**3. Methodology**

Concepts are an answer to the "what is" question and they are about meaning, semantic and ontology [5]. Developing valid concepts for social science involves analyzing descriptive, normative, and causal aspects, concurrently. To answer the "what is" question, one must identify the necessary and jointly sufficient conditions which constitute the defining features of the concept [5]. To look for the constitutive elements of the concept of corporate sustainability, we use the Necessary Condition Analysis (NCA) methodology brought forward by Jan Dul [9], according to which, a condition is necessary when its absence results in the absence of the phenomenon. Therefore, the condition is necessary for the presence of the phenomenon. Such hypotheses are rarely formulated and tested in organizational sciences, and are different from conventional analysis where the complex interrelation of all factors attempts to explain the presence of the outcome in a probabilistic relationship. NCA ignores the causality that predicts the presence of the outcome with a large number of factors and only makes simple theoretical statements to predict the guaranteed absence of the outcome when the condition is absent [9]. The method is intuitive and straightforward, it triggers a new way of theoretical thinking that is based on necessity logic and it works in isolation from the rest of the causal structure, that is why it is necessary [9]. When searching the constitutive features, we are not concerned with the level of presence of the condition, but with whether the condition is present or not. Therefore, in the dichotomous interpretation of Dul's [9] framework, the condition [X] and the outcome [Y] can assume only two values: absent or present. The contingency matrix, represented below, is a common way to present dichotomous necessary conditions [9]. The dashed lines are the 'ceiling lines' that separate the area with observations from the area without observations. For a condition to be necessary, the top left square requires to be empty, meaning there are no observations where the condition [X] is absent and the outcome [Y] is present, therefore the condition is necessary. Necessity does not equal sufficiency: the condition can be present, yet the outcome can be absent—bottom right square. The set of all necessary and jointly sufficient conditions constitute the concept [5]. See Figure 3.

**Figure 3.** Contingency Matrix.

To build the hypothesis over the potential necessary conditions, we draw the data from the extensive systematic literature review performed by Meuer et al. [2] on corporate sustainability between 1983 and April 2018, which provides 101 articles. We extend the review until December 2021, with additional 31 articles for a total of 132 articles. See Figure 4. We adopted Meuer et al.'s [2] literature review, as they engaged in a similar mission of identifying the essential attributes of corporate sustainability with the aim

to provide conceptual clarity. The purpose and the selection criteria of their articles are therefore pertinent and adherent to ours.

**Figure 4.** Data search strategy and qualification criteria.

To identify the necessary conditions of the concept of Corporate Sustainability, we focused on the 'What' question [9], and prescind from the 'How' and 'Why' questions. In addition, as definitions, which in the context of CS can also be contested and conflicting, transcend the required analysis over meaning and context [5,73,80–82] to build complete concepts we analyze all 132 papers, even if they do not provide a definition. To this point, for example, one of the most cited definitions of corporate sustainability is from Dyllick and Hockerts [53]: "Corporate sustainability can be defined as meeting the needs of a firm's direct and indirect stakeholders such as shareholders, employee, clients, pressure groups, communities, etc. without compromising its ability to meet the needs of future stakeholders as well" (p. 131). When reading the full article, the authors performed a considerable amount of analysis on the natural and social capital elements of corporate sustainability, which are not mentioned in the definition. Other examples include Sterman [83]: 'Sustainability initiatives that are framed as also helping to heal the world' (p. 3), which does not provide any indication of what it takes to heal the world

Dul [9] suggests two ways for testing or inducing necessary conditions with the data sets of observation. Either only successful cases—where the output is present—are purposely sampled and the omni-presence of the condition is an indication of necessity, or only cases with the absence of the condition are selected and the absence of the outcome is an indication of the necessity. Given the complexity and magnitude of the concept, we adopted the first approach, and built the following framework for guiding the allocation of papers into the contingency matrix—See Figure 5.

Principle 1: Based on lessons drawn from the ontological analysis, we classified a paper as 'present output' when CSR, business sustainability, and other similar terminologies were used as to relate to the concept of the sustainability of a corporate, and treated it as 'absent output' when they meant to talk about some specific without a link to the concept of corporate sustainability. For example, Dahlsrud [84], Kleine [85], and Cheng [86], use CSR as the broader terminology for corporate sustainability, and they are treated as output present. Articles such as Delmas [87], Herva [88], and Ehrenfeld [89] that meant to talk specifically about the environmental aspects of a corporate with no reference to corporate sustainability are treated as output absent. Articles that are not specifically aimed

at clarifying the concept of corporate sustainability are included as 'present output' when they clearly embrace a definition or a specific meaning of CS for their analysis.

Principle 2: Articles that have corporate sustainability in their discussion, but do not espouse any of the existing definition or meaning of CS are treated as 'outcome absent' and 'condition absent'. This last point is necessary to allow for the reconciliation of the full number of the analyzed papers. We assume that something does not exist if its meaning is not defined or spelled out. For example, to explain corporate sustainability, Ahi [27,90], Urdan [13], and Swarnapali [16] mention a few definitions, but do not clarify which one of those will be used for their analysis, and they are therefore treated as output absent. Conversely, papers of Boiral [90] that speak of auditing practices in sustainability, or Figge [91] that talk about sustainability value added measurement, embrace a specific meaning of CS, and are treated as output present.

Principle 3: Articles that do not have CS as a core subject but are rather useful for its understanding are treated as output absent. See Figure 4 for qualification criteria.

1. Articles that have CS as a core subject and embrace a meaning or definition

2. Articles that have CS as a core subject but Do Not embrace a meaning or definition

3. Articles that do not have CS as a core subject

**Figure 5.** Framework for allocating the articles in the contingency matrix.

When looking for the constitutive features, words with similar meaning are coded within their macro, most used, terminology. For example, ecology and nature are coded within environment, profit and organizational objectives are coded within economic, and organizational culture is coded within governance.

We performed three rounds of reviews and coding, through repeated critical reading, to confirm the findings and to ensure consistency with the guidelines. The third round was performed after confirming the validity of the scoping and allocation with Professor Jan Dul.

Based on the ontological analysis of the concepts of CS as well as on the frequency of the constitutive features cited in the literature, we formulate the hypothesis as follows: Constitutive features of CS may have Environmental, Social, Economic, Governance and Time dimensions.

The environmental dimension refers to the preservation of natural resources: when the production system utilizes more energy and materials that can be reproduced and when more emissions are emitted that can be absorbed, the industrial system becomes ecologically unsustainable [92]. The social dimension refers to the corporate responsibility to achieve a balance between a firm's economic operations and the society's aspiration and requirements for community welfare: social responsibility occurs when business firms, through the decisions and policies of its executive leaders, consciously and deliberately act to enhance the social well-being of those whose lives are affected by the firm's economic operations [54].

The Economic dimension refers to the viability of a company from profitability standpoint: "Economically sustainable companies guarantee at any time cashflow sufficient to ensure liquidity while producing a persistent above average return to their shareholders" [53]. The Governance dimension refers to the arrangements a company needs to establish in order to guarantee the integrity of the organization as well as the integrity of internal management processes [93]. The Time dimension refers to the ability of firms to respond to short-term financial needs without compromising theirs, or others, ability to meet their future needs [94].
