Organizational Slack, Corporate Social Responsibility, Sustainability, and Integrated Reporting: Evidence from Korea

Abstract

This paper examines whether organizational slack is associated with firms’ voluntary disclosure of corporate social responsibility (CSR), sustainability, and integrated reporting. This is an empirical research study using archival data based on a sample of public firms listed on the Korea Exchange from 2005 to 2016. We manually collected CSR reports, sustainability reports, and integrated reports (IRs) that were published during our sample period. We found that human resource slack was highly related to the publication of corporate social responsibility, sustainability, and integrated reports. Firms initiating such disclosure in their industry groups were likely to have slack in permanent employees as well as financial slack. Additionally, integrated reporting, which is a recent improvement in the delivery of financial and non-financial information, was positively associated with an excess number of regular employees. This study provides evidence that slacks in regular employees are related to a greater degree of voluntary disclosure via standalone CSR or sustainability reporting as well as integrated reporting. These findings suggest that slacks or excess human resources play a crucial role in voluntary corporate disclosure.

1. Introduction

Recently, there has been an increasing demand for information beyond financial statements by capital providers and stakeholders, because of the limitation of financial reporting in reflecting important business strategies, intangibles, and long-term corporate value [1]. One of the major channels of non-financial voluntary disclosure has been standalone corporate social responsibility (CSR) and sustainability reports. More recent movement has been to integrate financial and non-financial information in a single report (i.e., so-called integrated reporting, <IR>), thereby enhancing the conciseness of corporate disclosures [2].

There is voluminous literature on non-financial voluntary disclosures. Many previous studies have documented that voluntary disclosures are affected by several firm characteristics, such as size, leverage, listing, managerial ownership, large negative earnings surprises, cost of equity, and debt capital [3,4,5,6]. Especially, for voluntary non-financial disclosures like sustainability and CSR reporting, the literature also shows that such disclosures provide value-relevant information [7,8,9]. One of the most important motivations for such disclosures is to reduce the cost of capital. For example, Dhaliwal et al. [10] showed that CSR disclosure is initiated by firms with a higher cost of equity capital in the previous year, and that initiating firms with superior CSR performance exhibit a subsequent reduction in the cost of equity capital. In addition, companies facing economic pressure from the general public and government are more likely to disclose [11]. Recently, Baloria et al. [12] found that the shareholder activism mechanism motivates such ESG (Environmental, Social, Governance) disclosure. Additionally, prior studies have documented the economic consequences of CSR reporting, such as higher liquidity (lower information asymmetry) and expected future cash flows [13,14], as well as positive returns to shareholders [15].

Despite all these benefits and incentives, managers do not adequately provide their private information in financial and non-financial disclosures due to the costs of such disclosure activity, including proprietary costs. In consideration of trade-off between benefits and costs, empirical evidence substantiating the affordability of voluntary non-financial disclosure remains sparse. As companies are not mandated to publish CSR, sustainability, and integrated reports in many countries, managers may modify their disclosure behavior depending on how much slack resources they have. Hence, this paper examines whether organizational slack affects the voluntary disclosure of CSR, sustainability, and integrated reports.

This paper explores the explanation for why firms differ in their voluntary disclosure of non-financial information based on the “slack resource theory” [16]. Organizational slacks include financial and human resource slacks [17]. Prior research [18,19] has focused on slack in these two resources, which are closely related to firms’ emergence and development. Financial slacks usually represent unabsorbed slack that is currently uncommitted and available for redeployment [19], such as cash and receivables that are highly flexible and can be disbursed on managerial discretion. The more flexible the capital, the more affordable any additional voluntary disclosure becomes [20]. Human resource slack is considered as absorbed slack, as human resources are highly idiosyncratic to context and difficult to redeploy [19]. Usually, human resource slack refers to having more than average human resources or experienced personnel [21]. Firms having more than average employees will be able to adjust the slack in human resources to collect relevant information and create an additional publication of CSR and sustainability reports. Hence, we hypothesize a positive association between human resource/financial slack and firms’ publication of CSR, sustainability, and integrated reports.

Testing our research question in the Korean setting is justified because of the nature of the Korean employment environment. According to the OECD (Organization for Economic Cooperation and Development) reports, South Korea lags behind principal members of the OECD in job security for regular employees in the private sector. For example, South Korean companies are more active in layoffs when they undergo financial distress due to lax state regulations on the business sector’s retrenchment. Moreover, it is more difficult for temporary workers to become permanent employees than it is for those in other countries to make the same transition. While the average proportion of a job status transition after three years of work is 53.8% in OECD countries, only 11.1% of Korean nonregular workers were hired for regular positions after a year, and 22.4% after three years (this is based on the article retrieved from The Korea Herald, titled “Korea behind OECD average in job security” on 8 December 2014). However, the share of regular workers in the total labor force of Korea has grown from 43.9% in 2005 to 55.5% in 2016, while that of nonregular workers has dropped from 56.1% to 44.5% during this period (employment data were retrieved from the “Economically Active Population Survey” in December 2017, Korean Statistical Information Service). This hike in permanent employment has been viewed as unproductive and inefficient from a strategic perspective. On the contrary, we challenge the perspective by investigating whether slack in the traditional sense could create value. Specifically, we pay attention to the firm valuation effect of voluntary corporate disclosure regarding CSR and sustainability through facilitating future cash flows prediction [3] and/or reducing the cost of capital [4,10]. Hence, our investigation on how sustainability and CSR disclosures are affected by human resource slack (especially slack in regular employees) will have social and policy implications in countries with low job security like South Korea. Moreover, despite the importance of the topic, none of the prior studies examined the relationship between organizational slack and voluntary non-financial disclosure in many countries, including Korea.

The sample used for empirical analyses contained 4186 firm-year observations, consisting of firms listed on the KSE (Korea Stock Exchange) and KOSDAQ from 2005 to 2016. Our empirical findings are summarized as follows. First, human resource slack was significantly and positively associated with the publication of CSR, sustainability, and integrated reports. Second, among these firms, those with slacks in permanent employees were more likely to initiate such voluntary non-financial disclosures in their industry groups. Additionally, initiating firms had a greater extent of financial slacks than non-initiating firms. Lastly, the industry-level analysis showed that slacks in permanent employees were related to the firm’s compliance with a recent movement to integrate financial and non-financial disclosure in one report by publishing integrated reports. These results indicate that excess human resources and financial flexibility could be utilized to provide additional non-financial information about long-term firm value to various stakeholders. More importantly, employees’ experience in their job functions and firm-specific knowledge are crucial factors in determining the firm’s initiation of non-financial voluntary reporting when none of their industry peers provide such information, as well as in publishing integrated reports, which could enhance the conciseness of financial and non-financial information.

This paper makes several contributions to the existing literature in business and accounting. First, our findings provide new evidence that organizational slacks are significantly related to firms’ non-financial voluntary disclosure behavior. By viewing excess human and financial resources as the explanatory variables of voluntary disclosure, this paper differentiates from previous studies that have focused on incentives related to capital markets as the explanatory variable (e.g., [10]). Next, the findings in the current study have social and practical implications related to permanent employment. Traditional views on excess permanent employees are generally negative due to inefficiency and inflexibility. However, we found that slacks in regular employees were related to a greater degree of voluntary disclosure via standalone CSR or sustainability reporting. Additionally, integrated reporting—which is a recent improvement in the delivery of financial and non-financial information—was positively associated with an excess number of regular employees. Hence, we provide fresh evidence that human resource slack or excess of permanent employment can make it easier for firms to develop voluntary disclosure, which could positively affect firm value.

The rest of this paper is organized as follows. Section 2 explains the background of CSR and sustainability reporting practice in South Korea. Section 3 reviews prior studies and develops our hypotheses. Section 4 presents our research methods and sample selection, and Section 5 shows the results of empirical analyses. Section 6 concludes the study.

2. Background

Since the early 2000s, companies have provided information regarding their social and environmental performances and business strategies through non-financial voluntary disclosures [2,22]. Specifically, one of the major channels of non-financial voluntary disclosure has been standalone corporate social responsibility and sustainability reports. These reports, in general, cover firms’ environmental performance, working condition, social responsibility, human rights, products, and so forth. Many companies produce these reports as an attempt to increase corporate transparency and accountability, and to improve internal processes and engage stakeholders.

In Korea, ever since POSCO (a leading manufacturer of primary metal products in Korea) published a sustainability report in 2003 for the first time, the incidence of the publication of CSR and corporate sustainability reports has increased, as shown in

Table 1. Publication of corporate social responsibility (CSR) and sustainability reports by industry and year.

	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016	Total
Manufacture of Basic Chemicals	0	0	2	2	4	3	3	3	3	3	4	5	32
Manufacture of Other Chemical products	0	0	0	1	2	2	2	2	2	4	3	3	21
Manufacture of Man-Made Fibers	0	0	0	0	0	0	0	1	0	1	0	1	3
Manufacture of Rubber and Plastic Products	0	0	0	0	0	0	0	2	2	2	2	2	10
Manufacture of Other Non-Metallic Mineral Products	0	0	0	0	0	0	0	0	1	2	1	1	5
Manufacture of Basic Metal Products	1	0	1	2	1	2	1	1	1	1	1	0	12
Manufacture of Electronic Components	0	0	0	0	1	1	1	2	2	2	3	3	15
Manufacture of Telecommunication and Broadcasting Apparatuses	1	1	0	2	2	2	2	2	2	2	2	2	20
Manufacture of Other Machinery and Equipment	0	0	0	0	1	1	2	1	1	1	1	1	9
General Construction	0	0	0	0	0	0	0	1	1	1	1	2	6
Air Transport	0	1	1	1	1	2	2	2	2	2	2	2	18
Postal Services and Telecommunications	0	0	0	0	0	0	0	1	1	1	1	1	5
Professional Services	0	1	2	2	2	3	3	4	4	4	3	3	31
Architectural, Engineering, and Other Scientific Technical Services	0	0	0	0	0	0	0	2	1	2	1	2	8
No. of firms	2	3	6	10	14	16	16	24	23	28	25	28	195
No. of industries	2	3	4	6	8	8	8	13	13	14	13	13	105

. We manually collected CSR, sustainability, and integrated reports published by Korean companies from 2005 to 2016 from the sustainability report database of the Korean Standard Association (https://www.ksa.or.kr/ksi/4982/subview.do) and sampled firms’ websites. One notable observation from our data is that such non-financial voluntary disclosures are concentrated in few industries, such as the manufacture of basic chemicals, manufacture of telecommunication and broadcasting apparatuses, and air transportation. This observation is in line with prior studies [23] documenting that specialist industries such as chemical and petroleum industries seek greater CSR performance to prevent boycotting by stakeholders, including environmentalists.

Though there is no audit requirement, there are common guidelines for preparing these reports, such as SIGMA, Corporate Responsibility Guidelines, ETI Base Code, GRI (Global Reporting Initiative), and BSR guideline. However, the relevance of CSR and sustainability reports is under criticism because of the sheer volume of information contained without identification of relation to information contained in the annual reports [2]. This lack of integration of financial and non-financial information especially reduces usefulness to capital providers. Hence, a fairly new movement is to enhance the conciseness of current corporate reporting, and one of them is the integrated reporting framework—<IR> [24].

In 2014, the European Commission proposed new corporate disclosure by mandating <IR>, and the first company reports were to be published in 2018 covering the fiscal year 2017–2018 (GRI 2018). In Korea, few companies are publishing integrated reports that contain both financial and non-financial information in a single report. The first publisher of an integrated report was SK Telecom in 2005. As shown in

Table 2. Publication of integrated reports by industry and year.

	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016	Total
Manufacture of Basic Metal Products	0	0	0	0	0	0	1	1	1	1	1	2	7
Manufacture of Other Machinery and Equipment	0	0	0	0	0	0	0	1	1	1	1	1	5
Postal Services and Telecommunications	1	1	2	2	2	2	2	2	2	2	2	2	22
Professional Services	0	0	0	0	0	0	0	0	0	0	1	1	2
No. of firms	1	1	2	2	2	2	3	4	4	4	5	6	36
No. of industries	1	1	1	1	1	1	2	3	3	3	4	4	25

, companies in postal services and telecommunications appear to be the leaders in the publication of integrated reports. However, companies in other industries did not start issuing integrated reports until 2011. Note that POSCO has published integrated reports instead of sustainability reports since 2011.

3. Literature Review and Hypotheses Development

Firms engage in voluntary communication with investors and other stakeholders, such as management forecasts, analysts’ presentations and conference calls, press releases, internet sites, and other corporate reports. These voluntary disclosures may contain financial and non-financial information. One of the major channels of non-financial voluntary disclosure has been standalone sustainability and CSR reports. Standalone CSR reports have been documented to be value-relevant, although they are subject to limited regulation. Prior studies [7,8,9,25] have documented a positive relation between CSR disclosure and firm value (or performance). CSR disclosure also reduces information asymmetry related to factors affecting firm value [13,14,26], thereby reducing the cost of equity capital [10,24].

Previous studies have documented various motivations for such voluntary non-financial disclosure. For instance, according to Dhaliwal et al. [10], CSR disclosures are initiated by firms with a higher cost of equity capital in the previous year. Among these firms, those with superior CSR performance exhibit a subsequent reduction in the cost of equity capital. In addition, there is a signaling hypothesis (e.g., [27]) that corporate accountability reporting signals insiders’ private information about firms’ future financial prospects to outsiders. Moreover, various studies have examined other sources of corporate motivation for CSR and sustainability disclosure. Luo et al. [11] show that economic pressure is significantly associated with the disclosure, based on the findings that climate change disclosure is driven by pressure from the general public and government. Shareholder activism also motivates such disclosure, in that investors express their needs for non-financial disclosure through shareholder proposals [12]. Voluntary CSR disclosure is also found to be affected by politics, specifically company political interests [28]. Finally, based on reputation theory, prior studies have argued that CSR activity or disclosure is used for crisis management. For example, positive CSR offsets the damage from product harm crisis [29], the 2008–2009 financial crisis [30], and declining stage of corporate life cycle [31].

However, voluntary publication of CSR and sustainability reports requires a firm’s “special effort and commitment to improving transparency regarding long-term performance” [10]. For firms without sufficient financial or human resources at their disposal, non-mandatory corporate disclosures may not be affordable. Likewise, firms having more operational slack (i.e., less financial constraints) are more likely to undertake special projects [21]. These special projects can include the voluntary disclosure of CSR and sustainability reports.

It is vital to distinguish CSR disclosures and CSR performance ratings. Third parties generally provide CSR performance ratings. For instance, CSR rating agencies like the KLD Research & Analytics assess large firms’ environmental, social, and governance characteristics and provide social indices such as KLD Social Index and Dow Jones Social Index. Prior research shows that these indices are associated with the cost of equity capital [32,33]. There are also several prior studies examining the relationship between organizational slack and CSR performance (e.g., [34,35,36]). For example, Hong et al. [37] show that financial constraints are an important driver of corporate goodness. Firms are more likely to do good when they do well. Likewise, focusing on cash flow, which is the most discretionary type of organizational slack, Seifert et al. [38] show that such slack positively affects monetary donations that proxy for corporate philanthropy. Additionally, Shahzad et al. [20] found a negative relationship between financial and innovational slack and corporate social performance, giving credence to the agency theory perspective. On the other hand, Symeou et al. [39] and Xu et al. [40] found a positive (negative) association between unabsorbed (absorbed) slack and environmental and corporate social performance, based on the behavioral theory of the firm. As such, the relation between slack and CSR is inconclusive.

Whereas CSR performance index or ratings are limited in providing detailed information about firms’ overall CSR activities, CSR and sustainability disclosures contain comprehensive information and sufficient details for stakeholders to understand the summary ratings. Information in these disclosures is incrementally useful for investors to evaluate firms’ long-term sustainability and long-term value. For this reason, we focus on standalone CSR and corporate sustainability reports, which are the new forms of voluntary nonfinancial disclosure.

Organizational slack can be considered a prerequisite for firms’ efforts and commitment to CSR and sustainability reporting. Specifically, firms having more than average employees will be able to adjust their slack in human resources to collect relevant information and create a voluntary publication of CSR and sustainability reports. Additional voluntary disclosure is also more affordable for a firm with financial slacks that are highly flexible and disbursed at managers’ discretion (i.e., cash and receivables). Hence, we hypothesize a positive association between human resource/financial slack and firms’ publication of CSR and sustainability reports.

Hypothesis 1.

There will be a positive association between organizational slack and the publication of standalone CSR and sustainability reports.

Additionally, prior research has shown that voluntary corporate disclosure such as earnings warnings is affected by industry peer firms’ actions [41]. The phenomenon that managers’ decisions to provide voluntary disclosure such as pro forma earnings or conference calls are affected by industry peers’ decisions is called “herding” or “social learning” [42]. Such herding behavior is also observed in providing non-financial disclosures like CSR or sustainability reports. Yang et al. [23] state the peer effects of CSR strategies by showing that firms tend to improve the CSR performance or disclosure when there is a CSR gap between the firms and their peers. Their reasoning is based on market recognition, in that the market distinguishes high-quality firms from low-quality firms according to their CSR performance or disclosure. As a result, low-quality firms imitate high-quality peers in order to receive more favorable recognition from the market.

Managers are likely to lack their reference to observe and consider peers’ actions when initiating the voluntary publication of CSR and sustainability reports if they are the first one in their industry. Hence, it will require additional efforts and resources to be the first one publishing CSR and sustainability reports in an industry.

Hypothesis 2.

Firms with organizational slack are more likely to initiate the voluntary publication of standalone CSR and sustainability reports in their industries.

Our final hypothesis focuses on integrated reporting, which is an emerging corporate disclosure approach that integrates both financial and non-financial information. Integrated reporting combines a firm’s separate reporting strands into a single concise report, which contains strategically relevant information on a firm’s ability to create and sustain short-term and long-term value [43]. The number of companies voluntarily producing integrated reports and the regulations that pay attention to integrated reporting are growing around the world. In addition, as shown in Zhou et al. [24], the information contained in an integrated report is incrementally useful to investors and analysts over and above the current reporting suite.

We postulate that it will require additional human resources to collect both financial and non-financial information that is relevant and material, and to then process them into readable and concise reports. Moreover, in-depth firm knowledge will be premises for the integration of financial and non-financial information that should improve the firm’s information environment. Such knowledge and skills are likely to be possessed by regular employees who have held their positions relatively longer than temporary employees. Hence, we predict slacks in permanent employees to be positively related to the issuance of integrated reports. Accordingly, the final hypothesis is stated as follows:

Hypothesis 3.

There will be a positive association between permanent employee slack and integrated reporting.

4. Empirical Methodology

4.1. Data and Sample

The sample for this study was the companies listed on the KSE and KOSDAQ from 2005 to 2016. We collected financial data from the FN-Guide and KIS-Value databases (which are equivalent to COMPUSTAT in the U.S.). We only included firms of industries in which non-financial disclosure (CSR report, sustainability report, and integrated report) were published at least once. Next, we excluded firms with insufficient data. Therefore, our final sample was 4186 firm-years.

Table 3. Incidence of CSR, sustainability, and integrated reporting by year.

Year	Not Published	Published	Total
2005	84	3	87
2006	158	4	162
2007	149	8	157
2008	231	12	243
2009	338	16	354
2010	355	18	373
2011	372	19	391
2012	424	28	452
2013	441	27	468
2014	457	32	489
2015	468	30	498
2016	478	34	512
Total	3955	231	4186

presents the incidence of the publication of CSR, sustainability, and integrated reports by year. The number of firms issuing such non-financial disclosure grew each year. The percentage of firms publishing CSR, sustainability, and integrated reports were 3.44% and 2.46% in years 2005 and 2006, respectively. This percentage went up to 6.02% and 6.64% in recent years 2015–2016.

4.2. Research Models

We employed a logit analysis to examine the relationship between organizational slack and the non-financial voluntary disclosure (Hypotheses 1 and 2). We also adjust the standard errors for firm-level clustering and controlled for the year effects in testing our hypotheses. The following Equation (1) was estimated:

$$\begin{gathered} \mathit{NON}\_{\mathit{FIN}}_{\mathit{it}}\mathit{or}{\mathit{INITIATE}}_{\mathit{it}}={\beta }_0+{\beta }_1\mathit{SLACK}\_{\mathit{HR}}_{\mathit{it}}+{\beta }_2\mathit{SLACK}\_{\mathit{PERM}}_{\mathit{it}}+ \\ {\beta }_3\mathit{SLACK}\_{\mathit{FIN}}_{\mathit{it}}+{\beta }_4{\mathit{SIZE}}_{\mathit{it}}+{\beta }_5{\mathit{ROA}}_{\mathit{it}}+{\beta }_6{\mathit{LARGEST}}_{\mathit{it}}+{\beta }_7{\mathit{FOR}}_{\mathit{it}}+ \\ {\beta }_8{\mathit{EMP}}_{\mathit{it}}+{\beta }_9\mathit{SD}\_{\mathit{SLACK}1}_{\mathit{it}}+{\beta }_{10}\mathit{SD}\_{\mathit{SLACK}2}_{\mathit{it}}+{\beta }_{11}\mathit{SD}\_{\mathit{SLACK}3}_{\mathit{it}}+ \\ \mathit{Year}\mathit{dummies}+{\epsilon }_{\mathit{it}}, \end{gathered}$$

(1)

where

• NON_FIN: 1 if a firm i publishes voluntary non-financial disclosure (CSR, sustainability, integrated report) in year t, otherwise 0;
• INITIATE: 1 if a firm initiates voluntary non-financial disclosure (CSR, sustainability, integrated report) in the industry, otherwise 0;
• SLACK_HR: slack in the number of employees = ln(number of employees) − ln(industry average number of employees);
• SLACK_PERM: slack in permanent employees = ratio of permanent employees − industry average ratio of permanent employees;
• SLACK_FIN: financial slack = (current liability—current assets)/total assets − (industry average current liability − industry average current assets)/industry average total assets;
• SIZE: ln(total assets);
• ROA: return on assets (net income/lagged total assets);
• LARGEST: the largest shareholders’ ownership (sum of ownership of the largest shareholders and related parties);
• FOR: foreign investors’ ownership (sum of ownership of the largest shareholders and related parties);
• EMP: ln(number of employees);
• SD_SLACK1: standard deviation of slack of number of employees;
• SD_SLACK2: standard deviation of slack of ratio of permanent employees;
• SD_SLACK3: standard deviation of financial slack.

In testing Hypothesis 1, our dependent variable (NON_FIN) was based on non-financial voluntary disclosure, which was set to 1 if a firm published a CSR report, corporate sustainability report, or integrated report, and 0 otherwise. For Hypothesis 2, the dependent variable (INITIATE) was the indicator of whether a firm was the leader of non-financial voluntary disclosure in its industry, based on the first publication of CSR, sustainability, and integrated reports.

Our test variable was the organizational slack. Following management and organization research [17,44,45], we focused on two types of organizational slack: human resource slack and financial slack. We measured the human resource slack (SLACK_HR) using the natural logarithm of the number of employees more than the industry average. This assumes that a firm with an abnormally greater number of employees than others will have more available human resources (i.e., slacks). Financial slack (SLACK_FIN) is usually defined as financial assets available (e.g., cash on hand) that can be quickly deployed to varied uses [46]. Many studies (e.g., [36,47]) have used current ratio as the proxy of potential slack since it indicates that the firm has liquid resources above and beyond what is required to meet the immediate obligation. Hence, we used the ratio of (current liability - current assets)/total assets, by taking the difference from the industry average. In addition, we defined and measured the additional measure for slack in regular employment (SLACK_PERM) to examine the effect of an excess number of permanent employees on the corporate disclosure decision. SLACK_PERM was calculated by the difference between the firm’s ratio of permanent employees and its industry’s average. If firms having greater levels of human resource slack or financial slack are more likely to produce (or initiate) non-financial voluntary disclosure, the coefficients of SLACK_HR and SLACK_PERM would be positive and that of SLACK_FIN would be negative.

We controlled for several firm-specific characteristics associated with non-financial voluntary disclosure according to prior literature [3,4,5,6], such as firm size (SIZE, EMP), return on assets (ROA), most substantial shareholder’s ownership (LARGEST), and foreign investors’ ownership (FOR). Lastly, we added the standard deviations of slack variables (SD_SLACK) to control for changes in a firm’s human resource slacks and financial slack.

Next, we estimated the following regression Equation (2) to examine Hypothesis 3 on the relationship between organizational slack and integrated reporting. The variables used in Equation (2) were measured at the industry level because we had only 36 firms that had published integrated reports.

$$\begin{gathered} I{R}_{id,t}={\beta }_0+{\beta }_{1}SLACK\_PER{M}_{id,t}+{\beta }_{2}SIZ{E}_{id,t}+{\beta }_{3}RO{A}_{id,t}+ \\ {\beta }_{4}LARGES{T}_{id,t}+{\beta }_{5}FO{R}_{id,t}+{\beta }_{6}EM{P}_{id,t}+{\beta }_{7}SD\_SLACK{2}_{id,t}+ \\ Yeardummies+{\epsilon }_{id,t}, \end{gathered}$$

(2)

where IR has a value of 1 if an integrated report is published in the industry id in year t, otherwise it has value 0.

In Equation (2), all independent variables were measured using the industry average. If the firms in the industry having greater levels of permanent employee slack are likely to engage in integrated reporting, the estimated coefficient of SLACK_PERM would be positive.

5. Empirical Test Results

5.1. Descriptive Statistics

Table 4. Descriptive statistics.

(a) Full sample (N = 4186)
Variables	Mean	Median	Max	Min	SD
SLACK_HR	−1.3583	−1.4125	2.1980	−4.2473	1.4799
SLACK_Perm	0.0031	0.0128	0.2132	−0.7122	0.1076
SLACK_Fin	0.0033	0.0154	0.5280	−0.5311	0.2151
SIZE	26.1237	25.8454	30.6861	23.1827	1.4893
ROA	0.0230	0.0327	0.2323	−0.6229	0.0991
LARGEST	0.4165	0.4160	0.8000	0.0797	0.1581
FOR	0.0736	0.0221	0.5848	0.0000	0.1133
EMP	5.4451	5.3519	9.4011	2.0794	1.4561
(b) Published vs. Not-published firms
	Published (N = 231)	Not-Published (N = 3955)			Difference
Variables	Mean	Median	Mean	Median	Mean (t-Statistics)
SLACK_HR	1.2483	1.6183	−1.5105	−1.5090	2.7589 (30.43)
SLACK_Perm	−0.0132	0.0047	0.0040	0.0142	−0.0172 (−2.36)
SLACK_Fin	0.0922	0.0834	−0.0019	0.0093	0.0941 (6.49)
SIZE	29.4325	29.8095	25.9304	25.7460	3.5021 (41.17)
ROA	0.0401	0.0364	0.0221	0.0326	0.0181 (2.69)
LARGEST	0.3654	0.3482	0.4194	0.4186	−0.0540 (−5.06)
FOR	0.2627	0.2430	0.0625	0.0192	0.2002 (28.53)
EMP	8.3491	8.7114	5.2755	5.2679	3.0736 (35.58)
(c) Initiating vs. Not-initiating firms
	Initiating (N = 114)	Not-initiating (N = 117)			Difference
Variables	Mean	Median	Mean	Median	Mean (t-Statistics)
SLACK_HR	1.3925	1.6527	1.1078	1.5577	0.2847 (2.04)
SLACK_Perm	0.0056	0.0077	−0.0315	0.0000	0.0372 (2.34)
SLACK_Fin	0.0614	0.0167	0.1222	0.1283	−0.0608 (−3.31)
SIZE	29.8369	30.2391	29.0385	29.3307	0.7984 (4.82)
ROA	0.0435	0.0389	0.0369	0.0360	0.0066 (0.82)
LARGEST	0.3642	0.3380	0.3666	0.3630	−0.0024 (−0.12)
FOR	0.3093	0.3117	0.2174	0.1385	0.0920 (3.83)
EMP	8.6483	8.8423	8.0575	8.5057	0.5908 (3.67)
(d) Integrated reports <IR> publication vs. Non-<IR> publication
	<IR> (N = 25)	Non <IR> (N = 89)			Difference
Variables	Mean	Median	Mean	Median	Mean (t-Statistics)
SLACK_HR	0.2304	−0.1869	0.1812	−0.2029	0.0492 (0.22)
SLACK_Perm	−0.0028	0.0000	−0.0120	0.0000	0.0092 (1.03)
SLACK_Fin	0.0078	0.0018	0.0294	−0.0009	−0.0216 (−0.86)
SIZE	26.3471	26.4390	26.6186	26.4665	−0.2715 (−1.00)
ROA	0.0061	0.0107	0.0230	0.0286	−0.0168 (−1.97)
LARGEST	42.8253	43.5156	42.3761	42.6906	0.4492 (0.34)
FOR	0.1199	0.1030	0.0703	0.0710	0.0496 (6.04)
EMP	5.6170	5.5905	6.1334	5.6621	−0.5164 (−1.90)

Notes: Variables are defined in Equation (1).

reports the descriptive statistics of the key variables used in the study. Based on the full sample (panel a), the mean of variable SLACK_HR was −1.3583, implying that, on average, sample firms had fewer employees than the industry average. However, as shown in panel b (panel c), publishing (initiating) firms had significantly greater human resource slack than non-publishers (followers). The mean SLACK_HR for publishers (leaders) was 1.2483 (1.3925), while that for non-publishers (followers) was −1.5105 (1.1078), and the t-statistic for the mean difference was 30.43 (2.04). The mean values of SLACK_PERM and SLACK_FIN were 0.0031 and 0.0033, respectively. The difference of SLACK_PERM (SLACK_FIN) between leaders and followers was positive (negative) and statistically significant (0.0056 > −0.0315, t-stat. = 2.3474; 0.0614 < 0.1222, t-stat. = −3.3165), consistent with our prediction. However, we did not find significant differences in all three slack variables between firms publishing integrated reports and those not publishing, as shown in panel d. Moreover, companies publishing (initiating) CSR, sustainability, and integrated reports had larger firm size, higher profitability, greater foreign ownership, larger employment, and smaller ownership by the largest shareholders than others, on average (panels b and c).

Table 5. Correlations matrix.

	NON_FIN	INITIATE	1	2	3	4	5	6	7	8	9	10
1 SLACK_HR	−0.0736	−0.3346	1.0000
	0.0000 ***	0.0000 ***
2 SLACK_PERM	−0.0366	0.1533	−0.0692	1.0000
	0.0180 **	0.0198 **	0.0000 ***
3 SLACK_Fin	0.0999 *	−0.2141	−0.0986 *	0.0006	1.0000
	0.0000 ***	0.0011 ***	0.0000 ***	0.9714
4 SIZE	0.5370	0.3035	−0.3613	−0.0266	0.2413	1.0000
	0.0000 ***	0.0000 ***	0.0000 ***	0.0850 *	0.0000 ***
5 ROA	0.0416	0.0541	−0.0096	−0.0175	−0.3387	0.1160	1.0000
	0.0070 ***	0.4130	0.5346	0.2585	0.0000 ***	0.0000 ***
6 LARGEST	−0.0780	−0.0079	−0.1488	0.0279	−0.0432	0.0504	0.1629	1.0000
	0.0000 ***	0.9044	0.0000 ***	0.0709 *	0.0052 ***	0.0011 ***	0.0000 ***
7 FOR	0.4036	0.2460	−0.1973	0.0096	−0.0232	0.5611	0.1877	−0.0902	1.0000
	0.0000 ***	0.0002 ***	0.0000 ***	0.5340	0.1341	0.0000 ***	0.0000 ***	0.0000 ***
8 SD_SLACK1	−0.1229	−0.2782	−0.3798	0.0105	0.0209	0.0243	−0.0438	0.0558	0.0488	1.0000
	0.0000 ***	0.0000 ***	0.0000 ***	0.4953	0.1766	0.1161	0.0046 ***	0.0003 ***	0.0016 ***
9 SD_SLACK2	0.0236	−0.0475	−0.0802	0.0236	0.1424	0.1119	−0.0140	−0.0256	0.0965	0.1661	1.0000
	0.1263	0.4724	0.0000 ***	0.1267	0.0000 ***	0.0000 ***	0.3644	0.0975 *	0.0000 ***	0.0000 ***
10 SD_SLACK3	−0.1757	−0.0793	0.1049	−0.0145	0.1434	−0.3285	−0.2739	−0.1730	−0.1987	0.1145	0.0703	1.0000
	0.0000 ***	0.2302	0.0000 ***	0.3499	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***	0.0000 ***

Notes: Variables are defined in Equation (1). ***, **, * represent the statistical significance at 1%, 5%, and 10% levels, respectively.

presents the correlation of coefficients between test variables. The results show that human resource slack (SLACK_HR) was negatively correlated with non-financial disclosure (NON_FIN), whereas financial slack (SLACK_FIN) was positively related to such disclosure. Since these correlations are opposite each other, multivariate regression analysis would provide more accurate inference on the association between slacks and non-financial voluntary disclosure. Additionally, the positive correlations between NON_FIN and firm size (SIZE) and ROA imply that larger or more profitable firms are likely to provide CSR and sustainability disclosure. Moreover, foreign investor ownership (FOR) was positively correlated with non-financial disclosure. This may indicate the need of foreign investors for such information. Finally, unreported results show that the values of the variance inflation factor (VIF) were roughly between 3.26 and 3.69, suggesting that multicollinearity was not a severe issue in the regression analyses.

5.2. Multivariate Test Results

Table 6. H1: Organizational slack and CSR, sustainability, and integrated reporting.

	Dependent Variable = NON_FIN
SLACK_HR	0.5063	0.5284
	(2.369) **	(2.483) **
SLACK_PERM	−1.4513		−1.3832
	(−1.277)		(−1.458)
SLACK_FIN	0.5686			0.8239
	(0.365)			(0.531)
SIZE	1.1783	1.1664	0.9419	0.8812
	(4.592) ***	(5.608) ***	(4.753) ***	(4.215) ***
ROA	2.1817	1.5620	2.2118	2.4625
	(0.903)	(0.690)	(0.966)	(0.964)
LARGEST	1.6002	1.5556	2.2748	2.1961
	(1.170)	(1.111)	(1.662) *	(1.530)
FOR	1.9546	1.8623	2.2984	2.2434
	(1.291)	(1.239)	(1.654) *	(1.599)
EMP			0.8155	0.8338
			(3.153) ***	(3.245) ***
SD_SLACK1	−0.7657	−0.7201
	(−1.611)	(−1.592)
SD_SLACK2	−1.5448		−0.8189
	(−1.048)		(−0.502)
SD_SLACK3	−0.4427			−1.7620
	(−0.095)			(−0.391)
Constant	−37.5380	−37.7858	−37.5470	−36.3230
	(−5.433) ***	(−6.482) ***	(−7.492) ***	(−6.282) ***
Chi2	149.5194	92.5177	86.5252	118.1952
Pseudo R2	0.5968	0.5930	0.6086	0.6077
N	4186	4186	4186	4186

Notes: Variables are defined in Equation (1). All t-statistics are based on two-tailed tests and reported in the parentheses below the coefficient estimates. Standard errors are corrected for heteroscedasticity. ***, **, * represent the statistical significance at 1%, 5%, and 10% levels, respectively.

presents the logit regression results to examine the relationship between organizational slack and non-financial voluntary disclosure. When all three variables of organizational slack were included in the model as independent variables, only the coefficient for SLACK_HR was significant and positive (coef. = 0.5063, t-stat. = 2.369). We found the same result when we included one slack proxy in the model at a time. SLACK_HR was significantly and positively related to firms’ publication of CSR, sustainability, and integrated reports (coef. = 0.5284, t-stat. = 2.483). On the contrary, we found no association between SLACK_PERM (SLACK_FIN) and firms’ non-financial voluntary disclosure. These results imply that human resource slack is significantly and positively associated with the publication of CSR, sustainability, and integrated reports.

Table 7. H2: Organizational slack and first-mover of CSR, sustainability, and integrated reporting.

	Dependent Variable = INITIATE
SLACK_HR	−0.1127	−0.3298
	(−0.200)	(−0.608)
SLACK_PERM	4.0590		3.9756
	(2.176) **		(1.747) *
SLACK_FIN	−6.3198			−5.5028
	(−2.338) **			(−2.205) **
SIZE	1.2885	0.9699	1.0905	0.9611
	(2.626) ***	(2.021) **	(1.689) *	(1.520)
ROA	−11.8416	−4.1810	−4.2188	−11.5897
	(−2.885) ***	(−0.798)	(−0.886)	(−2.820) ***
LARGEST	9.3024	8.7302	8.9310	8.9132
	(2.166) **	(2.120) **	(2.153) **	(2.284) **
FOR	6.2598	4.2057	4.1666	5.5723
	(2.296) **	(1.740) *	(1.682) *	(2.191) **
EMP			0.0998	0.0241
			(0.165)	(0.038)
SD_SLACK1	0.6320	−1.4068
	(0.255)	(−0.635)
SD_SLACK2	−6.6748		−5.1951
	(−2.040) **		(−1.700) *
SD_SLACK3	5.7197			5.3940
	(0.552)			(0.468)
Constant	−40.0734	−32.6469	−35.6162	−33.3267
	(−2.595) ***	(−2.174) **	(−2.067) **	(−1.985) **
Chi2	67.6403	27.4031	33.1282	45.8958
Pseudo R2	0.3014	0.1950	0.2338	0.2416
N	231	231	231	231

Notes: Variables are defined in Equation (1). All t−statistics are based on two-tailed tests and reported in the parentheses below the coefficient estimates. Standard errors are corrected for heteroscedasticity. ***, **, * represent the statistical significance at 1%, 5%, and 10% levels, respectively.

shows the test results for Hypothesis 2, which examines the firms initiating such non-financial voluntary disclosure in their industry groups. We found that SLACK_HR was no longer significant in explaining the firms’ initiation of CSR, sustainability, and integrated reporting (coef. = −0.3298, t-stat. = −0.608 in column 2). However, as shown in column 3, the regression coefficient of SLACK_PERM was significantly positive, 3.9756 (t-stat. = 1.747), suggesting that firms with slacks in permanent employees are more likely to initiate publishing CSR, sustainability, and integrated reports. These results imply that human resource slack does not matter in deciding to be the first publisher of non-financial voluntary disclosure; instead, what matters is the excess employment of regular workers. Moreover, SLACK_FIN was negatively related to the initiation of CSR, sustainability, and integrated reporting (coef. = −5.5028, t-stat. = −2.205). This result indicates that initiating firms have greater extent of financial slacks than non-initiating firms.

We provide our final test results in

Table 8. H3: Permanent employee slack and integrated reporting.

Dependent Variable = IR
SLACK_Perm	36.0807			65.4323
	(1.759) *			(5.031) ***
SLACK_HR		−0.7038		−7.1916
		(−0.955)		(−5.771) ***
SLACK_Fin			−0.3305	−47.0466
			(−0.043)	(−5.254) ***
SIZE	−0.6999	−1.2534	−0.7717	0.4508
	(−0.424)	(−0.936)	(−0.463)	(0.132)
ROA	−57.4628	−25.8019	−23.0660	−89.1653
	(−2.505) **	(−2.207) **	(−2.110) **	(−8.425) ***
LARGEST	0.3464	0.1201	0.1626	0.6339
	(2.050) **	(2.229) **	(2.339) **	(1.519)
FOR	71.8053	56.5602	50.7237	135.6369
	(3.998) ***	(1.648) *	(1.859) *	(3.767) ***
EMP	−1.4858		−1.4426
	(−1.909) *		(−2.254) **
SD_SLACK1		0.6985		1.3449
		(0.369)		(0.492)
SD_SLACK2	−56.2077			−148.2205
	(−0.883)			(−2.652) ***
SD_SLACK3			22.2304	128.1959
			(0.718)	(2.431) **
Constant	8.4189	19.9197	10.2783	−72.3307
	(0.214)	(0.653)	(0.262)	(−1.111)
Pseudo R2	0.5746	0.4455	0.4700	0.6669
N	113	113	113	113

Notes:IR = 1 if an integrated report is published in the industry id in year t, otherwise 0; All independent variables are the industry average of the following variables. All t−statistics are based on two-tailed tests and reported in the parentheses below the coefficient estimates. Standard errors are corrected for heteroscedasticity. ***, **, * represent the statistical significance at 1%, 5%, and 10% levels, respectively.

. Hypothesis 3 examines whether permanent employee slack is associated with integrated reporting. As shown in column 1, the coefficient of SLACK_PERM was positive and statistically significant at the 10% level (36.0807, t-stat. = 1.759). In addition, when SLACK_PERM was replaced with SLACK_HR or SLACK_FIN, as reported in columns 2–3, the regression coefficients were not significant. In sum, industry-level analyses indicate that slacks in permanent employees are related to the firm’s compliance with a recent movement to integrate financial and non-financial disclosure in one report by publishing integrated reports.

5.3. Robustness Test Results

Our main tests were based on the measurement of human resources slack, the natural logarithm of the number of employees minus the industry average, which was used in the previous study [20]. For robustness, we scaled the number of employees by total assets. Unreported test results were not qualitatively different from the main results provided in the earlier section.

6. Discussion and Conclusions

This paper examines whether organizational slack is associated with voluntary disclosure of non-financial information. Using Korean data from 2005 to 2016, we found that excess human resources and financial slack were highly related to the publication of CSR, corporate sustainability, and integrated reports. Firms with slacks in permanent employees were shown to be the first-movers in producing such non-financial disclosures. This result suggests that firm knowledge and job experience contained by regular employees are useful to collect/produce relevant information about long-term firm value and performance.

Our findings in the present study are subject to several limitations. First, we performed the industry-level analysis on integrated reporting because very few companies have been producing integrated reports in South Korea. During our sample period, we counted only 36 firms having published integrated reports. Hence, these results need to be updated by future studies using a larger sample of firm-year observations. Next, we acknowledge that it might be more efficient to reduce the number of human resource and introduce technology-based solutions to develop additional information. In other words, the use of new technologies could reduce the need of human resource. If this is the case, the positive association between human resource slacks and non-financial disclosure may not be observed. However, for empirical analysis, we need to collect data on whether new technologies are adopted to develop additional information. Thus, this issue will be addressed in future research. Finally, our findings may not be generalized to firms outside of South Korea because our empirical analyses were limited to publicly traded Korean companies. However, prior work such as Vanacker et al. [48] has argued that a country’s formal institutions could affect the way managers deploy their slack resources. Hence, the findings based on the Korean research setting could provide new insights on this issue.

Additionally, the findings in this study add new evidence to the literature suggesting that organizational slacks are significantly related to firms’ non-financial voluntary disclosure. Many prior studies have focused on capital market-related incentives as the explanatory variable of non-financial voluntary disclosure. This paper differentiates from those papers by focusing on organizational resources that can be considered prerequisite to prepare and produce additional information voluntarily.

Our test results on the second and third hypotheses shed light on the positive side of permanent employment. We found evidence indicating that the knowledge and experience of permanent employees are related to a greater extent of voluntary disclosure and consequently contribute to reduced information asymmetry between managers and stakeholders. Such findings add new evidence to the existing research that argues holding excess numbers of employees may benefit the firm. For instance, Lecuona and Reitzig [49] found that the value of excess human resources increases as firms face competitive pressure and that excess numbers of employees who possess important tacit knowledge that is specific to firms could benefit the company. This study provides practical and managerial implications for how management can utilize human resources—especially permanent employee slack—to improve the information environment. It is essential for firms to provide useful non-financial disclosures as well as concise integrated reports so as to reduce information asymmetry. Such disclosure quality can be assured when the information is prepared and developed by employees with firm-specific experience and knowledge. Hence, firms having slack in permanent employees could facilitate disseminating their information through non-financial disclosures and/or integrated reporting. Finally, our findings may be of interest to policy makers and regulators since permanent employees could play a role in enhancing the information environment by promoting integrated reporting that delivers both financial and non-financial information.