In his report ‘Learning regimes and capacity formation in an advanced peripheral economy: Romania’s chemical industry’, sociologist and researcher Norbert Petrovici plays the role of economic geographer, exploring the stalled potential of Romania's chemical industry, drawing on extensive research and interviews with key players from legacy plants to multinational outposts. In this interview, he speaks about the sector's critical yet invisible role, the "residual knowledge" that persists from its socialist past, and how its future is inextricably linked to the green transition and strategic state policy, arguing that its true value lies not in nostalgia but in its unrealized capacity for foundational innovation.
Chemical Bonds: Your research moves from Romania's automotive and textile industries to its chemical sector. What drew you to this transition, and why is the chemical industry a critical subject of study now?
Norbert Petrovici: The chemical industry was absolutely central during socialism, a technological linchpin that enabled other sectors. Today, it’s economically significant yet largely overlooked—*visibly invisible*. It’s structurally stuck, grappling with its socialist legacy while being crucial for the future. The green transition, the shift from a petro-based economy, and current geopolitical energy tensions are impossible to imagine without chemistry. It’s a core sector. Yet, unlike automotive, which transitioned successfully into global value chains with technological upgrading, chemistry has not. I wanted to research why it failed to find its footing.
CB: For our readers who may be unfamiliar, what does Romania’s chemical industry look like today?
NP: Its structure reveals its history. Under socialism, it was predominantly B2B, providing raw materials for other industries. Today, it still retains that B2B core but has shifted slightly from basic chemicals to more technical intermediate products. However, it rarely reaches the final consumer good. This is highlighted by a massive trade imbalance: we import twice as many chemical products as we export. The Romanian market is flooded with imported cosmetics, plastics, and intermediates needed for our own automotive and textile sectors. The sector has capacity, but it’s not positioned to cover internal demand, let alone drive innovation.
CB: Let’s go back to that socialist inheritance. How was the industry structured, and what kind of expertise was built into it?
NP: It was built on vertical integration—controlling everything from mineral extraction to the final product. This required a specific, hierarchical corporate structure. But the key difference was the intense need for technology and learning. Chemical processes couldn’t be easily experimented with; patents were controlled by Western multinationals. Furthermore, many products had dual uses (civilian and military), meaning critical knowledge, like specific catalysts or mixing formulas, was often locked away by Western powers.
This forced the state to develop a dense, robust national innovation system. There was a vast network of about 25 national research institutes, plus laboratories within every factory, all dedicated to reverse-engineering Western patents and adapting technologies to local inputs. When a licensed factory was installed, often by German companies like Siemens or Zimmer, the impulse was to train local engineers to fine-tune and operate it independently. Factories even developed their own mechanical workshops to produce spare parts, creating a deep, practical, organizational knowledge.
CB: What happened to this extensive infrastructure after 1989?
NP: It experienced a severe institutional collapse. The national institutes limped along until around 2002 but without investment. When European grants arrived, they prioritized knowledge production for publication, not innovation for deployment. The critical link between the research institute and the factory floor was severed. The on-site factory labs were decimated; where 200 engineers once worked, you might find five today.
This collapse created a new dependency. Today, if a local company develops a new process, it must often go to Germany for expensive verification and licensing because the national capacity for this is gone. To truly innovate and scale, they need to partner with a multinational that has the capital and institutional power to handle global IP licensing. The system that once enabled autonomous adaptation now funnels innovation back to multinational headquarters.
CB: You use the powerful concept of "residual knowledge." Can you give an example of a plant still running on this inherited expertise?
NP: Certainly. The most successful residual knowledge lies in the mechanical factories that were within the plants. These teams, skilled in operating, repairing, and adapting machinery, were sometimes able to spin off and become players in engineering design and installing factories for multinationals, leveraging their deep understanding of local adaptation.
Another form is the knowledge of vertical integration. Managers and engineers who understood the precise sequence of inputs needed for complex production tried to replicate these business verticals. Companies like Chimcomplex or those in the food sector have attempted this, with varying success. This inherited expertise is a form of organizational memory that is both a residual of the past and a precarious resource for the present.
CB: Your work describes Romania’s re-integration into global markets as a form of dependency. Can you explain how this "lock-in" works?
NP: The key difference lies in where research and development (R&D) and intellectual property (IP) are controlled. In automotive, R&D can be decentralized geographically. In chemicals, it is highly concentrated at multinational headquarters because it is incredibly expensive. This means Romania is often integrated only through low-autonomy roles: production outposts that follow strict standard operating procedures or packaging hubs.
There are exceptions, like in generic pharmaceuticals where companies repurpose expired patents. But even this is a form of dependency. We export generics, but we import twice as much in higher-value, specialized medical products. The learning that happens in these integrated roles is about cost reduction, particularly energy input, not about groundbreaking innovation. The high-value, IP-generating work remains elsewhere.
CB: You’ve developed a typology of five "learning regimes" in these companies. Without getting bogged down in jargon, what does this typology reveal?
NP: It reveals a spectrum of autonomy and innovation. On one end, you have firms that are mere executors—production outposts or packaging hubs with no real autonomy or IP. On the other, you have a few residual legacy firms and generic pharma companies that still attempt some domestic R&D, trying to innovate around energy efficiency or process adaptation.
Our initial typology was too rigid. Through ongoing fieldwork, I’ve found surprising learning even in unexpected places. For instance, a company classified as a mere sales distributor might be innovating in marketing, developing IT tools to teach clients how to use chemicals, or even inventing new compounds through consultancy. The landscape is more fluid, but the central problem remains: the learning is often insular, fragile, and reliant on individual experts. If one key engineer leaves or passes away, an entire knowledge domain can be lost because it’s stored in people, not in institutionalized, replicable processes.
CB: This feels like a deeply human story. Beyond structures and systems, what is the human cost of this structural stagnation?
NP: Human geography is different from automotive. Chemical plants are often massive, contaminated sites in city centers. They are no longer mass employers but have become real estate assets. Their future is now part of a political economy of urban development, contamination cleanup, and deliberative democracy—a form of industrial heritage that is deeply entangled with strategies of urban planning.
Yet, within the companies that remain, I was struck by the passion. I spoke with doctors and engineers with a real pioneer, almost "cowboy" mentality—a desperate, energetic drive to be the first, to try and catch up, to survive against immense barriers. There is a palpable will to innovate.
CB: If you could fix one thing within the triple bond of policy, industry, and research, what would unlock the potential you see in these people?
NP: The state is the most important and currently absent actor. The market alone cannot de-risk knowledge production in this sector. The state must act as a catalyst to re-establish those linkages. The global economy still operates on verticals, and for chemicals, relying on imported inputs makes products uncompetitive. For the green transition—which is utterly dependent on chemical innovations for everything from solar panels to semiconductors—having a nationally bounded production network isn’t a socialist relic; it’s an economic necessity. The state must create conditionality in aid, foster ties between cutting-edge European initiatives and local industry, and strategically support the integration across business verticals. Without this catalytic role, the immense energy and passion of those on the ground will remain trapped, and the industry’s potential as a catalyst for a new future will remain unfulfilled.
