Clusters as Tools to Measure Sustainable Value Chains

Abstract

The literature and practice have proven the connection between competitiveness at all levels (company, region, national) and its position in international value chains, hence the need to “measure” their economic impact. Traditionally, this has been conducted by using complex quantitative data based on statistical sources translated into input/output tables that are difficult to calculate and interpret and rely on outdated data. While the contribution of clusters as drivers of economic competitiveness has been extensively debated over the last 30 years, it is more recently, after the COVID-19 pandemic, leading to tremendous disruptions in international value chains, that their role of generators and drivers of international value chains has been recognized, proven by the rapid response they have been able to provide in “repairing” the disturbances. The current paper proposes a cluster-based value chain analyses method in which the main measurement unit is the density of the chosen indicator along the value chain links (number of enterprises, turnover, R&D expenditure, exports). The results were checked by classical methods and proven to be congruent. The method allows for a rapid response to sudden disruptions and can be used for both cluster managers as well as economic policymakers at regional and national levels.

1. Introduction

The term “value chain” was introduced by Michael Porter in 1985 [1] as a fundamental concept for business strategy development meant to increase the competitiveness of the company by considering all activities implied by production and consumption. It is closely related to the “twin” concept of “supply chain”, which refers to the network of activities, enterprises, resources, and technologies involved in the creation and delivery of products and services from the suppliers toward the client. The main difference between them lies in their concentration. The “value chain” concentrates on the created added value in every step of the product life cycle. Activities such as design, marketing, and others are taken into consideration; the “supply chain” refers to required processes on the way from the supplier to the client [2]. Supply chains, as governance mechanisms, provide firms with the tools to navigate and thrive in dynamic industrial ecosystems, ensuring that they can meet both the current and future challenges effectively [3]. As to the relation between GVCs and supply chains, a firm can participate in international production sharing either by exporting its domestic value added in intermediate goods for domestic consumption (a), export to third countries (b), using the value added of other countries to produce its gross exports (c), or for domestic consumption (d) [4]. As global value chains (GVCs) become more and more complex, firms are displaying higher interdependence in terms of supply chain collaboration, where the results are seen mainly in times of market turbulence and uncertainties, resulting in positive operational outcomes needed to avoid the disaggregation of GVCs [5]. Still, the two terms are often either used in tandem or in an interchangeable manner both in the scientific literature and in practice.

The literature has extensively treated the relationship between value chains and sustainability in terms of the open innovation performance of enterprises [6,7,8,9], innovation ecosystems [10], and regional smart specialization [11]; the adoption of Industry 4.0 [12,13] and its impact on resilience [14,15,16], particularly in the post-COVID-19 era [17]; digitalization [18,19]; and emphasizing how sustainable chain integration fosters firm green innovations [20,21].

From another perspective, the role of intercultural connections in GVCs is twofold, as besides the obvious gains of integrating competing values coming from developmental, group, and hierarchical cultural patterns [22], shared culture may result in learning traps and may limit the infusion of knowledge [23]. Finally, social capital is found to be an important vector of value chain sustainability [24].

The COVID-19 pandemic placed massive stress on value chain/supply chain resilience, leading to discontinuities of interactional, organizational, supply-side, and demand-side types [25]. The pandemic was also associated with long-term risks on the supply (production disruptions, factory closure, shortages of raw materials, etc.) or demand side (geographical concentration of suppliers, order cancellations by first tier customers and order cancellation propagation caused by the demand shocks in the customer market, etc.), or both (fluctuation of the exchange rate, transportation failure, container shortage) [26]. On a very general level, it forced companies to innovate [27,28].

On another note, it allowed new concepts to be developed such as the “survivability” of the supply chains, defined as the ability to stay alive in a temporary non-viable equilibrium during a large disruption [29], or “responsiveness” and “robustness” triggered by “flexibility” and “agility” [30,31], while employment, consumption, trade facilitation. and capital remained the main variables used to determine the impact of the pandemic on GVCs [32]. By emphasizing risk awareness, greater transparency, and agility, firms can build production networks that are not only more resilient to disruptions, but are also more competitive in the long-term [33]. Furthermore, the pandemic represented a window of opportunity for more sustainable and circular supply chains [34].

As shown by the recent study on the supply chain disruptions in the EU [35], finance support for new technologies and new production capacities, mapping of the supply chains, and support to R&D projects have been identified as concrete measures toward sustainable and resilient value chains.

Examples of concrete actions taken by clusters toward sustainable and resilient value chains can be given. ASTRICO North East, a traditional Romanian textiles cluster, has developed a collaborative marketing-sales system by setting up a trading company owned by members of the cluster. Through this system, the group of companies benefit from services such as marketing, design, prototyping, technical assistance, logistic and financial services, thus lengthening its own value chain https://www.astricone.eu/ (accessed on 3 September 2024) In order to overcome the lack of skilled workers, the Slovak Association of Photovoltaic and Renewable Energy Industry has started to facilitate several 1–2 day courses in the field of photovoltaic systems, photothermic systems, or both https://www.sapi.sk/ (accessed on 3 September 2024) In the agrifood sector, the Smart Agri Hubs Initiative https://www.smartagrihubs.eu/about (accessed on 3 September 2024) aims to increase the digitalization level of European agriculture by promoting a sustainable agro-food ecosystem.

Of particular interest for our endeavor is the “metrics” of value chains. The hierarchical relationship along the processes defining input/output flows inside a value chain is one of the proposed methodologies [36]. The United Nations Industrial Development Organization has identified four methods of measuring value chains: (a) the approach based on business management strategies, which looks at the value chain from the perspective of an individual company; (b) the approach based on industrial clusters, which states that spatial organization, strategic alliances between companies, and cooperation are sources of systemic competitiveness; (c) the global value chain (GVC) approach, which is a framework that analyzes how production processes are fragmented and distributed across different countries, emphasizing the economic returns and governance structures within these chains. This approach recognizes that value creation, capture, and distribution are influenced by the relationships between the different actors involved in the chain, particularly by the dominance of key players such as buyers or sellers, who operate on an international scale; and (d) the innovation systems approach, which considers that access to knowledge and technology enables individual actors to participate in global value chains [37].

Several factors affect the complexity of value chains including the number of levels, the quantity of production, and the number of time periods that can be considered. In this regard, a supply chain model that is gaining popularity is the Bayesian network model, a directed acrylic graph in which each node is associated with a random variable [38]. In the same line of thought, the new network data envelopment analysis model considers multiple stages in a supply chain and links across the stages [39]. Karmaker et. al identified the lack of enthusiasm in top management regarding sustainable practices as a major risk factor for supply chain resilience [40]. Yan et al. identified several risks for international supply chains: (a) information risk in terms of adverse consequences of information asymmetry and/or pollution in the process of sharing information; (b) logistics risk including logistics timelines, security, costs, customs clearance, and so on; (c) decision risk considering possibility and consequences of decisions not leading to the achievement of the desired purpose; (d) credit risk (i.e., the risk of breach of contract); (e) operational risk when due to complexity and variety of the external environment and limitation of cognitive ability, so the enterprise may fail in operation; and (f) technology risk such as technology shortage, technology development, technology precision, technology use, and technology acquisition and technology transfer [41]. There is a critical aspect of global economic integration: the extent to which a country participates in global value chains (GVCs) and international economic cycles directly impacts its dependence on foreign inputs, particularly in its domestic production processes. This foreign dependence is especially pronounced in high-tech industries, which rely heavily on international collaboration, specialized inputs, and advanced technologies [42]. In addition, the improvement of a country’s GVC participation can effectively drive the development of its own economy as well as have spillover effects on neighboring countries [43].

Of particular interest for our study is the “smile curve” concept, introduced by economist Gary Gereffi [44], which is a framework that illustrates how value is distributed along the different stages of a product’s life cycle, from the initial design and development phase through to production and the final stages of marketing, sales, and after-sales services. More recent studies have shown that value added growth is higher in regions specialized in scarce natural resources or scarce human skills in a GVC [45].

As a conclusion, crises like the recent COVID-19 or the war in Ukraine affect economic sustainability by inflicting disruptions into global/regional supply and value chains. To mitigate and repair the negative effects, quick responses are needed. In turn, actions must be based on data, ideally real-time data. While being reliable, official statistics data have the great disadvantage of being slow and outdated. Therefore, there is a need for a more agile measuring method for the economic performance of value chains. Industrial clusters have proven themselves to be both generators and drivers of innovative value chains and recognized as such at the international level both in theory and practice. The current paper proposes a simple cluster-based method of analyzing the economic structure and impact of value chains that was applied as a pilot in several cases.

2. Materials and Methods

As proven by both the literature and practice, “measuring value chains” is an overly complex process requiring elaborated data analytics, effort, and time. Unexpected events that have happened in recent years such as the COVID-19 pandemic or the war in Ukraine have led to sudden disruptions in global supply and value chains, whose effects need to be measured and mitigated in real-time modus.

Clusters. Smile Curve. Density.

Michael Porter’s concept of “clusters” is a key element in his theory of competitive advantage, particularly as outlined in his work “The Competitive Advantage of Nations.” Clusters refer to geographic concentrations of interconnected companies, specialized suppliers, service providers, firms in related industries, and associated institutions (such as universities, trade associations, and standards agencies) in a particular field. These entities co-locate in a region, benefiting from shared resources, knowledge spillovers, and synergies that enhance their productivity and innovation capacity [1].

Clusters are recognized as tools for economic resilience based on the twin transition, as resilience is most certainly a property of the network [46], and it has been proven that rapid response clusters were able to provide to value chains disruptions during the COVID-19 pandemic. Between March and October 2020, the European Clusters Alliance, the representative body of clusters and cluster networks at European level, organized over one hundred webinars to analyze the implications and help industrial clusters with post-pandemic regeneration [47].

Furthermore, on behalf of DG GROW, in 2022, the European Cluster Collaboration Platform issued two survey-based reports on the disruptions in value and supply chains at the European level along the 14 industrial ecosystems identified by the European Commission, the clusters being credited with finding solutions for these disruptions and developing roadmaps for strengthening the European economy [35,48].

Given the above, this study proposes a cluster-based value chain analytical methodology, the main indicator of which is the density of the selected indicator along each link of the value chain.

The density is calculated in percentage as the ratio between the absolute value of the indicator on the link i and the total value of the indicator at the cluster level.

$$Density\left(\%\right)={\displaystyle \frac{{value}_i}{TotalValue}}\times 100$$

(1)

The taxonomy of value chain links (Research-Development-Innovation, Branding, Product Design, Suppliers, Production, Distribution, Marketing-Sales, and Maintenance) followed in the current proposed methodology was anchored in the Porterian model [1], which is used by the OECD [49] and Gereffi [44] to define the “smile curve” of the value chain, explaining the distribution of added value along the chain.

In the proposed methodology, the analysis of the results is based on the “smile curve” approach, which states that the highest value-added activities are placed in the pre- and post-production stages of the value chain. This allows for a direct correlation between the density of the indicator on a specific link and the level of added value at the specific stage of the value chain. Thus, value chains/regions/countries showing higher densities of companies in RDI, Branding, Product Design or Marketing/Sales, and Maintenance find themselves in a better competitive position than the ones concentrating the companies in Production, where the added value is the lowest, according to the proposed “smile curve” model.

In conclusion, the proposed methodology links the Porterian clusters and Gereffi’s smile curve by describing the distribution of added value along the links of a value chain via a proposed indicator, the density. To give an example, Figure 1 shows the distribution of company density in selected cluster-based value chains (Energy-Renewables, Agrifood, Electronics, Digital, Furniture, Creative and Cultural, Textiles, Construction, and Health) in Romania. In this case, the concentration of companies on the Production link in almost all sectors highlights a comparative disadvantage and eventually a competitiveness threat.

Data for the analysis were provided by the cluster management team. This was carried out via a dedicated questionnaire including a table requiring inputs for the analyzed indicators.

Respondents were instructed that:

-

The overall value of each indicator (Total) represents the aggregated value at cluster level for the respective indicator;

-

In the case of the indicator related to “number of companies”, if a company acts on more links—as is usually the case—it should be allocated to the link where the company concentrates its highest activity to avoid duplication.

3. Results

The methodology was used in the elaboration of the Industrial Strategy of Romania 2023–2027 [50], based on 29 clusters, representing 1371 companies, an EUR 6 billion turnover, and EUR 800 million of exports, leading to some relevant results. Considering as indictors the “number of enterprises”, “turnover”, and “exports” in the value chain links of Romanian clusters, the analysis showed the highest density of each indicator on the Production link, which represents a competitive disadvantage as it bears the lowest added value according to the “smile curve” model. A notable exception was the Creative and Cultural ecosystem, which, at least at the cluster level, focused 38% of enterprises on the Product Design link (compared to only 27% Production, generated 43% of turnover on the RDI link, 34% on the Product Design link (compared to only 18% Production), and not less than 50% exports on the Product Design link. These confirm the empirical observations and lead to the need for much more intense support and structure of this sector at the level of economic policy. Data were aggregated at the level of 29 Romanian clusters, as shown in

Table 1. Density of companies across Romanian cluster-based value chains [50]. A darker shade corresponds to a higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Energy-Renewables		20.73%	2.44%	35.37%	12.20%	14.63%	4.88%	6.10%	3.66%
Agri Food		5.50%	5.50%	8.26%	10.09%	54.13%	5.05%	9.17%	2.29%
Electronics		20.81%	7.51%	10.98%	9.25%	23.12%	6.36%	10.98%	10.98%
Digital		6.15%	4.47%	16.76%	5.03%	44.13%	4.47%	9.50%	9.50%
Furniture		10.40%	1.60%	3.20%	16.00%	67.20%	1.60%	0.00%	0.00%
Creative and Cultural		11.54%	7.69%	38.46%	3.85%	26.92%	0.00%	11.54%	0.00%
Textiles		3.85%	2.56%	1.28%	7.69%	76.92%	2.56%	2.56%	2.56%
Construction		2.04%	2.04%	16.33%	16.33%	57.14%	6.12%	0.00%	0.00%
Health		5.88%	0.00%	0.00%	11.76%	79.41%	0.00%	2.94%	0.00%

,

Table 2. Density of the indicator “RDI expenditure” across Romanian cluster-based value chains [50]. A darker shade corresponds to a higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Energy-Renewables		7.42%	0.00%	0.29%	25.94%	49.14%	6.22%	0.33%	10.66%
Agri Food		1.51%	0.06%	0.56%	1.45%	80.61%	2.02%	13.31%	0.49%
Electronics		3.43%	0.03%	0.04%	10.79%	81.95%	2.14%	0.24%	1.38%
Digital		12.74%	2.48%	24.34%	0.03%	45.34%	10.43%	2.34%	2.30%
Furniture		0.66%	0.31%	0.01%	3.94%	95.05%	0.02%	0.00%	0.00%
Creative & Cultural		43.35%	0.17%	33.75%	0.18%	18.36%	0.00%	4.19%	0.00%
Textiles		4.93%	1.23%	0.42%	2.08%	86.77%	0.00%	3.88%	0.70%
Construction		1.91%	0.08%	70.96%	10.44%	16.30%	0.30%	0.00%	0.00%
Health		0.88%	0.00%	0.00%	8.73%	89.50%	0.00%	0.89%	0.00%

and

Table 3. Density of the indicator “Exports” across Romanian cluster-based value chains [50]. A darker shade corresponds to a higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Energy-Renewables		47.50%	0.00%	25.00%	0.00%	27.50%	0.00%	0.00%	0.00%
Agri Food		0.00%	0.00%	50.00%	0.00%	50.00%	0.00%	0.00%	0.00%
Electronics		6.82%	3.65%	5.87%	4.25%	69.60%	2.72%	4.50%	2.58%
Digital		30.75%	6.13%	28.69%	0.05%	28.62%	0.01%	5.64%	0.10%
Furniture		0.90%	0.82%	0.72%	5.33%	92.20%	0.03%	0.00%	0.00%
Creative & Cultural		10.00%	0.00%	50.00%	0.00%	40.00%	0.00%	0.00%	0.00%
Textiles		3.14%	0.92%	0.18%	0.00%	95.58%	0.00%	0.00%	0.18%
Construction		0.00%	0.00%	0.00%	5.44%	94.21%	0.36%	0.00%	0.00%
Health		5.08%	0.00%	0.00%	0.00%	94.92%	0.00%	0.00%	0.00%

.

Moving further, the methodology allows for comparisons between the relative competitive advantages of clusters/regions/countries on specific links of the value chain. Between 2020 and 2022, in the frame of the Danube Peer Chains Project run in the Danube Transnational Program, the methodology was used to map three value chains in the Danube region: wood and furniture, mechanical engineering, and mechatronics [51]. This was based on data collected from twenty-two clusters from across the region, gathering 3319 companies.

Density values were gathered in quantiles (0–20%), (21–40%), (41–60%), (61–80%), (81–100%). The region scoring in the highest density quantile showed a competitive advantage on the specific link and vice versa.

The analysis of the wood and furniture industry highlights several relative strengths and weaknesses across different regions, as shown in

Table 4. Density (%) of companies in the wood and furniture value chain in the Danube region. Elaborated based on [51]. A darker shade corresponds to a higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Austria		6.72%	5.04%	7.98%	23.53%	47.48%	0.42%	0.84%	7.98%
Hungary		6.56%	0.00%	23.77%	12.30%	24.59%	20.49%	8.20%	4.10%
Montenegro		0.00%	0.00%	20.00%	20.00%	50.00%	0.00%	10.00%	0.00%
Romania		4.27%	0.85%	6.84%	11.11%	64.10%	4.27%	3.42%	5.13%
Slovenia		18.67%	0.00%	0.00%	0.00%	81.33%	0.00%	0.00%	0.00%
Croatia		6.90%	5.17%	8.62%	12.07%	51.72%	10.34%	0.00%	5.17%
Total		7.58%	2.58%	10.16%	15.00%	50.65%	5.97%	2.74%	5.32%

. With relation to the weak points, Branding represented the main issue with an extremely low presence across the industry (only 1.85% of companies) and was completely absent in Austria, Bavaria, Hungary, and Slovenia. In addition, only 4% of the companies were present on the Marketing/Sales link, missing completely from Austria and Slovenia. Only 6.83% of the companies were active in Outbound logistics, with no recorded activity in Slovenia. Finally, Inbound Logistics involved only 11.81% of the companies and was absent in Slovenia. Turning to the relative strengths, Research, Development, and Innovation (RDI) was strong in Slovenia, with 29% of companies engaged in this area; Product Development saw a high participation in Austria (24.76%) and Bavaria (24.59%); Manufacturing showed significant strength in Bavaria (45.90%), Baden-Württemberg (49%), and Slovenia (41.67%), while companies active on the Service link of the chain were particularly strong in Hungary, with 23.95% of companies involved in service-related activities.

Overall, the analysis suggests that while some regions showed strong specialization in certain aspects of the value chain, there were notable gaps, especially in branding and logistics, which could be addressed to improve the overall industry competitiveness in the wood and furniture sector.

Similarly, in the mechanical engineering sector, the analysis revealed distinct relative strengths and weaknesses across various regions, as shown in

Table 5. Density (%) of companies in mechanical engineering value chain in the Danube region. Elaborated based on [51]. A darker shade corresponds to higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Austria		5.22%	0.00%	12.17%	24.35%	46.09%	4.35%	0.00%	7.83%
Baden Wuerttemberg		4.74%	4.74%	9.48%	33.18%	37.91%	2.84%	4.74%	2.37%
Hungary		14.93%	0.00%	0.00%	0.00%	35.82%	5.97%	7.46%	35.82%
Romania		24.06%	5.26%	15.04%	0.00%	19.55%	0.75%	18.80%	16.54%
Serbia		8.00%	0.00%	9.14%	6.29%	38.29%	2.29%	5.71%	30.29%
Slovenia		17.69%	3.08%	7.69%	13.85%	36.92%	4.62%	4.62%	11.54%
Total		8.06%	3.64%	9.55%	24.30%	36.90%	2.99%	5.73%	8.84%

. Branding was the main weak point, with an extremely low involvement of companies engaged (only 3.80%) and a complete absence in Austria, Hungary, and Serbia. The Marketing/Sales link was also weak, with limited presence (only 5.96% of companies active) and no activity in Austria. Product Development involved only 9.65% of companies and was missing entirely in Hungary. Relative strengths were found in Research, Development, and Innovation (RDI), which was notably strong in Romania, with 24.06% of companies engaged in RDI activities. Inbound Logistics had high involvement in Austria (24.34%) and Baden-Württemberg (33.17%), Manufacturing showed significant strength in Austria (46.08%) and Slovenia (36.92%), and Service had a strong presence in Hungary (35.82%) and Serbia (38.46%).

Certain regions have developed considerable expertise in specific areas such as RDI, manufacturing, and services, but there are notable deficiencies in branding, marketing, and product development, which could potentially be addressed to increase the overall performance and competitiveness of the mechanical engineering sector in these regions.

Finally, in the mechatronics sector, the situation looks as follows. Branding, again, showed a particular weakness, with only 2.58% of companies participating, and was completely missing in Hungary, Montenegro, and Slovenia. The Marketing/Sales link had a limited presence, with only 2.74% of companies involved and no activity in Slovenia and Croatia. Unlike the other two cluster-based analyzed value chains, Service and Research, Development, and Innovation (RDI) showed relative weaknesses with 5.32% of companies engaged on the Service link and missing entirely in Montenegro and Slovenia, while RDI saw a limited 7.65% of company involvement and was absent from Montenegro. Strong points could be found in Hungary, with 23.77% of the companies acting in Product Development and 20.49% in Outbound Logistics. Austria had a significant 23.52% of participation in Inbound Logistics, and Slovenia showed a particularly robust participation in Manufacturing (81.33% of the companies), as shown in

Table 6. Density of companies in the mechatronics value chain in the Danube region. Elaborated based on [51]. A darker shade corresponds to higher value of the indicator.

	Link	RDI	Branding	Product Design	Inbound Logistics (Suppliers)	Production	Outbound Logistics	Marketing	Maintenance
Industrial Ecosystem
Austria		12.26%	0.00%	24.76%	14.62%	27.12%	4.72%	0.00%	16.51%
Bavaria		11.48%	0.00%	24.59%	3.28%	45.90%	3.28%	4.92%	6.56%
Baden Wuerttemberg		1.00%	5.00%	10.00%	10.00%	49.00%	10.00%	5.00%	10.00%
Hungary		9.38%	0.00%	13.54%	15.63%	23.96%	4.17%	9.38%	23.96%
Slovenia		29.17%	0.00%	12.50%	0.00%	41.67%	0.00%	0.00%	16.67%
Romania		15.13%	3.36%	13.45%	10.08%	20.17%	11.76%	11.76%	14.29%
Total		9.38%	1.86%	17.77%	11.82%	33.89%	6.84%	4.00%	14.45%

.

This suggests that while the mechatronics sector shows significant strengths in specific areas such as product development, inbound logistics, manufacturing, and outbound logistics in certain regions, there are substantial weaknesses in branding, marketing, service, and RDI, especially in regions like Slovenia and Montenegro. Addressing these gaps could enhance the overall performance and competitiveness of the sector.

These findings led to relevant conclusions in terms of integrating companies across cross-border, interregional value chains. In the selected cases, the results showed (a) complementarity of company profiles in the Danube region (strength), showing higher cooperation potential (opportunity); (b) the East–West technology divide, where companies in the Western part are concentrated on links with higher added value (weakness), which in the long run will bring higher levels of economic polarization (threat); and (c) the low number of companies acting on Branding and Marketing along the analyzed cluster-based value chains, which hints toward a call for action at the economic policy level.

It can be noted that in both cases, the conclusions of the cluster-based value chain analysis were double checked and complemented by classical methods such as input–output tables in the case of the Romanian Industrial Strategy [50], or the value of the exports/imports of the analyzed regions and sectors in the case of the Danube analysis [51].

4. Discussion

While the role of clusters in defining and driving international value chains has recently gained in recognition and importance, in the aftermath of the COVID-19 pandemic, the actual “measuring” of the economic impact of value chains by means of clusters has not been approached yet. In this sense, the current proposed methodology represents a novelty.

Linking cluster economic performance with the sustainability and resilience of value chains through the density of the analyzed indicator(s) (turnover, exports, number of companies, RDI expenditures, labor force) on each link of the chain represents the main theoretical contribution of the current study.

From a theoretical perspective, the methodology follows the same line of thought, among others such as Porter [1], UNIDO [37], and Gereffi et. al. [44] and adds on in the sense of structuring the cluster-based value chain approach. It complements similar recent studies on value chain metrics using indicators such as resistive-, absorb-, endure-, and recovery duration; the depth of impact and cumulative absorptive impact, critical threshold, residual capacity; residual-, restored-, and cumulative recovery performance; failure ratio and rate; and the recovery ratio and recover rate [52] were used to assess resilience. Regarding innovation in value chains, recently used indicators address the following topics: business sustainability and responsibility, disruptive innovation, sustainable business strategies, eco-innovation, and business dynamics. In addition, the relative majority of indicators are sector specific, having been developed to address sustainable food systems [53]. On the other hand, based on the classical “input/output tables” methodology, Takayabu used productivity as an indicator to assess the performance of 18 manufacturing sectors in 43 countries [54].

Practical results were confirmed by the classical approach of “measuring” the value chains used, for example, by the OECD [49] or DG Grow [35,48], as it is meant to be a tool for practitioners in the first place.

By applying this method, cluster managers are able to better assess the status-quo of their membership numbers in terms of “missing” or, in contrast, “overcrowded” chain links (1), and the distribution of added value, innovation, and exports (2), giving them a clear indication of where they should intervene by means of different support actions like innovation audits, export promotion activities, qualification measures for the labor force, etc. Having arrived at this point, it is worth mentioning that the methodology could be extended to any competitiveness indicator that the cluster manager might see fit such as the number of employees, IPR results, etc.

On the next level, regional and national policymakers can use the results in better shaping industrial and innovation strategies and actions, allowing them to quickly react to disruptions, as the main advantage of the cluster-based method compared to the classical (input–output) ones lies in its rapidness: asking cluster managers to provide data is much faster than consulting relatively outdated statistical data, which is also more effective: in the case of the Romanian industrial Strategy [50], 29 “interviewed” cluster managers provided data for 3771 enterprises (a ratio of 1:130), while in the case of the Danube analysis [51], 22 cluster managers accounted for 3319 companies (a ratio of 1:150). If we compare the 500 responses received by the direct survey questionnaire on value chain disruptions addressed to companies by the European Cluster Collaboration Platform in 2022 [35,48], we can obtain an image on the effectiveness of the proposed method.

This brings us to the main conditions for the validity of the method. First, there must be a critical mass of clusters/companies to be interviewed. It was beyond the scope of the current paper to debate on the exact numbers, but it is obvious that if more stakeholders are being interviewed, the greater data gains in relevance.

Second, it is “the real-time” aspect that makes the proposed method attractive. In the case of sudden disturbing events, reactions must be quick.

The above presented aspects also represent the limitations of the current research. While we have argued on the representativeness of the results, we must state that they were obtained by a top–down “pull” effort: cluster managers “passively” provided the requested data when asked by public bodies or industrial associations.

This leads us to future research directions for the development of the proposed method, which could be embedded into an AI empowered cluster cooperation platform aimed at enhancing international cooperation between cluster members, allowing for an exchange of best practices and fostering individual and institutional learning processes, thus strengthening the resilience of international value chains. It will also turn cluster managers into active users, “pushing” bottom up for change in economic policies based on the data they would provide. The idea has already started to motivate the scientific community, as shown by a recent study analyzing the way B2B platforms can turn value chains into value networks [55].

5. Conclusions

The proposed cluster-based value chain analysis method provides a much more rapid alternative to the classical input/output measurement methods. It was applied in the development of the New Industrial Strategy of Romania and in evaluating three value chains in the Danube region (mechatronics, wood and furniture, mechanical engineering). The results have allowed for relevant conclusions regarding the competitiveness of the analyzed sectors and have been translated into policy measures. As it is easy to apply, relies on the cluster managers as the interface to member companies, and uses the density of the selected indicator as the main measurement unit, the method requires a critical mass of “interviewed” stakeholders and a quick response from the clusters. The use of AI tools and embedment into a larger transnational cooperation platform, especially its functionalities, needs further research.