*Article* **A Systemic Approach for Sustainability Implementation Planning at the Local Level by SDG Target Prioritization: The Case of Quebec City**

**David Tremblay 1,\* , Sabine Gowsy <sup>2</sup> , Olivier Riffon <sup>1</sup> , Jean-François Boucher <sup>1</sup> , Samuel Dubé 2 and Claude Villeneuve <sup>1</sup>**



**Citation:** Tremblay, D.; Gowsy, S.; Riffon, O.; Boucher, J.-F.; Dubé, S.; Villeneuve, C. A Systemic Approach for Sustainability Implementation Planning at the Local Level by SDG Target Prioritization: The Case of Quebec City. *Sustainability* **2021**, *13*, 2520. https://doi.org/10.3390/su 13052520

Academic Editors: Margarita Martinez-Nuñez and Mª Pilar Latorre-Martínez

Received: 23 January 2021 Accepted: 22 February 2021 Published: 26 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

**Abstract:** The success of the 2030 Agenda hinges on mobilization at the local level. The localization of sustainable development goals (SDGs) and their targets involves adapting them to local contexts. This case study of Quebec City, Canada, illustrates how the use of a systemic sustainability analysis tool can help integrate SDGs in the building of a sustainable development strategy at the local level. Our approach focuses on the use of an SDG target prioritization grid (SDGT-PG) and begins with the mobilization and training of a group of officers representing various city services. We first used an original text-mining framework to evaluate SDG integration within existing strategic documents published by the city. The result provides a portrait of existing contributions to SDG targets and identifies potential synergies and trade-offs between services and existing policies. A citywide prioritization workshop was held to assess the relative importance of SDG targets for the city. Priorities were then identified by combining the importance of the targets as viewed by stakeholders, the current level of achievement of SDG targets as determined by the analysis of existing documents, and the jurisdiction and responsibilities given to Quebec City in regard to federal and provincial legislation. We identified the main focus areas and related SDG targets. Furthermore, we observed whether actions needed to be consolidated or new actions needed to be implemented. The identification of synergies and trade-offs within the city service actions provides information on the links to be made between the different municipal services and calls for partnerships with other organizations. The use of the SDGT-PG allows the vertical and horizontal integration of the SDG targets and demonstrates how participation and inclusion facilitate stakeholders' appropriation of the applied sustainable development strategy.

**Keywords:** 2030 Agenda; sustainable development goals (SDGs); systemic sustainability analysis; SDG targets prioritization

#### **1. Introduction**

In 2015, members of the United Nations unanimously adopted the 2030 Agenda for Sustainable Development [1]. The 17 Sustainable Development Goals (SDG) and 169 targets represent a global framework to guide the implementation of sustainable development (SD) by 2030 [2,3]. While the SDGs and targets were first designed for the global level [4], the 2030 Agenda is a universal program that applies to all governments and actors, regardless of their level of intervention [1,5]. Because cities represent the level of government closest to the population [6], they have the capacity to intervene quickly and concretely, according to the powers assigned to them, [7–9], and they are considered essential actors for sustainability [4]. Furthermore, urbanization is accelerating globally, and 68 percent of the world's population is expected to live in an urban area by 2050 [10]. Moreover, all human activities at the city level affect the economy, the environment, people, culture, governance, etc. [4,8,9,11–13].

Implementing the 2030 Agenda and achieving the SDGs require an integrated and systemic approach [1,14,15]. The principle of integration applies (i) horizontally between different policy areas, (ii) vertically from global to national to local levels, and (iii) territorially between local governments [6]. SDG implementation at the local level is termed "localization". In the context of implementing the 2030 Agenda, SDG localization refers to "the process of defining, implementing, and monitoring strategies at the local level for achieving global, national, and subnational SDGs and targets" [16]. Although there are numerous approaches and tools at the national level [11], the localization of SDGs requires documented [3] approaches adapted to applying sustainable development at the local level [17]. Scientific research on cities and the SDGs has increased [7,18,19]; nonetheless, there remains a knowledge gap in regard to how best to implement SDGs at the local level [3,11].

Approaches and tools dedicated to SDG localization involve some critical elements. Although the 17 SDGs and their 169 targets are set as "universal and indivisible," they must be applied in line with the realities, capacities, levels of development, and priorities specific to national or local contexts [1,20]. Cities face various challenges when implementing SDGs. These challenges include contextualizing their approaches to the specific environmental, economic, social, political, and cultural conditions [7,21,22]. This contextualization involves adapting SDG content and their targets to make them locally relevant [3,6] while maintaining integrated and systemic thinking to keep a holistic perspective of the system as a whole [1,14,15,23]. The systemic approach implies horizontal, vertical, and territorial integration [24]. Horizontally, integration aims to maximize synergies and diminish trade-offs [4,24–27]. Vertically, at the local level, integration involves the principle of subsidiarity [7,11,28] and a clear understanding of the division of powers between the various levels of government [11,14,28]. In addition to implementing SDGs, invoking these elements will ensure policy coherence and integrated multilevel governance [6,7,28].

The issue of prioritizing SDG targets is critical for local authorities, as the needs are multiple and urgent; yet, there are often limited capacities and resources. Prioritization is a complex exercise that combines assessing the importance of a target and determining its level of achievement at a specific time, for a given territory [2]. The integrated approach of the SDGs induces a paradigm shift in the elaboration of development plans and strategies at all levels. Today, it is a matter of contributing to the achievement of a global and shared vision. Thus, SDGs and their targets may constitute a normative framework. Given the inability and, most likely, the irrelevancy for cities to implement all the targets, it is essential to use approaches and tools to determine priorities [4,29].

A growing number of studies are looking at a systemic approach for prioritizing and implementing the SDGs [12,27,30,31]; however, very few tools for SDG and target prioritization are documented in the scientific literature. Among papers presenting tools, three include a set of criteria for establishing priorities. First, Allen et al. [30] proposed a multicriteria analysis for prioritizing SDG targets through the use of three main criteria:


The Stakeholder Forum [20] submitted a method of analysis addressed to developed countries to assist them in identifying the goals and targets representing the biggest transformational challenges. They proposed three criteria:


3. Transformationalism: determining whether the achievement of the goal/target requires new and additional policies beyond those currently in place.

Finally, the Sustainable Development Solution Network (SDSN) [6] proposed broad guidelines to define local SDG targets:


This paper presents the case study of Quebec City (Quebec, Canada) using an original and adapted systemic sustainability analysis tool. Quebec City is located in the province of Quebec, Canada. Canada is a federal state where responsibilities are shared between federal and provincial jurisdictions. Local governments, such as cities, are a provincial responsibility; however, as issues related to sustainable development touch multiple jurisdictions, our analysis includes the federal, provincial, and local (city) levels. This approach aims to bolster the implementation of the SDGs at the local level and integrates the key elements of contextualization, adaptation, systemic thinking, subsidiarity, and policy coherence.

The method focuses mainly on the use of the SDG target prioritization grid (SDGT-PG), a participatory prioritization tool for SDG targets applicable at local, national, and regional scales. The SDGT-PG methodology is inspired by the sustainable development analytical grid (SDAG) used at local and national levels since 1988 [32]. The SDAG methodology, which emphasizes participatory processes and scientific robustness, was developed in a partnership between academics (Université du Québec à Chicoutimi, Canada), an international organization (Organisation internationale de la Francophonie), and an international consulting firm (GlobalShift Institute Ltd., Quebec City, QC, Canada). This approach was tested in both developed and developing countries at national and local levels. Burkina Faso, Benin, Niger, and Togo refer to the use of the SDGT-PG as a prioritization tool in their Voluntary National Reviews presented at the United Nations High-level Political Forum on Sustainable Development. In this study, we test our central hypothesis that the use of the SDGT-PG allows the vertical and horizontal integration of the SDG targets. We also demonstrate how participation and inclusion permit stakeholder appropriation.

#### **2. Materials and Methods**

*–*

The applied approach is an iterative process (Figure 1) inspired by two analytical tools: the sustainable development analytical grid (SDAG) [32] and the rapid integrated assessment (RIA) [33]. Our approach also makes use of best-known practices and guidelines [6,20,30]. The different advancements of this action–research process, in coconstruction with Quebec City officials, aimed to generate data and information related to the three main SDGT-PG criteria being evaluated:

**Figure 1.** Stages of the applied approach.

1. Performance: Relying on SDG target indicators, what is the current level of achievement of the targets?

anager's office coordinated the project. The project team, established by the

ized 27 leaders from 19 Quebec City's administrative units.


#### *2.1. Mobilization and Capacity Building*

To eliminate silos and apply a systemic approach, we took the initial step of forming a group of leaders. We mobilized 27 leaders from 19 Quebec City's administrative units. To be selected, a leader needed to represent one of the main administrative units and embody the concepts of sustainability through their values, interests, personal, and/or professional activities. The selected leaders were readily available and also had the authorization of their respective managers. The 27 leaders included 17 advisers, 6 managers, 2 engineers, 1 analyst, and 1 police officer. The leaders comprised 15 men and 12 women. The city manager's office coordinated the project. The project team, established by the office, worked in partnership with our research team to organize and structure the complete process.

A series of workshops was co-managed by the partners. The main objective of these workshops was to raise awareness about the SDGs [6]. The specific objectives were to:


#### *2.2. Diagnosis–Performance*

The internalization of the SDGs requires identifying those already-existing actions that can be linked to the SDGs [6,14]. To carry out this diagnosis, we analyzed 89 strategic documents produced by Quebec City (A list of the analyzed documents is available in the Supplementary Materials). We aimed to:


We identified the initiatives using WordStat, a content-analysis and text-mining software within ProSuite (Provalis Research, Montreal, QC, Canada), a collection of integrated text-analysis tools [34]. For our analyses, we developed a specific dictionary linked to the content of the 2030 Agenda. Our dictionary included 1602 expressions found in the labels of the SDGs, SDG targets, and SDG indicators (The full dictionary is available in the Supplementary Materials).

First, we prepared each document for analysis by removing figures, hyphens, brackets, and braces. We imported the strategic documents into QDA Miner, a qualitative data analysis component of ProSuite. We conducted content analysis using WordStat for each document separately. Each SDG target represented a category in the dictionary. To transform textual data into keywords or content categories, we used a lemmatization substitution process. Lemmatization is a "process by which various forms of words are reduced to a more limited number of canonical forms like conversion of plurals to singulars and past tense verbs to present tense verbs" [35].

For each occurrence identified by the software, an expert in charge of the processing validated the result to retain only relevant occurrences. These retained occurrences were then classified within a matrix where they were associated with corresponding targets. The matrix is based on the rapid integrated assessment developed by the United Nations Development Programme [33] (Table 1).

**Table 1.** Excerpts of the matrix of links between the SDG targets (6.1 to 6.6) and the analyzed Quebec City strategic documents. The X mark indicates an occurrence between a strategic document and an SDG target (The full matrix is available in the Supplementary Materials). The title of each strategic document has been translated by the authors from the original French title.


We processed the data from the matrix to obtain a portrait of the SDG target coverage by the existing strategic documents and to group those documents influencing the same targets. Through identifying occurrences between the strategic documents and the SDG targets, we could assess, in light of the identified actions, the degree to which SDG targets had been achieved (performance). We assessed performance on a four-level scale:


We automatically assigned a performance score of 1 to targets having no occurrences. We awarded a performance score of 2 or 3 when occurrences existed between the city's strategic documents and a target. Our assessment of performance varied in accordance with the number of strategic documents associated with a target and the quality of actions mentioned in those documents. We never assigned a maximum score of 4, as we found no indicator, with verified metrics, for which we could attribute this performance.

#### *2.3. Identification of Synergies and Trade-Offs*

To apply systems thinking, we organized a workshop with the aim of identifying the potential interactions between the applied activities within the various city services. The research team identified themes for the 107 operational targets on the basis of target content and their indicators. SDG targets fall into two categories: "operational" and "means of implementation" (MoI). Operational targets relate to those to be achieved, whereas MoI refers to conditions that help attain targets [36]. MoI includes the mobilization of financial resources, technology development and transfer, capacity building, inclusive and equitable trade, regional integration, and the creation of a national environment conducive to the implementation of the sustainable development agenda [1]. MoI targets apply to national competences, which deviate from those at the local level. We therefore discarded MoI targets and SDG 17 [31,37].

As an initial step, the 27 leaders identified areas of activity undertaken both inside and outside of their administrative units and associated these areas with the themes of the SDG targets. Then, they identified potential interactions with other SDG targets. All interactions are directional; hence, for each interaction, there is a source target and an impacted target. In the case of bidirectional interactions, we used two-directional interactions, reversing sources and impact. City leaders characterized interactions as synergies or trade-offs. A

member of the research team validated each of the interactions and completed the exercise by adding interactions and adjusting some interactions associated erroneously with targets. Once the validation was completed, we analyzed the synergies and trade-offs by SDG and target. We used a cross-impact matrix [27,30,31,38] where the weight given to an interaction corresponds to the number of activities associated with the interaction.

#### *2.4. Importance and Prioritization*

Localizing the SDGs requires adapting the global reference framework to ensure that it is relevant to the local context [3,6]. This contextualization of the 2030 Agenda will promote ownership and mobilization of stakeholders [14,39]. To increase the understanding of the SDGs and their targets for the stakeholders in our study, we adapted the wording of the targets without changing the original meaning. The target labels were adjusted to change references from a national scale to a local scale. For example, Target 1.2: "By 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to **national** definitions" becomes "By 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to the **city** definitions."

Targets adapted to the local context were prioritized during a workshop that brought together 182 city employees. The employees occupied positions at various levels throughout the city services. We sampled city employees to ensure the representativeness of employment sectors, age group, gender, and workplace location. The selected employees did not need any particular skills or knowledge to participate in the workshop. Prioritization is an essential stage in identifying the relevant actions to be implemented at the local level [4].

Twenty-four tables of seven to eight employees, separated across two three-hour sessions, discussed the level of importance of SDG targets in the context of Quebec City. The table composition was predetermined to maximize diversity. Each table weighted the targets of three to four SDGs, for a total of 21 to 22 targets per table. An animator facilitated the discussion, while a second person recorded notes on a prepared canvas. Each participant was provided access to a set of four cards, representing the four levels of importance, to help them judge the importance of a target:

N/A: Not applicable;


3: Essential: Priority in the short term.

For each target, (1) the animator announced and explained the appropriate target; (2) the employees expressed their views on the level of importance to be given to the target; (3) the animator initiated a dialogue in regard to the employees' justifications for this importance; and (4) the employees then expressed their final scoring for the importance of the target after these discussions.

We recorded both employee assessments of importance (before and after discussions), and we noted the employees' justifications. To define the final level of importance of the targets to be entered in the SDGT-PG, we averaged (rounded to the nearest unit) the importance score of the final results.

Using the SDGT-PG, we produced a priority index for each target. The more participants that deemed a target to be significant and the poorer the target's performance, the greater the priority given to the target in question. The priority level corresponds to the table shown in Figure 2.

3) the animator initiated a dialogue in regard to the employees' justifications for this im-

sions), and we noted the employees' justifications. To define the final level of importance

deemed a target to be significant and the poorer the target's performance, the

**Figure 2.** Prioritization index grid in which an urgent target requires immediate intervention: a priority target should be addressed within a three-year horizon, a medium-term target should be addressed within seven years, a long-term target should be addressed within a 10-to-15-year period, and a target to be consolidated requires interventions that make it possible to maintain the current level of performance. The other priority levels do not require specific actions.

#### *2.5. Governance*

This stage aims to determine, for each target, the level of governance, from local to national, legally responsible for implementing actions required to achieve the target. In our case, the national level includes both the provincial and federal governments. The project team evaluated the target with reference to legislation at the Quebec (provincial) and Canada (federal) levels. To identify the governance level, we classified governance on a scale from 1 to 4:


#### *2.6. Localization*

The final information produced in the SDGT-PG considered the role of the different levels of governance in implementing initiatives; this governance level can affect whether a target can be achieved. Combining the priority level (Figure 2) and the governance assessment, we could determine what should be considered by local and national planners and what targets can be achieved jointly, in some form of multilevel governance (Table 2).


**Table 2.** Initiatives to be undertaken according to the level of priority and our governance assessment. These are proposals aimed at local (Quebec City) and national (Quebec, Canada) levels.

#### **3. Results**

#### *3.1. Performance Assessment*

Our diagnosis aimed to assess the potential achievement of the SDG targets by identifying those already considered within the city strategic documents and by assessing the degree to which they had been achieved. We found that the city documents dealt with 85 targets, representing 50% of the total SDG targets (Table 3). We noted a difference in coverage between the operational targets (71%) and the MoI (15%). No document covered the targets of SDG 14; however, five SDGs had 100% of their operational targets considered by at least one strategic document.


**Table 3.** SDG and targets covered by the analyzed Quebec City strategic documents.

Note: SDG 1: No poverty; SDG 2: Zero hunger; SDG 3: Good health and well-being; SDG 4: Quality education; SDG 5: Gender equality; SDG 6: Clean water and sanitation; SDG 7: Affordable and clean energy; SDG 8: Decent work and economic growth; SDG 9: Industry, innovation and infrastructure; SDG 10: Reduced inequalities; SDG 11: Sustainable cities and communities; SDG 12: Responsible consumption and production; SDG 13: Climate action; SDG 14: Life below water; SDG 15: Life on land; SDG 16: Peace, justice, and strong institutions; SDG 17: Partnerships for the goals.

> Table 3 illustrates that 84 targets (50%) are not covered, including 53 MoI targets (of a possible 62). SDG targets are also unequally covered by the strategic documents (Figure 3). Dividing the analyzed documents into blocks of five (for clarity's sake in Figure 3), we noted that 46 targets were represented only within 1 to 5 documents, 15 targets were found in 6 to 10 documents, 12 targets were covered by 11 to 15 documents, only 4 targets were found in 16 to 20 documents, and 8 targets were linked to 20 or more documents. Thus, most documents covered only a limited number of SDG targets.

**Figure 3.** Number of targets covered by the strategic documents.

Of the eight most covered targets, three come from SDG 11 (Sustainable cities and communities), two from SDG 9 (Industry, innovation, and infrastructure) and SDG 16 (Peace, justice, and strong institutions), and one from SDG 3 (Good health and well-being).

showed more than 80% of the targets had a performance ranking of 3 "in the process of

"achieved" becaus

being achieved" (

**–**

**–**

**–**

**–**

We used the potential coverage of SDG targets to assess performances in the SDGT-PG. Except for SDG 2 and 14, at least 60% of the targets showed some performance in terms of potentially achieving the target (Figure 4). No target could be labeled as "achieved" because we found no indicators confirming this level of performance. Targets of SDG 6 (Clean water and sanitation) and 11 (Sustainable cities and communities) showed more than 80% of the targets had a performance ranking of 3 "in the process of being achieved" (Figure 4). "achieved" becaus showed more than 80% of the targets had a performance ranking of 3 "in the process of being achieved" (

#### *3.2. Synergies and Trade-Offs*

The city officers and our research team identified 687 potential interactions, including 638 synergies and 49 trade-offs. These interactions involve 86 targets and 16 SDGs. Table 4 shows the targets that most influenced other targets on the basis of the number of times they are the source of interaction. The table also reveals the targets most influenced by other targets according to the number of times they are impacted by other targets. All these targets exhibited positive and negative interactions. Among all the analyzed interactions, the most influencing and most influenced targets were all strongly positive (Table 4).

**Table 4.** Most influencing and most influenced targets. Numbers in parentheses identify the number of positive/negative interactions, respectively, in which the targets are involved. (The full crossimpact matrix is available in the Supplementary Materials).


Influencing targets came from various SDGs. The exception was SDG 9, for which two targets were found in the five most influencing targets (Table 4). In the context of applying the SDGs at the municipal level, we expected and noted Target 11.3 (Sustainable urbanization; participatory and integrated planning and management) to be one of the most (the third) influential targets (Table 4).

The most impacted targets came from six SDGs. The most impacted target is 3.4 (Non-communicable diseases, mental health, and well-being). We found a single target (10.2. Empowerment; social, economic, and political inclusion) in both the most influencing and the most influenced targets (Table 4).

The targets included in Table 4 have the highest positive results. In terms of negative impacts, targets 7.3, 9.4, 9.5, and 15.1 most negatively affected other targets (sum: −4). The most often negatively impacted targets were targets 8.1 (sum: −7) and 10.2 (sum: −5). The highest positive interaction was between targets 11.3 and 10.2 (sum: 5). We observed the highest negative results (sum: −2) for interactions between targets 9.4 and 3.4, between 9.5 and 10.3, and between 15.1 and 11.1.

At the SDG level, targets from SDG 9 (sum: 100) and SDG 11 (sum: 99) had the highest positive net influence on other targets (Figure 5). Targets SDG 3 (sum: 76), SDG 11 (sum: 69), SDG 8 (sum: 53), SDG 10 (sum: 52), and SDG 12 (sum: 51) were most often impacted by other targets, according to their net influence. We noted that 94.7% of net influence was positive when excluding absent relationships. The greatest net-positive influence was from SDG 11 toward SDG 10 (sum: 23). The net influence of SDG 15 toward SDG 8 was the most negative (sum: −3) (Figure 5).


**Figure 5.** Cross-impact matrix of 16 SDGs. Numbers indicate the net influence of positive and negative interactions between targets of the corresponding SDGs.

#### *3.3. Importance, Prioritization, and Governance*

The participants at the prioritization workshop assessed the level of importance of the 107 operational targets (Figure 6). They found all targets to be relevant. The participants considered most targets as important (56.1%), with 32 targets deemed essential (29.9%) and 15 as unimportant (14%). The SDGs having the highest percentage of essential targets were

SDG 6 (66.7%) and SDGs 5, 12, and 16 (50%) (Figure 5). SDGs 3 and 14 had the highest percentage of unimportant targets (44.4% and 57.1%, respectively).

−1 −1 −1 −1

−1

−3

−1

**Figure 6.** Distribution of the level of importance given to the SDG targets by the Quebec City round tables.

We obtained a prioritization index by crossing performance with importance. Eight targets, among eight different SDGs, were prioritized as urgent (Figure 7), whereas 27 targets were deemed a priority. We noted five priority targets in SDG 15, four in SDG 12, three each in SDGs 4, 14, and 16, two each in SDGs 2, 5, and 8, and one each in SDG 1, 10, and 13. SDGs 3, 7, 9, and 11 did not have any urgent or priority targets (Figure 6). Thirty-four targets were prioritized in the medium term and fifteen in the long term. Additionally, Quebec City needed to consolidate 23 targets. SDG 11, with four targets, and SDG 8 and 9, each with three targets, showed the most targets to be consolidated.

Prioritization levels for each SDG. (A detailed table including all targets is available under the "detailed results" **Figure 7.** Prioritization levels for each SDG. (A detailed table including all targets is available under the "detailed results" tab of the SDGT-PG available in the Supplementary Materials).

The governance assessment showed that the project team members considered Quebec City to have exclusive power over six targets (5.6%) (Figure 8). These targets are found in SDGs 6, 11, and 12 (each having two targets). On the other hand, they assessed 29 targets (27.1%) as being exclusively national (provincial or federal) jurisdiction and responsibility. Among the SDG targets most associated with the national level, we noted five of the seven targets of SDG 14 (71%), two of the three targets of SDG 7 (67%), and four of the seven targets of SDG 4 and 10 (57%). Twenty-six targets (24.3%) represented a shared responsibility, and 46 targets (43%) were primarily national competence, although supported at the local level. Overall, the national level was better positioned to intervene on 75 targets (70%); for instance, the national level holds most of the authority to intervene in regard to all targets of SDGs 4 and 7 (Figure 7).

**Figure 8.** Distribution of the SDG targets among the levels of governance.

#### **4. Discussion**

The success of implementing the 2030 Agenda requires the mobilization of all actors at all levels. Our SDG localization approach focuses on the local level and includes an original systemic tool to identify priorities in a context of strategic planning. We used parameters found in the literature [6,20,30]; they were evaluated separately but integrated to define the priorities.

sions linked to the SDG targets identified the targets considered (or not) within the city's In our study case, assessing the current sustainability context for Quebec City is necessary to clarify the starting point and to develop a sustainable development strategy based on achievements [14]. The development and application of our dictionary of expressions linked to the SDG targets identified the targets considered (or not) within the city's strategic documents.

A proper analysis of performance requires contextualizing performance in terms of governance level. Local governments, depending on the effective distribution of powers in a given country, have varying levers on the SDGs. Quebec City is located in the province of Quebec and also falls within the Canadian national governance. The responsibility for municipalities resides with the provinces under the Canadian constitution. In the province of Quebec, cities have the legislative powers of development and urban planning, housing, roads, community and cultural development, recreation, urban public transport, and wastewater treatment [40]. We strongly recommend that an expert assessment of governance parameters, in accordance with the national/provincial legislative texts, be undertaken when applying an SDGT-PG.

Examining the distribution of powers among government levels allows an analysis of performance crossed with an evaluation of governance (Table 5). In this study, we observed that no exclusively local responsibility target was achieved. Among the exclusively national targets (at the provincial and/or federal level), however, 69% of the targets had not been achieved. In Canada, navigation, coasts, and inland fisheries are a federal responsibility. Five of the seven SDG 14 targets related to oceans and marine resources are exclusively a national responsibility and have not yet been achieved. The two other targets are considered as a shared responsibility. In contrast, Quebec City was on track to achieve 83% of the targets under its responsibility. These are targets of SDG 6 (Clean water and sanitation) and 11 (Sustainable cities and communities), which correspond to the fields of competence given to municipalities in provincial legislation. The other targets of exclusive local responsibility were partially achieved.


Note: For the pairs of percentage values, the percentage in bold represents the relative distribution of target performance related to the governance level, whereas the percentage presented in normal font represents the relative distribution of the targets' level of governance related to the performance.

> Agenda 2030 states that the SDGs and their targets are global and that [national] governments should define their priorities according to their particular contexts [1]. This contextualization applies to all levels of governance, from local to national. The successful implementation of SDGs requires multilevel governance implemented with communication channels that promote vertical integration [7,41]. Although cities have extremely varied contexts, they encounter common obstacles, such as issues of power [24], and can seize specific opportunities addressed by our approach. Moreover, localization allows local authorities to participate more effectively to achieve national SDGs.

#### *4.1. Obstacles, Limitations, and Challenges of SDG Localization*

The scope of the 2030 Agenda limits its localization. The formulation of targets is addressed at the national and global levels, and their text-based interpretation can have a demobilizing effect on the local-level actors. Local actors may see this agenda as being focused on global issues and, thus, they may ultimately reject the agenda outright [42]. Implementing SDGs at the local level requires localizing the targets by adjusting the labels without distorting their meaning. In our case study, the Quebec City project team modified the wording of targets, for which the scope was explicitly national, to provide a local-scale feel to the target. This adaptation increases the tangibility of targets for local actors, who must assess the importance of the targets and ensure that the targets are implemented at the appropriate—local—level. One could assume that targets explicitly mentioning a national scope would be assessed as less important or not applicable for local actors. For the MoI targets, however, adapting these targets to local contexts is difficult, as these targets often involve international partnerships for implementing the 2030 Agenda. From the governance parameter of the SDGT-PG, responsibility for the MoI targets occurs exclusively at the national level. For the sake of adaptation and contextualization, and not to give the impression to local actors that the 2030 Agenda is addressed only at the national level, we chose to exclude the MoI targets from our prioritization approach.

Localizing the SDGs involves implementing the SDGs in the logic of vertical, horizontal, and territorial integration. A siloed approach predominates, and moving toward an integrated approach is not straightforward.

Forming a group of leaders from different municipal services promoted horizontal integration. The leaders were not used to working in a multiservice group. Their collective work and dialogue broke down existing silos. The multiservice workshops greatly helped identify potential synergies and trade-offs. This horizontal integration occurred at several stages of our approach. During the diagnosis stage, our analysis of strategic documents, using the dictionary of expressions related to the SDG targets, identified the initial potential synergies. For example, we identified that the following targets touched all services: 4.4 (Skills for employment and entrepreneurship), 8.2 (Economic productivity), 9.1 (Sustainable infrastructure, economic development, well-being), 9.5 (Research, technological capabilities, innovation), 16.6 (Efficient, accountable, and transparent institutions), and 16.7 (Participation in decision-making). These shared targets do not systematically imply synergistic actions, but the diagnosis identified those actions carried out by several municipal services sharing common objectives. Our dictionary has proven to be a highly relevant and effective tool for undertaking this diagnosis.

The identification of 687 potential interactions formalized the links between city services and contributed to horizontal integration. The in-depth analysis and articulation of interactions illustrated the integrated nature of the actions of all services to members of the leader group. We observed that 92.8% of the interactions were positive by nature. This result closely matches the systemic analysis applied to the case of Sweden by Weitz et al. [31], where 96% of interactions were synergies. Referring to the classification of SDG targets to the five pillars of the 2030 Agenda (population, planet, prosperity, peace, and partnership) in Tremblay et al. [15], we observed that 83% of positive interactions (sum of +2 and greater) were linked to the same pillars. Half of the negative interactions related to different pillars. This illustrates the complexity of SDG targets and their interactions, and how the different pillars are integrated and indivisible.

The limits and challenges of vertical and territorial integration are multiple and complex. These types of integration refer to the principle of subsidiarity, "the search for the 'optimal scale of government,' [28] and the concept of multilevel governance, a system of continuous negotiation among nested governments at several territorial tiers" [43]. These limits and challenges are universal but vary depending on the context. Thus, there is not a single solution, but it is possible to provide adaptable reflections from our approach.

The actors of governance, at different scales, have variable levels of control and power over their context. This control varies from none (e.g., distribution of natural resources across the territory) to full (e.g., the adoption of policies). In addition, the actors interact according to different paradigms, at their level, in a complex system where the dominant paradigm of economic neoliberalism is omnipresent and, sometimes, underground [44–47]. It is well known that states tend to protect their powers despite the recognized importance of applying the principle of subsidiarity for implementing sustainable development [11].

The application of the principle of subsidiarity is linked directly to power issues, a very sensitive subject [24]. Local governments, to respond effectively to their sustainability challenges, must have the corresponding powers. From this perspective, Jones [48] writes, "Where national and state/provincial governments fail to act, city governments are severely limited in the implementation of [sustainability] policy." To address these issues, governments must collaborate. Using the governance assessment in the SDGT-PG, we guided local governments on the types of actions available to them on the basis of their specific governance context and target priority while also proposing actions at the national level. The terms "Search for opportunities" and "Advocacy at the appropriate governance level" apply to targets having a high priority level (urgent or priority) and whose governance is at the national level. This intersection between three parameters of the SDGT-PG helps guide the advocacy that local governments must undertake at higher levels. This observation does not guarantee success and an openness to dialogue; however, it provides guidelines for a structured argument based on an inclusive approach. The aim is to reduce what the OECD identified as the "policy gap" [49]. To achieve this, we

must establish mechanisms for collaborating between the levels of governance to make the implementation of public policies relevant and effective.

Localizing the SDGs requires an integrated commitment of human and financial resources [3]. SDSN [6] observed that, despite the importance of localizing SDGs, questions regarding capacities and mobilizing resources remain unanswered. Thus, the major constraints that cities face relate primarily to their limited political and fiscal powers, their lack of access to finance, the low levels of institutional capacity, the lack of multilevel government cooperation and integration, and the difficulty in establishing multi-stakeholder partnerships [6]. Becoming aware of these constraints is, however, a necessary step. Cities can act directly on a few aspects of sustainability, but they need the collaboration and openness of higher levels of governance to tackle the ensemble of issues. Open and empowered multilevel governance is essential for localizing SDGs horizontally, vertically, and territorially within an integrated approach [50].

#### *4.2. Opportunities*

The 2030 Agenda is mobilizing an enormous quantity of resources across the globe, and actors at all levels are developing appropriate tools and approaches. The number of scientific articles having "2030 Agenda" as a keyword has increased rapidly from 44 in 2015 to 246 in 2017 to 632 in 2020 (Scopus, search results using "2030 Agenda" as a keyword, 5 November 2020). The SDGs and their targets provide a relevant framework at all scales and are internationally recognized. The principle of integration is increasingly applied, and organizations (national, local, private) increasingly choose the SDG framework for the sake of multilevel consistency.

This willingness to join the SDG movement must be supported politically. In Quebec City, the mayor undertook the process, leading to a strategy and an action plan for sustainable development. This engagement at the highest levels of local government is essential for committing all necessary resources to ensure the success of the process [48]. Thus, the mayor's office established a competent project team that mobilized stakeholders, coordinated and analyzed activities, and developed the necessary strategy. Furthermore, a team of leaders, mobilized within all of the city's administrative units—because of the support of the directors of the various units mobilized by the mayor's office—has been trained in sustainable development issues. The team members communicated the progression of the approach and raised awareness with their colleagues. They sought their views at various stages of the process [48,51,52]. This multiservice mobilization was achieved through the mayor's commitment, through a top-down approach, to provide the means for achieving the results. The presence of a city councilor of the executive committee at every stage of the process testified to this political will. Mobilization at the highest level facilitates awareness of the efforts and actions to be implemented to vertically integrate the process. In our case study, Quebec City does not hold all the necessary powers to respond to the priorities that emerged from the prioritization exercise. City officers will be obliged to develop partnerships with higher governance bodies. As the mayor is the process holder in this case study, he will feel all the more invested and convinced of the need to carry out this task and to use the right communication channels to develop a multilevel collaboration. However, it is important to reiterate that the mobilization of the mayor alone cannot guarantee a successful implementation of sustainable development. It is also essential for all stakeholders to rally and face the challenges related to sustainability.

Cities must build on existing structures and actions already underway that fit within the sustainability framework to ensure optimal localization of the SDGs [14]. Our diagnosis provides a relatively rapid portrait of the situation, an exercise that can often be tedious. In our case study, we included the diagnosis at the stage of evaluating the performance parameter of the SDGT-PG. The use of the dictionary made it possible to undertake rigorous work with a minimum mobilization of human resources. It provided a solid starting point on which to build the remainder of the process and made it possible to identify a common starting point for all actors involved.

Crises can constitute opportunities to introduce a sustainability approach. Some previous crises (climate, financial, energy, etc.) have been drivers of change. For example, the 2008 financial crisis motivated some countries to embark on a transition movement [53,54]. The COVID-19 pandemic may also turn out to be an opportunity to provide arguments that favor the implementation of a sustainable development strategy. Quebec City, as most other local and national governments, must implement a post-containment/COVID-19 recovery strategy. This recovery strategy, linked to a sustainable development strategy, could offer an opportunity to facilitate ownership of the shift and the actions proposed by the city. In terms of sustainability, however, not all crises become opportunities. As stated in the 2030 Agenda: "There can be no sustainable development without peace and no peace without sustainable development" [1]. Thus, crises such as armed conflicts remain major obstacles to sustainability.

Local governance is the closest level of government to citizens and their issues. This reality allows, in theory, to quickly implement measures to respond effectively to identified problems. The local level involves fewer actors and fewer divergent issues than at the national level. This difference could explain why differing from "business as usual" can be easier to implement at the local level [55]. For example, in the context of local actions, actors are less influenced by the dominant paradigm of neoliberalism and thus allows the emergence of approaches considered more radical when compared with "business as usual" actions [47,55,56]. Cities should support grassroots initiatives [42] and socio-ecological transition projects [57] undertaken by local community groups in their territories. These partnerships are much easier to support by local governments that are in direct contact with these groups. Leadership at the top of the city hierarchy (top-down) and support of bottom-up initiatives are not contradictory and mutually reinforce each other [24]. In this sense, Quebec City has opened a dialogue with local partners from various civil society organizations with the objective of identifying challenges, issues, and opportunities, as well as proposals for action.

The identified limitations and opportunities routinely brought us back to the need for multilevel governance to ensure implementation of the 2030 Agenda [7,11]. The national level of governance, although holding most of the powers (Figure 7), must be aware that the national level is not always the most appropriate level in regard to local actors and issues [28]. The motivation of local governments can be hampered by the lack of collaboration of higher governance bodies [42]. The evaluation of the governance parameters shows higher authorities must collaborate with local governments. As Meuleman and Niestroy state [24], the issues and contexts differ at all levels, and a lack of integration and collaboration can lead to failure. A multilevel governance approach that relies on collaboration and cooperation will help promote vertical integration and policy coherence [50]. Our analyses identified targets representing opportunities to build such collaborations, allowing local and national authorities to optimize their contributions for achieving the goals of the 2030 Agenda.

#### **5. Conclusions**

Our approach aligns with the best practices for localizing SDGs and includes the concepts of contextualization, localization, systems approach, and integration. Although we apply this approach to the local level, it is flexible and adjustable enough to be applied at all levels of governance. Our approach provides a procedure that empowers sustainability actors in line with vertical and horizontal integration through capacity building, awareness, and direct participation, a procedure that, to our knowledge, has not been provided in previous studies focused on the local level.

Each application of our approach should be contextualized, as the opportunities and limitations differ from place to place. In our case, we were limited by a lack of data; the indicators of the SDG targets had yet to be assessed. Therefore, it was impossible to accurately assess performance. We stated that they were potential performances, and we remain conservative in our assessments by not describing any targets as being fully achieved.

The systemic tools and approach presented in our study will help planners develop strategies and action plans for implementing the 2030 Agenda. Although our approach is complete, it can only be implemented with a mobilization at the highest level and with the involvement of stakeholders who represent the complexity of the system in which the agenda is being implemented. SDG localization faces other challenges, in particular the adaptation of SDG tools and approaches to the private sector, where each particular sector comprises its challenges, contexts, opportunities and specific scopes of organizations governance. Future research could help define, as in the present study, good practices in localizing the SDGs, and methodologies for adapting the 2030 Agenda to the private sector.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2071-105 0/13/5/2520/s1, Table S1: List of analyzed documents, Table S2: Dictionary linked to the content of the 2030 Agenda, Table S3: Matrix of links between SDG targets and the analyzed Quebec City strategic documents inspired by the Rapid integrated assessment (RIA), Table S4: Cross-impact matrix, Table S5: Sustainable development goal target prioritization grid of Quebec City.

**Author Contributions:** Conceptualization, D.T., C.V., and S.G.; SDGT-PG methodology, O.R., D.T., and C.V.; WordStat, D.T.; Validation, D.T., J.-F.B., O.R., S.D., S.G., and C.V.; Formal analysis, D.T., S.G., and S.D.; Writing—original draft preparation, D.T.; Supervision, J.-F.B. and C.V.; Project administration, C.V. All authors commented all the sections and reviewed the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data is contained within the article or Supplementary Material.

**Acknowledgments:** The authors acknowledge the contribution of the project team and the group of city leaders. The authors thank the project director, Yohan Maubrun, for his leadership and openness. The authors acknowledge the support of the Organisation Internationale de la Francophonie and its subsidiary body, l'Institut de la Francophonie pour le Développement Durable, for their support in the development of systemic sustainability analysis tools since 2012. The lead author thanks the Université du Québec à Chicoutimi and the Chaire en éco-conseil de l'Université du Québec à Chicoutimi for their financial support. We also thank Murray Hay of Maxafeau Editing Services.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Áron Szennay 1,2 , Cecília Szigeti 3,\* , Judit Beke <sup>4</sup> and László Radácsi <sup>5</sup>**


**Abstract:** Small- and medium-sized enterprises (SMEs) play a significant role in the national economies of the EU member states. This economic activity has an inevitable environmental impact; however, environmental performance indicators are mostly measured at larger companies. Since the ecological footprint (EF) is a suitable measure of unsustainability, this paper considers it as a measure of the environmental impact of SMEs. An EF calculator for SMEs was developed that is freely available online, and it is a methodological innovation per se. Our previous research projects highlighted that the calculator must be easy-to-use and reliable; therefore, the calculator considers only the common, standardizable, and comparable elements of EF. Our results are based on validated ecological footprint data of 73 Hungarian SMEs surveyed by an online ecological footprint calculator. In order to validate and test the usefulness of the calculator, interviews were conducted with respondents, and results were also checked. The paper presents benchmark data of ecological footprint indicators of SMEs obtained from five groups of enterprises (construction, white-collar jobs, production, retail and/or wholesale trade, and transportation). Statistical results are explained with qualitative data (such as environmental protection initiatives, business models, etc.) of the SMEs surveyed. Our findings could be used as a benchmark for the assessment of environmental performance of SMEs in Central- and Eastern Europe.

**Keywords:** ecological footprint; environmental performance of SMEs

#### **1. Introduction**

There is a broad consensus around the need and usefulness of indicators and metrics to define the planetary boundaries. Humanity's demand on resources has been expanding, which has a significant impact on the Earth system; therefore, many researchers now believe that this era can be considered as a new geological epoch, the so-called Anthropocene [1]. The World Overshoot Day, calculated by Global Footprint Network (GFN), is a high-level and easy-to-understand indicator of global (un)sustainability, since it "marks the date when humanity's demand for ecological resources and services in a given year exceeds what Earth can regenerate in that year" [2]. Since 1970, this date occurs before 31st December each year, and, since the beginning of the 2010s, it lands around the 1st of August. This figure means that in the 2010s, humanity used up approximately 1.7 times more resources each year than the ecosystems of the Earth can regenerate. Although environmentally friendly (i.e., "green") consumption habits and technologies are becoming more common, recent studies show that even conscious consumers change their habits occasionally (e.g.,

**Citation:** Szennay, Á.; Szigeti, C.; Beke, J.; Radácsi, L. Ecological Footprint as an Indicator of Corporate Environmental Performance— Empirical Evidence from Hungarian SMEs. *Sustainability* **2021**, *13*, 1000. https://doi.org/10.3390/su13021000

Received: 26 November 2020 Accepted: 16 January 2021 Published: 19 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

during holiday) [3]. Considering this, it is not surprising that Mathis Wackernagel [4] called our economy the largest Ponzi scheme ever. However, as a result of the COVID-19 pandemic and the lockdown measures introduced in the developed and emerging world, the Overshoot Day landed on 22nd August in 2020.

Environmental sustainability (with special regard to the reduction of greenhouse gas emission and the increase of renewable energies) is one of the headline targets the Europe 2020 strategy of the European Union (EU) [5]. Since Europe's 25 million small- and mediumsized enterprises (SMEs) play a significant role in the economies of EU member states, their contribution to sustainable development is also crucial. SMEs make up over 99% of all enterprises in all EU countries, they generate around two-thirds of all jobs and account for more than half of EU's GDP [6]. Evidence shows that both the regulatory stakeholder pressure and organizational stakeholder pressure positively influence green production practices, corporate reputation, and the environmental performance of manufacturing SMEs [7], which means that the sustainability efforts of both the EU as a whole and the individual member states have a positive impact on the attitudes of SME managers towards sustainability. This finding is supported by evidence from the energy sector, i.e., debt increases the value of SMEs in countries with strong environmental commitment, which makes it possible to facilitate growth with additional external capital [8]. Italian evidence highlights, however, that decision-makers of SMEs "have a high school diploma mainly used bank loans or overdrafts as compared to those that received formal training" [9]. Nonetheless, firms with external capital must maintain financial capacity to repay it, which might create significant problems in case of a crisis situation [10], and capital structure considerations may also play a crucial role [9,11,12]. Another aspect is that a large share of SMEs are family businesses that make up between 57 and 66 percent of the enterprises with 3 and 99 employees in Hungary [13]. Evidence shows that Hungarian family businesses have better chances of survival and create higher value added than non-family businesses [14].

Experience has shown that, although several managers of SMEs are interested in metrics on environmental performance, their businesses/companies cannot afford paying for comprehensive environmental audit and advisory; therefore, they do not have enough experience in selecting the most appropriate measures. Our results suggest that the ecological footprint (EF) is a suitable metric for SMEs because (1) it is easy to understand, therefore making it easy for even managers who do not have enough relevant expertise to use it; (2) the calculation is standardizable, therefore capable of providing performance metrics at a low cost or even for free; and (3) quantitative performance indicators allow them to support the selection of the most appropriate projects or measures to enhance corporate environmental performance (CEP). Our aim was to develop an easy-to-use EF calculator for SMEs which could measure the common elements of corporate environmental impacts reliably. Based on experiences with carbon footprint calculators, it has been found that there is a trade-off between accuracy and simplicity. A calculator that measures the EF of SMEs is needed because, whereas large enterprises have sufficient resources to make unique calculations, it can be difficult for SMEs to find resources and expertise [15]. The results of standardized calculations can be complemented with unique items (e.g., material consumption or more sophisticated data on meals) or longitudinal assessment of CEP can be conducted. Based on the results of our previous analyses, the usefulness and accuracy of the calculator developed was validated, the results were discussed with the respondents, and we made attempts to improve the calculator [16]. Nevertheless, lacking benchmark data can be considered as the most critical problem. Therefore, this paper aimed to calculate sectoral comparative benchmark data.

This paper is structured as follows: Chapter 2 summarizes the concept of the EF and its potential role in measuring of corporate environmental performance. At the end of the chapter, some examples of sectoral EF calculations are presented. The third chapter gives an account of the methodology and the sample used, while the fourth chapter summarizes our results.

#### **2. Theoretical Framework**

#### *2.1. The Concept of Ecological Footprint*

The ecological footprint (EF) concept was developed by Mathis Wackernagel and William E. Rees [17] in 1996. Since the introduction of the concept, the EF has been used to measure environmental sustainability both at a global level and of individual consumption, as well [18–22]. Nonetheless, other indicators could also be used for measuring environmental sustainability [23–25], but it is only the EF that indicates the upper limit of growth properly [26]. The GFN started its National Footprint Accounts (NFA) program in 2003 based on Wackernagel's calculations, and, since then, the EF calculation methodology framework is regularly updated [27]. The most recent update, which contains data sets for most countries and the world from 1961 and 2017, was published in 2020 [28].

The indicator represents the size of land needed for humanity at a given level of technological development to satisfy its needs and absorb waste generated. Compared to other indicators of environmental impact, the most important advantages of the EF are the following: the EF is easy to understand, and it is relatively easy to determine the upper limit of sustainable consumption.

According to the concept of GFN, EF considers six land types: built-up land, forest products, grazing land, cropland, fishing ground, and carbon. Resource usage is expressed and measured by land usage, which are standardized with the help of equivalence factors (EQF) in global hectares (gha)—globally comparable hectares. This conversion number serves as a tool to compare different land types (e.g., cropland, forest, etc.). Since productivity of the particular land types may show regional differences, an adjustment-specific yield factor (YF) is applied [29].

Besides the spread of spatial calculations [30–34], corporate calculations were also introduced. The principles of corporate EF calculations were developed by Nicky Chambers and her colleagues in 2000 [35]. Although the concept of EF calculation was developed by examining (un)sustainability at a macro-level, it is equally useful at a micro-scale, for example, for corporations or other organizations. EF calculations could help corporations to find intervention fields [36] where environmental measures are the most effective, i.e., a particular amount of money spent has the greatest positive impact on corporate environmental performance.

A clear sign of global unsustainability of CO<sup>2</sup> emissions is that, although the usage of all land types has been increasing since the Industrial Revolution, the increase in carbon usage had the most significant role. Carbon usage grew from 43.8% to 59.9% of total land usage between 1961 and 2018, while it has an annual growth rate of 2.54%, the second highest among the land types [37].

#### *2.2. Ecological Footprint as a Possible Corporate Environmental Performance Indicator*

The usage of natural resources of business operations has an obvious impact. The concept of environmental performance attempts to measure and manage such impacts. Trumpp et al. [38] reviewed the related literature and identified 16 articles that give a definition of corporate environmental performance (CEP). Since 5 articles refer to the definition of International Organization for Standardization (ISO) standard 14031, and they capture the most important aspects of the 11 other definitions, the authors argue that "the ISO definition provides an encompassing and parsimonious definition". The ISO standard defines environmental performance as the "measurable results of an organization's management of its environmental aspects" [39]. However, the exact and comparable measurement of CEP is not easy because the ISO definition is "fuzzy enough to impose no clear conceptual boundaries" [40].

According to Jung et al., environmental performance measures can be grouped into five categories [41], where general environmental management (GEM) represents the strategic level, while the other four categories (input, process and operation, output, and outcome) are operational. Input measurement considers the raw material (for example, water, timber, metals, etc.) and energy (electricity, fossil fuels, etc.) consumption, while

output measures reveal desirable outputs (energy or pollutant savings) and undesirable outputs, for example, emission of air, water, or even land pollutants. As Schultze and Trommer summarize, these two measures refer to "companies' physical interactions with the natural environment" [42]. Process measures deal with optimization of corporate operations to enhance CEP, i.e., the increase in material efficiency and raising awareness of employees and suppliers. Outcome measures concern financial outcomes of the actions taken (for instance, avoided costs, fines, penalties, or even cost savings) and non-financial outcomes, which comprise mainly stakeholder relations, for example, complaints, lawsuits, or reputational issues [41].

We argue that the EF can be considered as an input/output environmental performance measure, since it focuses on the resources (raw material and energy consume, built-up land, etc.) that are consumed in business operations. Furthermore, we argue that the EF is a suitable tool to measure and manage CEP because ecological footprint:


Although standardized calculations and methodologies of EF calculators can be considered as an advantage, especially for SMEs and individuals, Harangozó and Szigeti found that online corporate carbon footprint calculators may have validity and reliability issues, even in the case of the simplest business operations [15]. The authors suggested that the reliability of online EF calculators can be enhanced with more detailed input data and using local data (e.g., electricity mix). Furthermore, while corporate carbon footprint calculations are more commonly used among SMEs than EF calculations [43], understanding further aspects of EF brings new insights to improving environmental performance at the SME level.

#### *2.3. Impact of Environmental Performance on Financial Performance*

Although some authors suggested that the EF can be reduced at low or no cost [44], further engagements consume scarce corporate resources (e.g., financial funds, human resources, managerial attention, etc.). Since these resources could be used for other projects with net present value, companies will engage only in environmental projects the benefits of which exceed their costs. The link between sustainability and corporate financial performance (CFP) is an empirically well-studied area (see References [45–52]). Meta analyses (e.g., References [53–55]) mostly showed a positive relationship. Although there is no consensus on which indicators measure sustainability the best, we have found no study that used EF as a proxy. We suggest, however, that EF could be a suitable indicator of CEP because, (1) as we mentioned above, the EF has some advantages over other indicators; and (2) the EF is measured in ratio scale, therefore making the link between CEP and CFP examinable with more sophisticated methods than in case of other proxies measured by dummy variables (e.g., certificates, non-financial disclosures, etc.).

According to the theoretical model of Schaltegger and Synnestvedt [56], up to a point, environmental efforts pay off (see point A in Figure 1); after that, marginal benefits will be decreasing. Nonetheless, further environmental protection efforts may be confirmed because the economic performance will be higher than at the starting point up to point B in Figure 1. Two other consequences are as follows: (1) due to managerial skills, attitudes towards and the ignorance of environmental performance may vary at a given level of economic performance; and (2) several factors (e.g., change of consumer attitude, technological development, etc.) may allow to implement further environmental protection efforts, i.e., it causes the curve to shift right (see dashed line in Figure 1).

**Figure 1.** Possible relations between corporate environmental protection and economic success. Source: Reference [56].

By analyzing a sample of 4186 companies in OECD countries, empirical evidence on the positive relationship between environmental protection efforts and financial performance has been found [57]. Furthermore, Zhang et al. [58] provide a more sophisticated version of the model by adding the effects of environmental uncertainty. The authors suggest that environmental uncertainty may influence both costs and benefits of CEP through several factors. Their empirical findings show that the link between the corporate environmental performance (CEP) and the CFP is "steeper and of a lower plateau in higher levels of environmental uncertainty characterized by high dynamism, low munificence, and high complexity".

#### *2.4. Sectoral Average of EF*

As it was mentioned earlier, EF was developed to calculate environmental impacts of larger areas (regions, states, countries, etc.) and individuals or their households. In addition, the EF concept was complemented with other, specific calculations to determine sustainability of industrial branches or companies, among others [44]. Although ecological and carbon footprint calculations may be suitable tools for measuring both environmental and economic improvements and related reporting [36], however, one of the main limitations of corporate footprint calculations is the lack of benchmark data; namely, there are no industrial or sectoral averages available to assess the calculated footprint value. Recent research [59–62] aims to fill this gap and to provide guidance for both advisors and managers to assess CEP. To highlight both methodological approaches and impacts of different business models on EF values, in this subsection, we provide a brief insight on the results from three different specific EF calculations.

Mining is one the most CO<sup>2</sup> intensive sectors; thus, there is a legitimate demand on calculating total EF and optimizing it. Murakami et al. [59] have found that underground mines (1) have significantly lower EF for built-up land due to their smaller land-use change, and (2) fossil fuel consumption is also much lower due to their electrification; therefore, the EF could be decreased by using renewable energy sources.

Residential homes have a rather high EF in the EU. Energy consumption of households makes 26.1 percent of total final energy consumption in the EU, out of which heating is the largest portion (63.6%) [63]. Residential buildings have an average energy intensity of 180 kWh/m<sup>2</sup> , but it shows significant differences among countries, even when they are

located in the same climate zone [64]. Another aspect that studies have shown is the high variability of emissions associated with construction and operation of buildings during their life cycle [65]. Since the Energy Performance of Buildings Directive requires all new buildings to be nearly zero-energy by the end of 2020 in the EU [66], EF minimalization measures should focus on the construction phase. Incorporating EF figures in construction cost databases could support in optimization of both environmental impact and costs of construction. A case study from Andalusia (Spain) highlights that the substitution of traditional construction units with lower EF solutions could result in 18% reduction of the EF, while the total cost increased only by 7% [60]. Using recycled materials (e.g., wood, concrete, steel) could reduce the EF significantly [61].

Since Hungary is an export-oriented, open, and small market economy, industrial parks can be considered as important engines of economic growth and regional development (see References [67,68]). In a case study from China [62] researchers claim that through eco-industrial transformation, EF of HETDA industry park of China can be reduced by 15.9 percent [62]. Nevertheless, other studies have shown that most eco-industrial parks are at a very early stage of development [69].

#### **3. Methodology**

A mixed methodology was used in this study. On the one hand, an online ecological footprint calculator was developed according to the special needs of the SME sector. A brief outline of the calculator can be found in the appendix. On the other hand, with a special regard to EF, we conducted interviews and mini case studies to gain deeper understanding of the unique features of SMEs operating in different sectors.

Both the monetary and employment figures are standardized. First, although financial data was collected in local currency (Hungarian forint, HUF) in the survey, results are expressed in euros. Since survey data considers both 2018 and 2019, an arithmetic average of daily exchange rates of the European Central Bank was applied (322.0932 HUF/EUR). Second, all employment data are expressed in full-time equivalents.

#### *3.1. Calculation of EF*

The Table 1 cites only articles in which figures, methodology, etc., were directly used in the calculator developed.

Although material usage was part of a previous version of the calculator, later it was excluded from the formula due to the fact that the 500+ materials we employed in the explorative phase could not be standardized in a proper way [16].


**Table 1.** Element of ecological footprint (EF) calculated, their short description, and calculation method.

The EF of meals was calculated on the basis of Hungarian average values of people's food consumption [70] (see Equation (1)). Average values do not take into consideration food consumption exceeding the minimal human needs (e.g., alcohol or candy consumption, import goods, etc.); therefore, they provide a rather lower estimate than the real figures. To achieve more accurate results, different EF factors were used for both females and males, as well as the characteristics of jobs (i.e., white collar or blue collar). Since the abovementioned values reflect the total food consumption of a given year, we assumed that employees have *n* working days a year, and they consume *i* percent of their meals at the workplace, where *n* and *i* values are given by the SMEs surveyed for each employee category. Calculation of the EF of food consumption was as follows:

$$\mathrm{EF\_{matls}} = \frac{n\_{femlsle}}{365} \times \dot{i}\_{femlsle} \times \sum \mathrm{E}\_{jib} \times \mathrm{EF} \, factor\_{jib} + \frac{n\_{male}}{365} \times \dot{i}\_{mle} \times \sum \mathrm{E}\_{jib} \times \mathrm{EF} \, factor\_{jib} \tag{1}$$

where:

*n*—number of working days of both female and male employees,

*i*—percent of at workplace consumed meals,

*E*—number of employees at a given job type (e.g., white collar or blue collar), and *EF factor*—EF factor of each job type (e.g., white collar or blue collar).

The EF of food consumption is one of those EF elements which could differ significantly among regions [73]. An EF calculation on food consumption conducted by a Polish research team showed a much larger EF per capita figure. (It is interesting to note that Poland is another Central Eastern European country and EU member state.) The higher number is partly due to methodological considerations.

Spanish and Chilean EF values on food consumption, both of them based on Food and Agriculture Organization (FAO) of the United Nations data, show significant differences too, 0.97 and 1.43 gha per person, respectively [61].

According to our methodology, the EF of water consumption calculates with the EF of building and maintenance of water pipelines, sewage, and wastewater treating facilities. Since exact measures are not available, we assumed that the EF of water consumption is a function of employee number (see Equation (2)).

$$\text{EF}\_{\text{water}} = \left(\text{E}\_{female} + \text{E}\_{\text{male}}\right) \times \text{EF} \, factor\_{water} \tag{2}$$

where:

*E*—number of both female and male employees; and

*EF factor*—EF factor of water consumption.

The EF of built-up area was calculated on the base of buildings' ground floor and other covered and non-water absorbent (e.g., asphalt or concrete) surface (see Equation (3)).

$$EF\_{\text{build}-up} = \left(\mathbb{S}\_{\text{building}} + \mathbb{S}\_{\text{other}}\right) \times EF \left[factor\_{\text{build}-up}\right] \tag{3}$$

where:

*S*—covered surface, both ground floor of buildings and other non-water absorbent surfaces, in square meters; and

*EF factor*—EF factor of built-up area.

The EF of electricity consumption is based on carbon intensity figure (264 g CO2e/kWh 2015) of International Energy Agency (IEA) [71]. CO2e (carbon dioxide equivalent) is a term for describing different greenhouse gases in a common unit. For any quantity and type of greenhouse gas, CO2e signifies the amount of CO<sup>2</sup> which would have the equivalent global warming impact. This value was adjusted from CO2e to CO<sup>2</sup> figures by the British organization called Department for Environment, Food and Rural Affairs (DEFRA) database 2018 [72] in order to determine carbon intensity values in CO2/kWh instead of in CO2e. After that we added estimated impacts of energy generation and losses of electricity transmission and distribution. Although the renewable energy generation of enterprises was taken into consideration, its EF factor was determined as 0. The EF of electricity consumption was calculated as follows:

$$EF\_{\text{electricity}} = El\_{\text{grid}} \times EF \left[ \text{factor}\_{\text{electricity}} + El\_{\text{remavailable}} \times 0\_{\star} \right] \tag{4}$$

where:

*El*—electricity consumed (i.e., bought from the electricity grid or generated by the enterprise); and

*EF factor*—EF factor of electricity consumption.

The calculation of EF of heating and boiling is based on carbon intensity factors of DEFRA database 2018 [72]. It includes the usage of different fossil energy sources, e.g., natural gas or even burning coal.

$$EF\_{\text{heating and boiling}} = \sum FE\_{\text{i}} \times EF \, factor\_{\text{i}\prime} \tag{5}$$

where:

*FES*—fossil energy source (e.g., megajoules of natural gas or tonnes of wood logs); and *EF factor*—EF factor of specific fossil energy source.

Besides heating and boiling, transportation and the related carbon footprint generally makes up the largest portion of EF [74]; therefore, our online EF calculator provides the following options to determine the EF:


Since SMEs in general use several different transportation modes, only the first two calculation methods are mutually exclusive. All carbon intensity factors are based on the DEFRA 2018 database [72].

*3.2. The Sample*

Enterprises in our sample were required to have the following attributes:


Data was collected from three sources: (1) SMEs known from our professional network or from our university networks; (2) commercial and industrial chambers in Hungary were asked to send calls for survey to their member companies, and we participated in some of their events; and (3) students were asked to assist with our study. Mini case studies were conducted about most of the companies surveyed to gather additional qualitative data.

Companies were filtered out from our sample as an outlier when one or more figures varied significantly from other companies of the same group and we had no plausible explanation for this (e.g., equipment used, working processes, etc.).

Anecdotical evidence suggest that the SMEs of different business activities may have similar EF. Therefore, a preliminary qualitative analysis was conducted to classify SMEs on the basis of the determining factors of their ecological footprint, i.e., based on the attributes of their CEF. This is inevitably different from statistical classifications (i.e., NACE in the EU or SIC in the USA). We suggest that a more detailed and more accurate result could be achieved by analyzing a larger database. For example, white-collar jobs have similar

environmental impact, regardless of whether the enterprise is involved in bookkeeping, software development, civil engineering planning, or even fashion design. The ecological footprint of white-collar jobs is determined mainly by (1) the conditions of the property used (place, size, insulation, effectiveness, and usage of air conditioning and heating, etc.), (2) commuting habits of employees and home office opportunities, (3) number and length of business trips and vehicles used, and (4) the number of employees. The study focuses on the following five groups of SMEs (see Table 2):


**Table 2.** Classification of SMEs analyzed.

Variation of EF among group of enterprises can be explained by several coexisting factors:


One of the limitations of the EF calculator is that it ignores all the factors that are beyond the control of companies. Accordingly, financial performance is measured by an adjusted value added, which is calculated on the available accounting data as the sum of personnel costs, amortization, and after-tax profit. Adjustment had to be made because of a simplified tax type eligible only for small companies. If a company chooses this tax type, it substitutes corporate tax and social contributions of employment. Since personnel costs of companies of different types are directly not comparable, we chose after-tax profit instead of pre-tax profit. We suggest that these kinds of calculations provide more comparable results among the analyzed SMEs but have the limitation that all value-added figures presented show an underestimation of real values.

#### **4. Results**

Our sample consists of 73 SMEs from the five groups. Four out of the five groups have 15–20 valid items, while the smallest sub-sample (transportation) comprises only 4 items. This can be explained by the relative simpleness of the sector; the EF of these SMEs is determined almost completely by fuel consumption (liters of diesel per 100 km). Detailed results are presented in the following subsections. For detailed numerical information see Tables 3 and 4.


**Table 3.**Descriptive statistics.


**Table4.**Correlations.

\*\*. Correlation is significant at the 0.01 level (2-tailed). \*. Correlation is significant at the 0.05 level (2-tailed).

#### *4.1. Construction*

Activities of construction enterprises in our sample, ranging from civil engineering, structural architecture, and some special construction firms (e.g., planning, implementing solar panels and other electric equipment on buildings, installing shading equipment, etc.), are also present. They have an average EF of 1.25 gha/employee (confidence interval (CI): 0.87–1.62), an eco-efficiency of 0.089 gha/thousand EUR adjusted value added (CI: 0.065–0.113), and specific value added of 15.4 thousand EUR/employee (CI: 11.94–18.85). Positive correlation between eco-efficiency and specific EF (*p* < 0.01) shows that more eco-efficient construction also has lower the EF per employee figures. Significant correlations between other variables could not be identified.

The EF of construction enterprises is determined mostly by the consumption and efficiency of vehicles and other equipment used. Our mini cases show that managers mostly aimed to reduce fuel consumption; therefore, vehicles are regularly replaced by more efficient ones, private vehicle use is restricted, and employees are collected by a company vehicle. It is interesting to note, however, that the prestige of driving a car is of great importance for many people, and they drive to work even if the commuting distance is less than a few kilometers. Nevertheless, a moderate vehicle use may be allowed in most construction enterprises, since the second half of 2010s is marked with a shortage of trained and experienced professionals. Another issue is that, although there is governmental aid for purchasing battery electric vans or cars, the managers interviewed are concerned about the higher price and the lack of experience; therefore, only a small car that was used for the everyday corporate errands was to be replaced.

If the company has a larger office building, it is often retrofitted or is even equipped with solar panels.

#### *4.2. White-Collar Jobs*

White-collar jobs include mostly financial and accounting services (bookkeeping, tax advisory services, auditing, etc.), but engineering, education, or even software development enterprises are present in the sample. The group has the smallest environmental impact—an average of 0.46 gha/employee (CI: 0.32–0.60) and average eco-efficiency of 0.051 gha/thousand EUR adjusted value added (CI: 0.029–0.074), while the average specific value added is less than in other sectors, 15.29 thousand EUR/employee (CI: 7.23–23.34). Results of the correlation analysis show that (1) more eco-efficient enterprises have significantly lower specific EF figures (*p* < 0.05) and (2) higher specific value added (*p* < 0.05). This latter result means that engagement in environmental protection measures and/or project may be profitable.

Since working in an office is a human capital-intensive activity, its EF is determined mostly by the energy-efficiency of the buildings used and by the commuting practices and working trips of the workforce. While the former figure can easily be reduced by insulation and/or renovation of the buildings, by using energy-efficient lightning or even by implementing solar panels, reducing the latter figure is a more complicated issue. On the one hand, the COVID-19 pandemic showed that personal contacts can be at least partly substituted by online meetings, but working trips could be necessary in some cases; for example, engineers must visit working fields or even cultural determinations may require personal meetings. On the other hand, the prestige of commuting by car and/or living in urban agglomerations may influence the habits of employees. Furthermore, employees mostly use their own cars; therefore, it is out of the managers' control. Based on our findings, we recommend promoting more sustainable ways of commuting. For example, when it is feasible, businesses should provide shower and changing facilities for cyclists in the workplace, but biking events and/or actions may influence commuting habits, as well. Of course, financial stimuli could also be used, for example, cutting contributions on commuting with a car and providing benefits for public transport usage instead.

#### *4.3. Production*

Producer companies in our sample are very diverse—they range from manufacturing spices, wooden toys for playgrounds to producing vehicles. The EF figures of these activities differ substantially. Specific EF is EF 1.47 gha/employee (CI: 0.85–2.08) on average in this group, while eco-efficiency is favorable, 0.067 gha/thousand EUR (CI: 0.033–0.100), and specific value added is 32.98 thousand EUR/employee (CI: 14.04–51.93) due to the higher adjusted value added. Just as in the case of construction enterprises, correlation analysis shows a significant relationship only between eco-efficiency and specific EF (*p* < 0.01).

Production is technology-intensive, so EF is also highly determined by working processes and equipment used. Our mini cases show that companies attempt to implement both up-to-date working processes and efficient equipment, but EF figures are influenced significantly by other factors, such as industrial specialties, level of market competition, managerial attitudes, and governmental and/or EU grants.

#### *4.4. Retail and Wholesale Trade*

Retail and/or wholesale trade companies range from pharmacies and other fastmoving consumer goods (FMCG) stores to wholesale of electronic components or even veterinary items. The most significant difference among companies is the following: (1) transportation and/or home delivery of goods with own vehicle or by a third party; and (2) special storage needs of goods sold (e.g., storage of frozen or chilled goods have much higher energy consumption than of recyclable waste). Specific EF of the sector is on average 1.10 gha/employee (CI: 0.73–1.47), and eco-efficiency is 0.088 gha/thousand EUR (CI: 0.050–0.126), while specific added value lies at 17.24 thousand EUR/employee (CI: 12.64–21.84). We found significant correlation between eco-efficiency and specific value added (*p* < 0.05). It means, as in the case of office activities, that more eco-efficient enterprises have generally higher added value per employee; namely, there is a positive relationship between corporate environmental performance and value-added creation.

Our cases reveal that companies of the clusters sector have similar challenges as of offices, namely energetical characteristics of the buildings used.

#### *4.5. Transportation*

The fifth group is transportation, which is the most EF-intensive sector in our analysis. Specific EF figure of transportation companies is 20.15 gha/employee (CI: 17.00–23.30), which is 16 times higher than of construction companies. Average eco-efficiency is 1.055 gha/thousand EUR (CI: 0.410–1.701), and specific value added is 20.64 thousand EUR/employee (CI: 11.79–29.49). Similar to other groups, our correlation analysis identifies significant relationship between eco-efficiency and specific value added (*p* < 0.05), establishing a positive connection between corporate financial and environmental performance.

Our cases show that there are four main routes to reducing the EF: (1) increasing the efficiency of vehicle technology, which means not only lower consumption in relative terms (liters per 100 km), but highway tolls and maintenance costs are significantly lower, as well; (2) monitoring fuel consumption could mitigate misuse of tanked fuel and provide data for route optimization; (3) route optimization could decrease mileage of trucks, which means lower consumption in absolute terms (liters per trip); and (4) using lower-carbon fuels (e.g., hydrogen).

#### **5. Conclusions and Discussions**

The paper aimed to develop an easy-to-use EF calculator for SMEs which could measure the common elements of corporate environmental impacts reliably. Results are based on a sample of 73 corporate EF calculated by an online EF calculator; thus, identical approach and methodology was assured.

Our results primarily have practical implications, as they show that it is feasible to develop an EF calculator for SMEs which can provide reliable figures and is easyto-use. As anecdotical evidence suggested, SMEs can be classified on the basis of the

determining factors of their ecological footprint. Considering their EF figure calculated with a standardized methodology, benchmark data could be also calculated to measure CEP. Using a larger sample, a more detailed classification and more accurate benchmark data could be provided.

According to our results, there is significant and negative correlation between ecoefficiency (EF/value added) and added value per capita in some groups (office activities, retail and/or wholesale trade, transportation). Similarly, significant correlation cannot be found in other analyzed groups (construction, production). These findings suggest that CEP does not influence the financial performance of the analyzed SMEs negatively; rather, there is a positive link in the case of some groups. A possible explanation for the difference of sectors' results may be that production and construction are both highly technology intensive sectors; therefore, environmental protection measures are either too expensive (e.g., more advanced production technology) or there is no available solution (e.g., heavy duty vehicles with electric powertrain).

It is remarkable that transportation enterprises have much higher EF figures than enterprises from other groups. On the one hand, it highlights the importance of locality [76]. On the other hand, transportation connect participants of the value chains. It means that activities with significantly different EF figures in a value chain are separated into several enterprises, so it would seem that CEP of a specific enterprise might be high, but actually only a more environmentally intensive element of the value chain is outsourced to it.

Our results have four main limitations. First, there is no widely used and accepted methodology of conducting easy-to-use and reliable EF calculator for SMEs; therefore, we could not lean on former experiences or calculators that we could have used for the development and testing of our calculator. To provide reliable results, only the common elements of corporate EF were taken into consideration. Although EF of material usage might make up a significant proportion of corporate EF, we suggest that the number and diversity of materials would make the calculator too complex and complicated to use. Second, the sample used in the analysis is small and does not represent the real environmental performance of Hungarian SMEs. Furthermore, we suggest that the sample is positively biased because companies with higher environmental performance are more willing to participate in the survey. Third, although the most financial data were validated on the basis of disclosed financial reports, and we adjusted them according to findings of qualitative methods, firm-specific parameters (e.g., part-time employment, tax optimization, accounting policies, business models, etc.) could significantly influence the results. Fourth, our calculator does not consider material usage of companies; thus, the provided EF values are consistently underestimated.

The results presented in this article show a transition phase between individual and mass calculations; therefore, our future research aims to provide more accurate benchmark data on EF values of SMEs based on a larger sample size. This step would make it possible to conduct more sophisticated analyses using moderating variables (such as corporate governance [77,78], family businesses [79], developed, emerging, and transitional countries, etc.). Another aspect could be to complement the data set with other sectors, for example, with services, agriculture, etc., and to compare EF values of SMEs based in different countries.

**Author Contributions:** Conceptualization and methodology, C.S.; data collection, Á.S., J.B., C.S.; validation and formal analysis, Á.S.; writing—original draft preparation, Á.S.; writing—review and editing, J.B., L.R.; supervision and project administration, C.S., L.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by a grant from the Higher Education Institutional Excellence Program of the Hungarian Ministry of Innovation and Technology to Budapest Business School (NKFIH-1259-8/2019). The APC was funded by Budapest Business School—University of Applied Sciences.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data are not publicly available because respondents did not permit data usage of third parties.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Sustainability* Editorial Office E-mail: sustainability@mdpi.com www.mdpi.com/journal/sustainability

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-2950-9