Finance

Industrial Policy in Data: Why Governments Are Back in the Driver's Seat

For a couple of decades, the dominant narrative around the digital economy was one of laissez-faire. Let the market innovate, let platforms scale, let data flow freely across borders. The role of government, many argued, was to get out of the way. That era is decisively over. We are now witnessing the full-throated return of industrial policy, but this time its primary theatre of operations isn't steel, cars, or semiconductors—it's data. Governments worldwide are no longer passive regulators; they are active architects, using a new toolkit of subsidies, regulations, and strategic investments to shape who controls data, where it is processed, and what value can be extracted from it. This isn't a minor regulatory shift; it's a fundamental reordering of the rules of economic competition in the 21st century.

What You'll Learn

Why Industrial Policy Is Back (For Data)

The comeback isn't random. It's driven by a convergence of fears and opportunities that policymakers can no longer ignore.

Geopolitical Anxiety and Data Sovereignty: The realization that economic and national security are now inextricably linked to data flows. If all your citizens' health data, corporate R&D, and financial transactions are processed on servers controlled by a foreign tech giant or a geopolitical rival, you have a vulnerability. This fear of digital dependence fuels the push for data sovereignty—the idea that nations must exert control over data generated within their borders. It's less about protectionism for its own sake and more about risk management in an unstable world.

The AI Arms Race: Artificial intelligence is the new “space race,” and data is its rocket fuel. Governments understand that leadership in AI, and therefore in future industries from biotech to autonomous systems, depends on access to vast, high-quality datasets and the compute power to process them. Letting private actors hoard this strategic asset without oversight is seen as a recipe for falling behind. Industrial policy aims to ensure domestic players have the data and infrastructure to compete.

A report by the World Economic Forum argues that data is now a factor of production on par with capital and labor, justifying state intervention to ensure its equitable and strategic use. It's not just an asset; it's infrastructure.

Market Concentration and the Platform Problem: The winner-take-all dynamics of digital platforms have created unprecedented concentrations of economic power and data. This isn't just an antitrust issue; it's an innovation issue. Policymakers worry that a handful of gatekeepers can stifle competition by controlling access to essential data. New industrial policies, like the EU's Data Act, are explicitly designed to break open these “data silos” and create a more level playing field for smaller firms and startups.

The New Toolkit of Data Industrial Policy

Forget the old image of industrial policy as just tariffs and direct subsidies to factories. The modern version for data is more sophisticated, blending regulation, investment, and soft power.

Regulation as a Strategic Tool

This is the most visible lever. Laws like the GDPR weren't just about privacy; they were a strategic move to set the global standard for data handling, forcing foreign companies to adapt to European rules. The new wave goes further:

  • Data Localization Mandates: Requiring certain types of data (e.g., financial, health, government) to be stored and processed within national borders. Russia, China, and India have variants of this.
  • Data Sharing Mandates: Forcing large companies, especially in sectors like IoT (Internet of Things), to share data with users and competitors. The EU's Data Act is a textbook example, aiming to unlock industrial data.
  • Interoperability and Portability Rules: Making it easier for users to switch services and take their data with them, reducing platform lock-in.

Direct Investment in Digital Infrastructure

Governments are becoming venture capitalists and infrastructure builders for the data age.

The US CHIPS and Science Act isn't just about semiconductors; it allocates billions for regional innovation hubs focused on key technologies, many of them data-intensive. The European Commission is funding the creation of common European data spaces in sectors like manufacturing, energy, and healthcare—essentially building sanctioned, sovereign data marketplaces. China's “East Data, West Computing” project is a massive state-driven initiative to build data center clusters in its less-developed western regions, balancing economic development with the compute needs of its eastern tech hubs.

The Big Shift: The goal is no longer just to regulate the private sector's use of data, but to actively create public or semi-public data assets and infrastructure that serve national economic objectives. It's proactive, not reactive.

Standards and Certification

Who gets to define what “trustworthy AI” or “green data center” means? Governments are increasingly stepping in to set these standards, creating a non-tariff barrier that favors domestic companies who design to the local spec first. It's a softer, but potent, form of industrial steering.

Real-World Cases: From Brussels to Beijing

Let's look at two contrasting models that define the poles of this new landscape.

The EU: The Regulatory Architect

The European Union's approach is fundamentally rule-based and defensive. Its strength is setting the regulatory “Brussels Effect,” where its rules become the global default. The GDPR was the first salvo. The Data Act and Data Governance Act are the next phase, explicitly designed to build a single market for data. The goal isn't to create European Google, but to ensure that any company, European or not, plays by rules that (in theory) foster SME innovation and citizen rights. The industrial policy element is clear: by mandating data sharing from large tech and industrial firms, the EU hopes to fuel a wave of European AI and SaaS startups that can thrive on now-accessible data.

China: The Systemic Integrator

China's approach is holistic and offensive. Industrial policy for data is seamlessly integrated into its broader national technology and security strategies. Data is treated as a national strategic resource. Through its “Digital China” blueprint, the state coordinates investments in 5G, data centers, and industrial internet platforms. It uses antitrust actions to discipline its own tech giants, redirecting their energies and data toward state-prioritized areas like semiconductor R&D and industrial AI. The “East Data, West Computing” project exemplifies this: it's a massive, state-planned geographical redistribution of computing power to support national goals in energy efficiency and regional development.

The contrast is stark: Europe builds fences to shape the behavior inside the field. China is redesigning the field itself.

How Businesses Should Adapt and Thrive

For company leaders, this new environment is less about compliance and more about strategic repositioning. Here’s a pragmatic approach.

1. Map Your Data Footprint Against Policy Maps: You can't manage what you don't measure. Create a detailed inventory of what data you collect, where it is stored and processed, and its classification (personal, industrial, sensitive). Overlay this with the regulatory requirements of every jurisdiction you operate in. This isn't a one-time audit; it needs to be a dynamic process. I've seen mid-sized manufacturers get blindsided because they didn't realize their IoT sensor data from German factories would fall under the EU's new data-sharing rules.

2. Rethink “Data Sovereignty” as an Operational Model, Not a Burden: Instead of fighting localization requirements, explore how they can be turned into an advantage. Can you build closer, more trusted relationships with local clients by using in-country data centers? Can you partner with a local cloud provider who is favored by government contracts? In some markets, “sovereign cloud” offerings are becoming a major selling point.

3. Engage in Standard-Setting: Don't wait for rules to be finalized. Participate in industry consortia and government consultations. The shape of future regulations around data interoperability, AI ethics, or carbon accounting for data centers is being decided now. Having a seat at that table is invaluable.

4. Diversify Your Data and Tech Stack: Over-reliance on a single cloud provider or a single geographic region for data processing is now a strategic risk, not just an IT one. Develop a multi-cloud or hybrid strategy. Consider edge computing to process data closer to its source, which can simultaneously address latency needs and data sovereignty concerns.

Common Mistakes and Advanced Strategies

After a decade advising firms on this transition, I see the same errors repeatedly.

The Compliance Silo Trap: The biggest mistake is handing this entire issue to the legal and compliance team and checking a box. Data industrial policy is a core business strategy issue, touching R&D, IT, product development, and market entry. Legal informs the constraints, but strategy decides the path. A compliance-led approach will only ever minimize cost; a strategy-led approach can find new revenue.

Misreading the Motive: Assuming these policies are just disguised protectionism is a shallow read. While that element exists, many policymakers are genuinely trying to solve real market failures—the lack of data mobility, the stifling of competition, the under-investment in shared digital infrastructure. Businesses that frame their engagement around solving these shared problems (e.g., “Here’s how we can help create a secure data-sharing ecosystem for the automotive supply chain”) build more productive relationships than those who just lobby against rules.

An advanced strategy is to voluntarily adopt the strictest data governance standards you see emerging (like those in the EU) as a global baseline. It becomes a competitive moat and future-proofs your operations, turning regulatory foresight into a trust signal for customers worldwide.

The Static Analysis Error: Treating policies like the Data Act as a fixed set of rules is a mistake. They are the opening move in a long game. The smart play is to model the second-order effects. If industrial data becomes more accessible, what new services can we build? If data localisation increases in Southeast Asia, should we pre-emptively partner with a local data center firm? Think in scenarios, not just checklists.

FAQ: Data Industrial Policy in Practice

Doesn't data industrial policy stifle innovation by adding bureaucracy and friction?
It can, if poorly designed. The classic fear is that red tape kills startups. But the counter-argument from Brussels and elsewhere is that the current situation stifles innovation by allowing a few giants to lock up essential data. The policy aim is to change the type of innovation that flourishes—from innovation within walled gardens to innovation that leverages open, shared data ecosystems. The net effect on innovation is contested and will depend entirely on how these policies are implemented on the ground.
We're a small SaaS company. How can we possibly navigate these complex, national-level strategies?
For SMEs, the key is focus and leverage. Don't try to become an expert on every global policy. Deeply understand the rules in your primary market and one or two key growth markets. Then, leverage industry associations and cloud service providers. AWS, Google Cloud, and Microsoft Azure have entire teams dedicated to helping customers understand regulatory landscapes. Use them. Their compliance blueprints and in-region infrastructure can offload a huge part of the operational burden, letting you focus on your product.
How do we balance data sovereignty requirements with the need for global data analysis to improve our products?
This is the core technical challenge. The solution increasingly lies in privacy-enhancing technologies (PETs) and federated learning. Instead of moving raw data to a central server, you can move the analysis to the data. Techniques like federated learning allow you to train AI models on data that never leaves its local jurisdiction. Similarly, synthetic data generation can create statistically accurate but non-real datasets for broader analysis. Invest in these technologies now; they are moving from academic research to commercial necessity.
Is the ultimate goal of these policies to create national “data champions” that can rival the US and Chinese tech giants?
For some countries, yes, that's an explicit hope (see France and Germany's support for GAIA-X, a European cloud initiative). For others, like the EU, the goal is more nuanced: to ensure that their industrial base (automotive, pharmaceuticals, machinery) remains competitive by securing access to the data they need, regardless of whether the tech platform is European or American. It's less about creating a European Facebook and more about preventing a non-European company from becoming the indispensable gatekeeper for European factory data.
What's the next frontier for data industrial policy that most businesses aren't seeing yet?
The convergence of data policy with climate and energy policy. The carbon footprint of data centers and AI compute is becoming a major political issue. We're already seeing policies in Singapore and Ireland that link data center expansion to energy efficiency and renewable energy use. The next wave of “green data” industrial policy will likely involve subsidies for data centers that use advanced cooling, locate near green energy sources, or provide waste heat to district heating systems. Your data strategy will soon need a full environmental, social, and governance (ESG) audit.

The return of industrial policy in the data realm is not a temporary blip. It is a structural response to data's ascendance as the critical resource of our time. For businesses, this means the playbook written in the 2010s—move fast, break things, scale globally with a uniform model—is obsolete. The new playbook requires geopolitical acuity, regulatory engagement, and technological flexibility. The winners will be those who stop seeing government as an obstacle and start seeing it as a complex, powerful, and permanent player in the data game—one whose rules you must not only follow but learn to anticipate and even shape.

Comments

0
Comments are moderated