their sustainability actions and performance in their disclo- sures, related to economic, social, and environmental dimensions. The economic dimension typically refers to financial resources and returns, the social one deals with how a company takes care of the diversity of social and human demands, and the environmental dimension refers to interactions with the natural environment (Daub, 2007). This content analysis distinguishes between disclo- sures of sustainability actions and sustainability performance, while also considering the type of measure (quantitative or not) and time orientation (forward-looking or not). At a second level of analysis, we verify the presence of biases between what is (formally) reported in corporate disclosures and (actual) corporate behavior. To evaluate the relation between IR sustainability dis- closures and variables that proxy for sustainability performance, corporate governance, and industry, we use the mean of a multi- variate statistical analysis.

We start by describing IR as the analysis context, the IIRC proj- ect, the requirements of its framework for disclosing sustainability issues, and how IR seeks to encourage the integrative management of sustainability. Then we present previous research on sustain- ability accounting and emerging literature on IR, and specify the perspective on which we base our research design. After we introduce our theoretical framework and research hypotheses, we outline the methodology for our content and multivariate statistical analyses, then present their results. Finally, we discuss the results and highlight both the contributions and limitations of this study.

2. Analysis context

IR is the process of communicating about sustainable value creation over time, which provides insights into the resources and

relationships used or affected by the organization, as well as the interaction between the organization and its external environment (IIRC, 2013a). Despite its voluntary nature, IR has gained consider- able prominence since the formation of the IIRC in 2010 (de Villiers et al., 2014). Approximately 100 firms have taken part in the IIRC pilot program since its launch in October 2011; 75 business firms uploaded their reports to the Emerging Integrated Reporting Database as of 30 April 2014 (http://examples.theiirc.org/home). Evidence about the development of IR at the organizational level is not yet available, but indications suggest notable movements in some countries (e.g., United Kingdom, the Netherlands, Spain, Australia, Singapore, Japan, United States); South Africa pioneered mandatory adoption of IR in 2010. The IIRC also has signed important agreements with international standard setters (e.g., Memoranda of Understanding with the International Accounting Standard Board and Global Reporting Initiative, both in 2013). More generally, global players devote attention and resources to the IR project, including international accounting firms (e.g., EY, 2013; KPMG, 2013; PWC, 2013) and national and international profes- sional organizations (e.g., ACCA, 2012; ICAEW, 2013).

Although IR is not intended to be a sustainability report, its focus on sustainability issues and purpose of stimulating integrated thinking are clear; as the CEO of the IIRC explains, “IR promotes a more cohesive and efficient approach to corporate reporting that draws on different reporting strands. Here, our innovators focus in on the sustainability perspective. They tell their journey, how they are connecting departments across their business, how they are developing their strategies for longer term value creation” (IIRC, 2013b, p. 2). Therefore, IR should favor an integrated approach to sustainability and extend “the traditional corporate focus on short term financial performance to the inclusions of the resources and relationships that are essential to the company's long-term success and viability. It is the integration of the social, economic and environmental dependencies and issues pertinent to the company's business” (King and Roberts, 2013, pp. 40e41).

The International Integrated Reporting Framework (IIRF) seeks to foster integration across information systems that support in- ternal and external reporting (Giovannoni and Fabietti, 2013); it assigns great importance to disclosing information used internally by managers. Firms should reveal not only their performance but also how they achieve these results, that is, through which re- sources and processes. An IR adopter therefore should provide in- formation about the firm inputs (resources, relationships), business activities, outputs (products, services), and outcomes (impacts on resources, relationships). The IIRF also requires a forward-looking focus (targets, forecasts, projections) and quantitative information (key performance indicators, monetized metrics). The United Na- tions Global Compact LEAD anticipates that adopting such report- ing practices “encourages and supports the integration of sustainability in strategic planning, decision-making and opera- tions” (LEAD, 2013, p. 3). The disclosures through IR should encourage communication about leading sustainability indicators, as well as greater understanding of corporate engagement across different dimensions of sustainability (Busco et al., 2013). Although companies also might offer SM disclosures in other corporate re- ports, firms that adopt the IIRF all should provide this information in their IR. Therefore, we investigate IR as a vehicle for disclosing information on sustainability actions and performance, favoring integrative SM.

3. Previous research

The ability of sustainability accounting to reflect corporate engagement in sustainability and determine corporate SM is an increasingly relevant concern for business and academia (Daub,

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110104

2007; Fernandez-Feijoo et al., 2014; Hahn and Kühnen, 2013; Lozano and Huisingh, 2011; Schaltegger and Zvezdov, 2015). On one side, the “management oriented path to sustainability accounting” (Burritt and Schaltegger, 2010, p. 831) recognizes sets of tools that managers use to make decisions, highlighting the potentials of SMA and SR to be managed in an “integrated manner” and the role that SMA can play in “linking” corporate SM and SR (Schaltegger and Wagner, 2006b). By integrating corporate sustainability in- dicators, the reporting corporation can better manage its sustain- ability (Daub, 2007; Roca and Searcy, 2012). Lodhia and Martin (2014) also highlight the use of corporate sustainability indicators to manage the company and inform stakeholders. Firms that identify, measure, and disclose sustainability indicators enhance internal and external understanding of their corporate perfor- mance on economic, social, and environmental issues. The process of developing key performance indicators for SR also favors the integration of sustainable performance data into decision making, risk management, and performance management (Adams and Frost, 2008).

On the other side, research focused on the “critical path of sustainability accounting development” (Burritt and Schaltegger, 2010, p.830) highlights the limits of SR for representing actual corporate engagement (Tregidga et al., 2014). Sustainability reporting thus may be a mere communication strategy, designed to repair organizational legitimacy (Deegan, 2007) or conform with institutionalized norms (Larrinaga-Gonz�ales, 2007). Moreover, it may act as a corporate veil to protect the firm's inner workings from external view (Hopwood, 2009). Symbolic uses of SR might advance the corporate image, in the absence of corresponding engagement (Neu et al., 1998; O'Dwyer, 2003). In this case, SR and SMA are not concomitant, which limits the implementation of integrative management for sustainability.

In summarizing these two sides, we acknowledge that “external reporting … frequently corresponds (as it should) with what is measured and reported internally.… However, the decision on the actual information to be disclosed is inherently judgmental so that the selection of the information to be reported is inevitably up to the discretion of the directors of each company” (Bennett et al., 2013, p. 4). By coming within the observational lens of both research perspectives, the sustainability-related nature of IR has prompted a primarily conceptual debate (Adams, 2015; Burritt, 2012; Flower, 2015; Milne and Gray, 2013). That is, IR may offer rich grounds for sustainability disclosures and favor the integration of sustainability accounting and SR (Adams et al., 2011; Busco et al., 2013), but these questions remain empirically unexplored. Existing empirical studies of IR focus on the determinants of IR adoption, in relation to institutional factors (Jensen and Berg, 2012; Frías- Aceituno et al., 2013), governance characteristics (Frías-Aceituno et al., 2012), SR assurance (Sierra-García et al., 2015), industry concentration (Frías-Aceituno et al., 2014), or legitimacy needs (Lai et al., 2014).

Investigating the integrative management of sustainability from the viewpoint of discretionary disclosure instead reflects two main competing schools of thought (Merkl-Davies and Brennan, 2007): incremental information and IM. The first one assumes that disclosure helps overcoming information asymmetries between corporate reporting preparers and users. It predominantly adopts an user perspective, and the main concerns relate to detecting the ways investors respond to discretionary disclosures. The latter questions the idea of an inherently unbiased disclosure and focuses on detecting manipulations in reporting sustainability achieve- ments. IM studies take a preparer perspective, aimed at detecting circumstances in which firms use impression management strate- gies. Despite some distinctive rationales, accounting scholars explicitly develop competing hypotheses (Cho et al., 2012;

Mahoney et al., 2013) or comply with only one school of thought (e.g. García Osma and Guillam�on-Saorín, 2011), with the implicit assumption that a rejection of related hypotheses constitutes an affirmation of the contrasting school's ideas. We are thus pushed to investigate opportunistic managerial behavior with regard to sus- tainability disclosure and privilege a preparer perspective, such that we base our research hypotheses on the impression manage- ment framework.

4. Theoretical framework and research hypotheses

The IM framework offers a theoretical and methodological means to address managers' use of SR to distract from their poor sustainability results. In this process, one entity attempts to control others' impressions (Leary and Kowalski, 1990) by presenting in- formation in a manner that distorts recipients' perceptions, moti- vated by a desire to present a self-serving image (Neu et al., 1998; Hooghiemstra, 2000). To derive conclusions about the potential presence of IM strategies, we investigate the association between disclosures that might be manipulated and specific corporate characteristics that favor manipulation. For this assessment, we focus on thematic manipulation (Merkl-Davies and Brennan, 2007) and investigate whether the choice to provide disclosures about actions taken to support SM processes, rather than disclosures about sustainability results, constitutes a manipulation strategy by a company that prefers to highlight the resources it has used and activities implemented, rather than its actual sustainability achievements.

Corporate sustainability results likely affect the use of thematic manipulation, such that poor economic performance incentivizes firms to manipulate their disclosures and “balance out” the nega- tive perceptions that might arise from their poor results (Clatworthy and Jones, 2003; Schleicher and Walker, 2010). Cho et al. (2010) demonstrate that firms with the worst environ- mental performance are more likely to disclose about environ- mental issues than firms with better performance; Mahoney et al. (2013) posit that firms with weaker social performance use disclosure strategies to alter public perceptions of their corporate behavior, but they fail to find support for this hypothesis. We similarly argue that weak sustainability performance (poor eco- nomic and social performance, high environmental impacts) leads managers to try to balance negative perceptions of the firm by increasing the amount they disclose about sustainability actions but avoiding disclosures about their sustainability performance. Formally.

H1. Firms with (a) poorer social performance, (b) greater envi- ronmental impacts, and (c) poorer economic performance are more likely to provide disclosures about sustainability actions rather than sustainability performance in their IR.

Weak corporate governance also might favor the use of manipulative disclosures. For example, a less independent board grants managers greater discretion to present information in a self- serving way. Previous studies rely on the proportion of indepen- dent directors on the board to represent a factor that likely miti- gates IM strategies, by improving transparency and reducing self- serving disclosures (García Osma and Guillam�on-Saorín, 2011; Mather and Ramsay, 2007). Because disclosures about actions are harder to verify by external stakeholders than disclosures about performance, we predict:

H2. Firms with less independent boards are more likely to provide disclosures about sustainability actions rather than sustainability performance in their IR.

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110 105

Finally, industry traits should affect IM, in that some industries are exposed to more public scrutiny than others. Firms in envi- ronmentally sensitive industries are more likely to use these stra- tegies (Cho et al., 2010), so membership in such industry groups should increase the use of sustainability action disclosures. Formally.

H3. Firms in environmentally sensitive industries are more likely to provide disclosures about sustainability actions rather than sustainability performance in their IR.

5. Methodology

5.1. Sample selection and data collection

We gathered IR by all firms that made their reports available on the IIRC website before 30 April 2014 and that included a section (usually entitled “Business Model”) with information about their sustainable value creation process. This section generally includes information about different dimensions of sustainability perfor- mance (i.e., environmental, social, and economic), actions taken, and performance achieved. Of the 79 reports available, we excluded 25 that did not provide a specific section about value creation pro- cesses. The 54 remaining firms belonged to ten industry sectors: 46% in environmentally sensitive industries (Oil & Gas; Basic Materials; Industrials; Utilities) and 54% representing non-environmentally sensitive industries (Consumer Goods; Consumer Services; Health Care; Technology; Telecommunication; Financials).

We manually collected data from these firms' IR, available in the Integrated Reporting Emerging Practice Examples Database. These reports belong to businesses from around the world and are pub- licly accessible through the IIRC's official website. We also collected data about corporate characteristics, for the multivariate analysis, from the Bloomberg database for 2011, 2012, and 2013, parallel with the release years of the IR.

5.2. Data analysis

The analysis consists of two steps. First, we developed a manual content analysis to assess disclosure quality. Second, we used this evidence in a multivariate statistical analysis to test the relation between disclosures and specific corporate characteristics.

5.2.1. Manual content analysis For the in-depth, manual content analysis, we chose text units as

recording units (Beattie et al., 2004), such that we highlighted and coded every sentence in the IR value creation process section. If a sentence featured more than one statement, we considered each statement (i.e., text unit) separately. We also included figures and diagrams that reported on value creation processes, because the IIRF recommends graphic depictions. Therefore, we analyzed 4426 text units from the 54 different reports.

To begin, we classified each text unit according to its thematic content: If the disclosure referred to outcomes of the firm value creation process, it constituted a sustainability performance disclosure (SPD), whereas if it pertained to resources used or ac- tions taken to drive sustainability performance, it was a sustain- ability action disclosure (SAD). Next, we classified the text units according to sustainability dimensions: economic, social, or envi- ronmental, where the economic dimension refers to financial re- sources and returns, the social one deals with how a company takes care of diverse social and human demands, and the environmental dimension pertains to interactions with the natural environment. Drawing on previous studies that assessed disclosure quality (Beattie et al., 2004), we also consider linguistic attributes, that is,

the semantic properties of the information provided, which distinguish the text units by type of measure (quantitative or non- quantitative) and time orientation (forward-looking or non- forward-looking). Table 1 contains disclosure examples. To ensure a clear, complete understanding of our classification process, we provide examples of the types of measures and time orientations and specify the three dimensions of sustainability (economic, so- cial, and environmental).

To ensure the reliability of the classification procedure, we applied a pretest, to clarify the coding rules and standardize the coders' classification capabilities. Two authors then independently coded each IR and repeated the analysis on a sample of 21 reports at different time periods. In support of the reliability of the process, the coefficient of agreement (i.e., ratio of pairwise interjudge agree- ments to total pairwise judgments) was above acceptable levels.

5.2.2. Multivariate statistical analysis For the multivariate analysis, we estimated a logit regression

model, using data for 2011, 2012, or 2013, depending on the IR release year. We investigated whether disclosures of sustainability activity, rather than sustainability performance, were associated with worse social, environmental, and economic sustainability performance (Models 1e3). We also tested if firms offered more SAD when they had weak corporate governance (Model 4) or functioned in an environmentally sensitive industry (Model 5). All the models included the linguistic attributes (type and time orientation) as control variables, to verify if disclosures tend to be quantitative and forward looking or not. Specifically, the regression models were:

SM_DISC ¼ a0 þ a1SOC_PERF þ a2DISC_TYPE þ a3DISC_TIME þ 3: ðModel 1Þ

SM_DISC ¼ a0 þa1ENV_IMPACT þa2DISC_TYPE þa3DISC_TIME þ3: ðModel2Þ

SM_DISC ¼ a0 þ a1ECO_PERF þ a2DISC_TYPE þ a3DISC_TIME þ 3: ðModel 3Þ

SM_DISC ¼a0 þa1BOARD_INDEP þa2DISC_TYPEþa3DISC_TIME þ3: ðModel4Þ

SM_DISC ¼a0 þa1ENV_SENS_INDþa2DISC_TYPEþa3DISC_TIME þ3: ðModel5Þ

The dependent variable is SM_DISC, a dummy variable equal to 1 if the IR text unit offers SAD and 0 otherwise (i.e., if it offers SPD). The explanatory variables are social performance (SOC_PERF), which is the amount of yearly community spending, and environ- mental impact (ENV_IMPACT), which is the amount of greenhouse gas emissions. Markets rely on such data to assess social and environmental performance (Eccles et al., 2011). For economic performance, we used the return on equity (ECO_PERF). Board in- dependence (BOARD_INDEP) was the proportion of board- independent directors. Finally, we considered an industry dummy (ENV_SENS_IND), equal to 1 if firms work in environmentally sen- sitive industries and 0 otherwise. For the linguistic attributes, we measured disclosure type (DISC_TYPE) as a dummy variable, equal to 1 if the text unit was non-quantitative and 0 otherwise. The disclosure time orientation (DISC_TIME) was another dummy

Table 1 Examples of IR disclosures.

Sustainability actions Sustainability performance

Quantitative and forward-looking

“We plan to start-up 26 new major fields in the next four years, mainly Goliat in the Barents Sea, the Block 15/06 West Hub in Angola, the heavy oil and gas Venezuelan assets and Jangkrik in Indonesia, which will add more than 500 kboe/d by 2017, supporting production growth and the replacement of mature production”a

“Target 2012: 1.2e1.5 times dividend cover; … Gross debit/Ebitda <3”a

Quantitative and non-forward-looking

“Improve service quality and safety through the completion of investment commitments underwritten in the Convention (actions envisaged worth over Euro 20Bn)”b

“[Company] assisted 37,415 families by paying out R299 million in funeral claims during 2011”b

Non-quantitative and forward-looking

“We will continue to implement energy efficiency projects to decrease our usage so that we comply with the South African Government's power conservation programme when it is implemented”c

“We will halve the environmental footprint of our products”c

Non-quantitative and noneforward-looking

“Key drivers of profitability: Global growth of engineering plastics; Exposure to high growth Asian markets; High utilization rates”a

“Outcomes: Forest residues; Solid waste; Impact on water resources; Impact on the soil; Impact on biodiversity; Impact on community”c

a Economic dimension of sustainability. b Social dimension of sustainability. c Environmental dimension of sustainability.

Source: Extracts of IR available on the IIRC website.

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110106

variable, equal to 1 if the text unit was non-forward looking and 0 otherwise.

To confirm the robustness of the results, we ran five sensitivity tests, such that we performed the multivariate analysis (Models 1e5) but also included a control variable for disclosure length (DISC_LENGTH), measured as the amount of text units about both sustainability actions and performance. The entire set of depen- dent, independent, and control variables appears in Table 2.

6. Results

6.1. Manual content analysis

Of the 4426 text units analyzed, 2611 (58.99%) offer SAD, and the remaining 1815 (41.01%) include SPD. These two types of infor- mation refer similarly to the three dimensions of sustainability (economic, social, and environmental): 38.64% to economic di- mensions, 40.60% to social issues, and 20.76% to environmental dimensions, overall. However, the amount of SAD focused on the economic dimension is lower than the amount of SPD, and the SAD is centered more on environmental dimensions than the SPD (Table 3).

In terms of the linguistic attributes, quantitative information appeared in 25.58% of the text units, whereas 74.42% were non- quantitative. Only 12.11% of the text units included forward- looking information, and the remaining 87.89% were not forward looking. The SAD embeds fewer quantitative indicators (17.58%)

Table 2 Variable definitions and measures.

Definition Measurement

Dependent Sustainability management disclosure (SM_DISC) Dummy variable, ¼ 1 if the t Performance Social performance (SOC_PERF) Amount of money spent by t Environmental impact (ENV_IMPACT) Amount of Scope 1 and Scop Economic performance (ECO_PERF) Return on equity capital in t Governance Board independent directors (B_IND) Percentage of board-indepen Industry Environmentally sensitive industry (ENV_SENS_IND) Dummy variable, ¼ 1 if firm Linguistic attributes Disclosure type (DISC_TYPE) Dummy variable, ¼ 1 if the t Disclosure time (DISC_TIME) Dummy variable, ¼ 1 if the t Additional variable for sensitivity tests Disclosure length (DISC_LENGTH) Number of text units in each

than the SPD (37.08%), and the amount of forward-looking infor- mation is lower in SAD than in SPD (10.99% and 13.72%, respec- tively) (Table 4). Pearson chi-square tests indicated significant differences between SAD and SPD in both the type of information and the time orientation they displayed (Pearson chi-square probabilities < 0.05).

We considered the possible combinations of the two linguistic attributes, as detailed in Table 5.

These results indicate that forward-looking SAD represents 6.48% of all disclosures. There is thus little space for the disclosure of the so-called leading indicators of sustainability. Even though companies provide 58.99% of their disclosures about sustainability actions, it is mainly non-forward looking, so IR readers gain limited insights into how the firm plans to manage its sustainability in the future. Quantitative, forward-looking SAD represents 0.72% of the whole disclosure universe, whereas non-quantitative, non- eforward-looking SAD represents 42.86%. We similarly find a limited volume of forward-looking SPD (5.63%). Quantitative SPD is 15.21% of the total, but forward-looking quantitative disclosures constitute only 1.36% of the whole universe.

6.2. Multivariate statistical analysis

6.2.1. Descriptive statistics As we show in Table 6, firm social performance, measured as the

amount of community spending, is equal to US$116,257 on average. The mean environmental impact of the firms in our sample is 0.144

ext unit offers SAD, ¼ 0 otherwise (i.e., offers SPD)

he company on community-building activities, in thousands of US$ e 2 greenhouse gas emissions in thousands of metric tons (divided by 1,000,000) housands of US$

dent directors

belongs to an environmentally sensitive industry, ¼ 0 otherwise

ext unit is non-quantitative, ¼ 0 otherwise ext unit is non-forward looking, ¼ 0 otherwise

report section providing sustainability management disclosure (SAD and SPD)

Table 3 Disclosures of sustainability actions and performance in IRs by sustainability dimensions.

Total Economic Social Environmental

Text units % Text units % Text units % Text units %

SAD 2611 58.99 938 35.93 1050 40.21 623 23.86 SPD 1815 41.01 772 42.53 747 41.16 296 16.31 Total text units 4426 100.00 1710 38.64 1797 40.60 919 20.76

Table 4 Disclosures of sustainability actions and performance in IRs by linguistic attributes.

Quantitative Non-quantitative

Text units % Text units %

SAD 459 17.58 2152 82.42 SPD 673 37.08 1142 62.92 Total disclosure 1132 25.58 3294 74.42 Pearson c2(1) ¼ 213.901, Pr ¼ 0.000

Forward looking Non-forward looking

Text units % Text units %

SAD 287 10.99 2324 89.01 SPD 249 13.72 1566 86.28 Total disclosure 536 12.11 3890 87.89 Pearson c2(1) ¼ 7.481, Pr ¼ 0.006

Table 6 Descriptive statistics: selected variables.

Obs. Mean Std. dev. Min Max

SOC_PERF 44 116,257 329,598 109 1,814,000 ENV_IMPACT 31 0.144 0.287 0.000 1.282 ECO_PERF 54 23.472 31.115 �5.838 197.158 BOARD_INDEP 43 63.155 20.031 22.222 100 D_LENGTH 54 81.963 97.464 5 465

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110 107

million greenhouse gas emissions. Their average economic perfor- mance is equal to US$23,472. In terms of governance, 63% of board members are independent. The control variables in the sensitivity tests showed that the report section contained an average of 82 text units, with a minimum of 5 and a maximum of 465.

We present the pairwise correlations among the same variables in Table 7. No statistically significant correlations arose between the continuous variables across the sample firms, with the exception of ECO_PERF and BOARD_INDEP.

6.2.2. Multivariate analysis With Model 1, we test H1a and confirm that poorer social per-

formance increases the probability that a firm will provide SAD, in that SOC_PERF exhibits a negative coefficient that is statistically significant at 5%. Similarly, firms with higher environmental im- pacts provide more SAD than SPD, according to the positive, sig- nificant coefficient of ENV_IMPACT in Model 2, in support of H1b. However, we find no significant association between firms' eco- nomic performance (ECO_PERF) and SAD in Model 3, so we cannot confirm H1c.

Model 4 supports our predictions about the governance variable in H2. The negative, statistically significant coefficient of BOARD_- INDEP (�0.004, significant at 5%) indicates that lower indepen- dence of the board increases the provision of SAD. The positive, statistically significant coefficient of ENV_SENS_IND (Model 5) is consistent with H3; when firms are in environmentally sensitive industries, they tend to provide more SAD than SPD (5% significance level). In all the models, the supply of SAD is significantly associated

Table 5 Disclosures of sustainability actions and performance in IRs: Combination of linguistic a

Forward looking

Quantitative Non-quantitative

SAD 0.72% 5.76% SPD 1.36% 4.27% Total disclosure 2.08% 10.03%

with the presence of information that features non-quantitative, noneforward-looking attributes. That is, we uncover positive co- efficients of DISC_TYPE and DISC_TIME (both significant at 1% in Models 1e5). We summarize the multivariate analysis results in Table 8.

6.3. Sensitivity tests

The sensitivity tests control for the effect of DISC_LENGTH in Models 1e5, which also might affect the supply of SAD (Table 8). These results reveal a negative association between disclosure length and SAD (significant in Models 1e3 and 5 but not Model 4): When the IR section is shorter, the information features SAD rather than SPD. Both SOC_PER and BOARD_INDEP negatively affect the supply of SAD. Similarly, a higher ENV_IMPACT is associated with increased SAD. However, we did not find a significant relationship between ECO_PERF and SM_DISC. Membership in ENV_SENS_IND, as highlighted in the previous analysis, significantly affected the amount of SAD. Finally, we provide confirmatory evidence of the significant and positive relation between non-quantitative and non-forward-looking disclosure attributes and the supply of SAD. Thus, the sensitivity tests support the results of the logistic regression models.

7. Conclusions

7.1. Discussion of the results

We sought to investigate how IR adopters communicate their SM and thus offer empirical evidence about the role of IR in sup- porting integrative management of sustainability by conveying unmanipulated disclosures. Our content analysis reveals first that firms offer not only limited quantitative indicators but also little forward-looking information about their sustainability actions and performance. Thus, IR adopters are not informing external stake- holders about the leading indicators they used to make decisions,

ttributes.

Non-forward looking Total

Quantitative Non-quantitative

9.65% 42.86% 58.99% 13.85% 21.53% 41.01% 23.50% 64.39% 100.00%

Table 7 Pairwise correlations of selected variables.

SOC_PERF ENV_IMPACT ECO_PERF BOARD_INDEP D_LENGTH

SOC_PERF 1.000 ENV_IMPACT �0.045 1.000 ECO_PERF �0.062 �0.236 1.000 BOARD_INDEP 0.030 0.114 �0.344* 1.000 D_LENGTH �0.150 �0.149 �0.058 0.236 1.000

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110108

despite the IIRC's emphasis on integrating information systems to support internal and external reporting (Giovannoni and Fabietti, 2013). Although some companies may simply be in the initial

Table 8 Multivariate analysis: main logistic regression and sensitivity tests.

Main model Sensitivity tests

Coefficient Std. err. Coefficient Std. err.

Model 1 Dependent variable: SM_DISC SOC_PERF �0.000** 0.000 �0.000** 0.000 DISC_TYPE 1.118*** 0.085 1.184*** 0.085 DISC_TIME 0.360*** 0.105 0.347*** 0.105 DISC_LENGTH �0.005* 0.000 _cons �0.814 0.120 �0.731 0.130 Number of observations 3343 3343 Prob > c2 0.000 0.000 Pseudo R2 0.047 0.047

Model 2 Dependent variable: SM_DISC ENV_IMPACT 0.545** 0.227 0.427* 0.235 DISC_TYPE 1.244*** 0.103 1.252*** 0.103 DISC_TIME 0.324*** 0.125 0.299** 0.126 DISC_LENGTH �0.001*** 0.000 _cons �0.840 0.145 �0.699 0.161 Number of observations 2389 2389 Prob > c2 0.000 0.000 Pseudo R2 0.052 0.053

Model 3 Dependent variable: SM_DISC ECO_PERF 0.001 0.001 0.001 0.001 DISC_TYPE 1.041*** 0.071 1.038*** 0.071 DISC_TIME 0.369*** 0.095 0.353*** 0.095 DISC_LENGTH �0.001*** 0.000 _cons �0.749 0.110 �0.604 0.120 Number of observations 4426 4426 Prob > c2 0.000 0.000 Pseudo R2 0.038 0.039

Model 4 Dependent variable: SM_DISC BOARD_INDEP �0.004** 0.002 �0.004** 0.002 DISC_TYPE 1.334*** 0.092 1.338*** 0.093 DISC_TIME 0.286*** 0.108 0.286*** 0.108 DISC_LENGTH 0.000 0.000 _cons �0.629 0.170 �0.629 0.170 Number of observations 3206 3206 Prob > c2 0.000 0.000 Pseudo R2 0.052 0.052

Model 5 Dependent variable: SM_DISC ENV_SENS_IND 0.157** 0.063 0.157** 0.063 DISC_TYPE 1.037*** 0.071 1.035*** 0.071 DISC_TIME 0.376*** 0.095 0.360*** 0.095 DISC_LENGTH �0.000*** 0.000 _cons �0.807 0.112 �0.670 0.121 Number of observations 4426 4426 Prob > c2 0.000 0.000 Pseudo R2 0.040 0.040

***, **, *The estimated coefficient is statistically significant at the 1%, 5%, and 10% level, respectively.

stages of SM, and lack quantifiable performance indicators (Bennett et al., 2013), in general firms offer minimal information about ac- tions they plan to implement to achieve better sustainability per- formance in the future. The reasons of these deficiencies could be track down in the managers' concerns to release internal strategic information to competitors and threaten corporate legitimacy by disclosing poor achievements in terms of sustainability: they cannot only be explained by data unavailability.

The multivariate statistical analysis indeed reveals opportu- nistic behaviors in selecting the sustainability information to disclose. In line with previous critical studies (e.g., Milne and Gray, 2013), our findings suggest that IR disclosures draw a corporate veil (Hopwood, 2009) over the limited achievements of sustain- ability practices. Firms prefer to provide disclosures about their actions rather than their performance when their social and environmental performance is poor (H1a and H1b). We found no evidence of a significant association between SAD and economic performance, so managers apparently do not manipulate their IR disclosures to deal with poor economic performance. That is, managers choose IR as an IM vehicle when social and environ- mental performances are low, but not for economic performance. Our results are consistent with IM arguments; firms with weak social and environmental records use IR disclosure to detract attention from these results. These findings strengthen the idea of the use of the IR as a mean to opportunistically manage public impression on corporate behavior.

In further support of the presence of IM, when firms have weaker governance, managers take advantage of their high discretion and report more SAD, which is less verifiable by external stakeholders. In other words, a board that is less independent is not able to mitigate managerial opportunism in presenting information in a self-serving view. Similarly, firms belonging to an environ- mentally sensitive industry provide more SAD, likely because they are subject to more public pressure than other industry groups (Cho et al., 2010). This finding confirms the role of external pressures in driving sustainability reporting practices (Fernandez-Feijoo et al., 2014).

In turn, it is conceivable that firms consider IR a communica- tion vehicle they can use to alter perceptions of their sustainability achievements, and managers employ the discretion associated with which information to disclose (Bennett et al., 2013). The findings demonstrate the use of IR for managing corporate image. Being IM a plausible explanation of IR adoption, unlikely IR is fa- voring the integration among the departments involved in the processes of (external) sustainability reporting and the de- partments that are daily engaged in (internal) sustainability measurement and management processes (Adams and Frost, 2008; Bennett et al., 2013). Thus, pessimism about the capability of this reporting process to encourage integrative management of sustainability is warranted, and we answer our focal research question in the negative. In this respect, we contribute to the debate on the role of sustainability reporting for the management of sustainability issues (Adams and Frost, 2008; Bebbington, 2007; Daub, 2007; Fernandez-Feijoo et al., 2014; Lozano and Huisingh, 2011; Roca and Searcy, 2012), revealing that IR is not a condition for enhancing corporate sustainability.

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110 109

7.2. Research implications and limitations

For both managerial practices and research related to sustain- ability (Burritt and Schaltegger, 2010), the findings offer several implications. First, this study contributes to research into how companies can integrate sustainability performance management, measurement, and reporting efforts (Adams and Frost, 2008; Burritt, 2012; Schaltegger and Wagner, 2006a, 2006b). In this respect, we answer calls for research that crosses the boundaries of accounting and management disciplines to explicate sustainable development (Bebbington and Thomson, 2013). This empirical ev- idence details how firms commit, in terms of communicating their sustainable value creation processes. Despite the expectation that IR should favor integrated thinking (Adams, 2013), we show that IR adopters do not actually integrate SMA and SR. To our knowledge, this study is the first work to analyze IR in an effort to detect dis- closures of SM. Accordingly, the findings enrich research into the disclosure of sustainability indicators (Lodhia and Martin, 2014; Lozano, 2013; Roca and Searcy, 2012) by highlighting the limited disclosure of quantitative and forward-looking indicators in IR.

Second, this article contributes to the growing body of IR liter- ature with an empirical addition to the conceptual debate about the sustainability-related nature of IR (Adams, 2015; Burritt, 2012; Flower, 2015; Milne and Gray, 2013). Although the IIRF provides a reference point for informative descriptions of corporate engage- ment in sustainability issues, disclosures by IR adopters appear inadequate to represent actual commitment to managing sustain- ability. Our work thus raises concerns about the ability of IR to overcome the limits of other SR initiatives and encourage genuine business involvement with sustainability (Milne and Gray, 2013).

Third, we provide empirical support for the IM framework and confirm the influences of poor social and environmental perfor- mance, weak corporate governance, and membership in environ- mentally sensitive industries on disclosure strategies (Cho et al., 2010; García Osma and Guillam�on-Saorín, 2011; Mahoney et al., 2013). We know of no other research that has applied an IM framework to investigate IR disclosure practices related to different dimensions of sustainability. Similar to other reporting formats, IR is subject to thematic manipulations. In this respect, we widen the context of research on IM strategies; previous studies adopting this framework have focused solely on sustainability reports, financial reports, letters from the president and CEO, or press releases.

Fourth, for practitioners, this study highlights the state of the IR art: IR adopters are not following the IIRF in terms of communi- cating quantitative, forward-looking indicators of sustainability. The IIRC may need to offer additional guidelines about how com- panies should describe their sustainable value creation process, especially to include quantitative indicators and forward-looking information. For managers who have adopted or are about to adopt IR, our results caution that a lack of sound commitment to providing SM information will lead to a sense that their IR is a vehicle for IM strategies, not a tool to manage sustainability in an integrated manner. Finally, for investors and stakeholders, our re- sults show that in the current state, they cannot assume IR provide accounts of the internal information that managers use to make decisions about sustainability processes.

This research also contains several limitations that suggest di- rections for further research. We manually analyzed the section of the IR dedicated to sustainable value creation; additional research should extend this analysis to other parts of the report, such as the strategy or governance sections, or in other corporate reports, such as the financial ones, to determine whether and how sustainability is embedded in them. Further research also could examine corpo- rate communications that detail how sustainability actions link to particular performance indicators. We focused on particular

measures of corporate sustainability achievements, so our findings might be augmented by choosing different proxies for economic, social, and environmental performance, then investigating the links among the three dimensions (Lozano, 2013; Lozano and Huisingh, 2011), or considering other explanatory variables to represent corporate governance. Furthermore, the path toward IR is just starting, and these findings constitute a first contribution. By highlighting the shortcomings of disclosure practices by early IR adopters, we hope to spark further research that tracks the evolu- tion of IR over time and compares these practices with those ach- ieved in other forms of corporate reporting. Finally, the use of content analysis as a research approach has limitations; a trian- gulation with interviews and field studies would strengthen the analysis. Consistent with recommendations for a“practice turn” in disclosure studies (Lodhia and Jacobs, 2013), additional research should analyze interactions among different corporate de- partments to determine their role in achieving (or not) integration between SMA and SR.

Acknowledgements

The authors are grateful to the four anonymous reviewers, the Guest Editors Karen Maas, Nathalie Crutzen and Stefan Schaltegger, and the attendees of the 17th EMAN Conference, Rotterdam (The Netherlands), 27e28 March 2014. While the article is the result of a joint effort of the authors, the individual contributions are as fol- lows: Riccardo Stacchezzini wrote x 1, 3 and 7.2; Gaia Melloni wrote x 4, 5, 6 and 7.1; Alessandro Lai wrote x 2.

References

ACCA, 2012. Integrated Reporting: the Influence of King III on Social, Ethical and Environmental Reporting. www.accaglobal.com/content/dam/acca/global/PDF- technical/integrated-reporting/tech-tp-iirsa.pdf.

Adams, C.A., 2013. Understanding Integrated Reporting: the Concise Guide to In- tegrated Thinking and the Future of Corporate Reporting. Doshorts, London.

Adams, C.A., 2015. The International Integrated Reporting Council: a call to action. Crit. Perspect. Account. 27, 23e28.

Adams, C.A., Frost, G.R., 2008. Integrating sustainability reporting into management practices. Account. Forum 32, 288e302.

Adams, S., Fries, J., Simnett, R., 2011. The journey toward integrated reporting. Ac- count. Dig. 558, 1e41.

Baginski, S.P., Hassell, J.M., Hillison, W.A., 2000. Voluntary casual disclosures: ten- dencies and capital market reaction. Rev. Quant. Finance Account. 22, 371e389.

Beattie, V., McInnes, B., Fearnley, S., 2004. A methodology for analysing and eval- uating narratives in annual reports: a comprehensive descriptive profile and metrics for disclosure quality attributes. Acc. Forum 28, 205e236.

Bebbington, J., 2007. Changing organizational attitudes and culture through sus- tainability accounting. In: Unerman, J., Bebbington, J., O'Dwyer, B. (Eds.), Sus- tainability Accounting and Accountability. Routledge, London, pp. 226e242.

Bebbington, J., Thomson, I., 2013. Sustainable development, management and ac- counting: boundary crossing. Manag. Acc. Res. 24, 277e283.

Bennett, M., Schaltegger, S., Zvezdov, D., 2013. Exploring Corporate Practices in Management Accounting for Sustainability. Icaew, London, pp. 1e48.

Burritt, R.L., 2012. Environmental performance accountability: planet, people, profits. Account. Audit. Acc. J. 25, 370e405.

Burritt, R.L., Schaltegger, S., 2010. Sustainability accounting and reporting: fad or trend? Account. Audit. Acc. J. 23, 829e846.

Busco, C., Frigo, M.L., Riccaboni, A., Quattrone, P. (Eds.), 2013. Integrated Reporting. Concepts and Cases that Redefine Corporate Accountability. Springer, Cham.

Cho, C.H., Roberts, R.W., Patten, D.M., 2010. The language of US corporate envi- ronmental disclosure. Account. Org. Soc. 35, 431e443.

Cho, C.H., Freedman, M., Patten, D.M., 2012. Corporate disclosure of environmental capital expenditures: a test of alternative theories. Account. Audit. Acc. J. 25, 486e507.

Clatworthy, M., Jones, M.J., 2003. Financial reporting of good news and bad news: evidence from accounting narratives. Account. Bus. Res. 33, 171e185.

Daub, C.H., 2007. Assessing the quality of sustainability reporting: an alternative methodological approach. J. Clean. Prod. 15, 75e85.

Deegan, C., 2007. Organizational legitimacy as a motive for sustainability reporting. In: Unerman, J., Bebbington, J., O'Dwyer, B. (Eds.), Sustainability Accounting and Accountability. Routledge, London, pp. 127e149.

de Villiers, C., Rinaldi, L., Unerman, J., 2014. Integrated reporting: insights, gaps and an agenda for future research. Account. Audit. Acc. J. 27, 1042e1067.

R. Stacchezzini et al. / Journal of Cleaner Production 136 (2016) 102e110110

Eccles, R.G., Serafeim, G., Krzus, M., 2011. Market interest in non-financial infor- mation. J. Appl. Corp. Finance 23, 113e123.

Elkington, J., 1998. Cannibals with Forks: the Triple Bottom Line of 21st Century. New Society Publishers, Gabriola Island.

EY, 2013. Integrated Reporting. Communicating Sustainable Value Creation. http:// www.ey.com/Publication/vwLUAssets/Integrated_reporting_-_ Communicating_sustainable_value_creation/$FILE/EY-Integrated-Reporting. pdf.

Fernandez-Feijoo, B., Romero, S., Ruiz, S., 2014. Commitment to corporate social responsibility measured through global reporting initiative reporting: factors affecting the behavior of companies. J. Clean. Prod. 81, 244e254.

Figge, F., Hahn, T., Schaltegger, S., Wagner, M., 2002. The sustainability balanced scorecardelinking sustainability management to business strategy. Bus. Strat. Environ. 11, 269e284.

Flower, J., 2015. The International Integrated Reporting Council: a story of failure. Crit. Perspect. Account. 27, 1e17.

Frías-Aceituno, J.V., Rodríguez-Ariza, L., Garcia-S�anchez, I.M., 2012. The role of the board in the dissemination of integrated corporate social reporting. Corp. Soc. Resp. Env. Manag 20, 219e233.

Frías-Aceituno, J.V., Rodríguez-Ariza, L., García-S�anchez, I.M., 2013. Is integrated reporting determined by a country's legal system? An exploratory study. J. Clean. Prod. 44, 45e55.

Frías-Aceituno, J.V., Rodríguez-Ariza, L., Garcia-S�anchez, I.M., 2014. Explanatory factors of integrated sustainability and financial reporting. Bus. Strat. Environ. 23, 56e72.

García Osma, B., Guillam�on-Saorín, E., 2011. Corporate governance and impression management in annual results press releases. Account. Org. Soc. 36, 187e208.

Giovannoni, E., Fabietti, G., 2013. What is sustainability? A review of the concept and its applications. In: Busco, C., Frigo, M.L., Riccaboni, A., Quattrone, P. (Eds.), Integrated Reporting. Concepts and Cases that Redefine Corporate Account- ability. Springer, Cham, pp. 21e40.

Hahn, R., Kühnen, M., 2013. Determinants of sustainability reporting: a review of results, trends, theory, and opportunities in an expanding field of research. J. Clean. Prod. 59, 5e21.

Hooghiemstra, R., 2000. Corporate communication and impression management e new perspectives why companies engage in corporate social reporting. J. Bus. Ethics 27, 55e68.

Hopwood, A.G., 2009. Accounting and the environment. Account. Org. Soc. 34, 433e439.

ICAEW, 2013. Integrated Reporting Is a Catalyst to Improve Business Reporting Says ICAEW. http://www.icaew.com/en/archive/about-icaew/newsroom/press- releases/2013-press-releases/integrated-reporting-is-a-catalyst-to-improve- business-reporting-says-icaew.

International Integrated Reporting Committee, 2013a. The International <IR> Framework. http://www.theiirc.org/wp-content/uploads/2013/12/13-12-08- THE-INTERNATIONAL-IR-FRAMEWORK-2-1.pdf.

International Integrated Reporting Committee, 2013b. Business and Investors Explore the Sustainability Perspective of Integrated Reporting. http://www. theiirc.org/wp-content/uploads/2013/12/IIRC-PP-Yearbook-2013_PDF4_PAGES. pdf.

Jensen, J., Berg, N., 2012. Determinants of traditional sustainability reporting versus integrated reporting. An institutionalist approach. Bus. Strat. Environ. 21, 299e316.

Kaplan, R.S., Norton, D.P., 1996. The Balanced Scorecard: Translating Strategy into Action. Harvard Business Press, Boston.

King, M., Roberts, L., 2013. Integrate. Doing Business in the 21st Century. Juta and Company, Claremont.

KPMG, 2013. Integrated Reporting. Performance Insight through Better Business Reporting. http://www.kpmg.com/Global/en/IssuesAndInsights/ ArticlesPublications/Documents/road-to-integrated-reporting.pdf.

Lai, A., Melloni, G., Stacchezzini, R., 2014. Corporate sustainable development: is “integrated reporting” a legitimation strategy? Bus. Strat. Environ. http://dx.doi. org/10.1002/bse.1863.

Larrinaga-Gonz�ales, C., 2007. Sustainability reporting: insights from neoinstitu- tional theory. In: Unerman, J., Bebbington, J., O'Dwyer, B. (Eds.), Sustainability Accounting and Accountability. Routledge, London, pp. 150e167.

LEAD, 2013. Comments to the IIRC's Consultation Draft of the International Inte- grated Reporting Framework. http://www.theiirc.org/wp-content/uploads/ 2013/08/125_UN-Global-Compact.pdf.

Leary, M.R., Kowalski, R.M., 1990. Impression management: a literature review and two-component model. Psychol. Bull. 107, 34e47.

Lodhia, S., Jacobs, K., 2013. The practice turn in environmental reporting: a study into current practices in two Australian commonwealth departments. Account. Audit. Account. J. 26, 595e615.

Lodhia, S., Martin, N., 2014. Corporate sustainability indicators: an Australian mining case study. J. Clean. Prod. 84, 107e115.

Lozano, R., 2013. Sustainability inter-linkages in reporting vindicated: a study of European companies. J. Clean. Prod. 51, 57e65.

Lozano, R., Huisingh, D., 2011. Inter-linking issues and dimensions in sustainability reporting. J. Clean. Prod. 19, 99e107.

Mahoney, L.S., Thorne, L., Cecil, L., LaGore, W., 2013. A research note on standalone corporate social responsibility reports: signaling or greenwashing? Crit. Per- spect. Account. 24, 350e359.

Mather, P., Ramsay, A., 2007. Do board characteristics influence impression manage- ment throughgraph selectivityaroundCEO changes? Aust. Account. Rev.17, 84e95.

Merkl-Davies, D.M., Brennan, N.M., 2007. Discretionary disclosure strategies in corporate narratives: incremental information or impression management? J. Account. Lit. 26, 116e196.

Milne, M.J., Gray, R., 2013. W(h)ither ecology? The triple bottom line, the global reporting initiative, and corporate sustainability reporting. J. Bus. Ethics 118, 13e29.

Neu, D., Warsame, H., Pedwell, K., 1998. Managing public impressions: environ- mental disclosures in annual reports. Account. Org. Soc. 23, 265e282.

O'Dwyer, B., 2003. Conceptions of corporate social responsibility: the nature of managerial capture. Account. Audit. Acc. J. 16, 523e557.

PWC, 2013. Integrated Reporting: Companies Struggle to Explain What Value They Create. http://www.pwc.nl/nl_NL/nl/assets/documents/ir-opmars- geintegreerde-verslaggeving-stagneert.pdf.

Roca, L.C., Searcy, C., 2012. An analysis of indicators disclosed in corporate sus- tainability reports. J. Clean. Prod. 20, 103e118.

Schaltegger, S., Wagner, M., 2006a. Integrative management and sustainability performance, measurement and reporting. Int. J. Account. Audit. Perform. Eval. 3, 1e19.

Schaltegger, S., Wagner, M., 2006b. Managing sustainability performance mea- surement and reporting in an integrated manner. In: Schaltegger, S., Bennett, M., Burritt, R. (Eds.), Sustainability Accounting and Reporting. Springer, Dordrecht, pp. 681e697.

Schaltegger, S., Zvezdov, D., 2015. Gatekeepers of sustainability information. Exploring the roles of accountants. J. Account. Organ. Change 11, 333e361.

Schleicher, T., Walker, M., 2010. Bias in the tone of forward-looking narratives. Account. Bus. Res. 40, 371e390.

Sierra-García, L., Zorio-Grima, A., García-Benau, M.A., 2015. Stakeholder engage- ment, corporate social responsibility and integrated reporting: an exploratory study. Corp. Soc. Resp. Environ. Manag. 22 (5), 286e304. http://dx.doi.org/10. 1002/csr.1345.

Tregidga, H., Milne, M., Kearins, K., 2014. (Re-)presenting ‘sustainable organiza- tions’. Account. Org. Soc. 39, 477e494.