Here is what progress looks like in Binh Duong, Vietnam, in 2025. A 44,000-square-meter factory under construction, designed to LEED Gold standards, with 50,000 solar panels generating power from a rooftop the size of 70 football fields, a rainwater harvesting system, a sewage treatment facility, and a commitment to run on 100% renewable electricity. When it opens, it will be the LEGO Group's first carbon-neutral factory anywhere in the world.

It is, by any measure, an impressive achievement.

Now look at the photograph from 500 meters above. The factory sits inside the VSIP Binh Duong industrial zone, one of 14 Vietnam Singapore Industrial Parks spread across Vietnam. The LEGO factory is surrounded by manufacturers who do not share its energy system, whose waste streams connect to nothing, whose material flows terminate at the loading dock and disappear into the regional infrastructure. The zone hosts electronics assembly, consumer goods production, and logistics. It is a collection of optimized objects. It is not a system.

This distinction matters enormously. And Vietnam is in the middle of building 410 of them.

In the 1960s and 1970s, Britain, Germany, and Japan were winning the post-war industrial competition with spectacular environmental costs. Rivers caught fire. Cities became uninhabitable. Respiratory disease spiked near industrial corridors. The economic model that built prosperity also exported its externalities into the air, water, and public health budgets of the communities around it.

The cleanup cost, once it came, was staggering.

By the 1990s, the European industrial sector was investing heavily in emission controls, effluent treatment, and waste management. These were good investments in the sense that they reduced harm. They were terrible investments in the sense that they were building expensive systems to manage the consequences of linear industrial design, rather than changing the design itself. Germany spent the equivalent of hundreds of billions of euros retrofitting industrial infrastructure built for a different economic model. Japan's Minamata disease settlements, acid rain damage remediation, and industrial site cleanup costs ran into trillions of yen. The United States Environmental Protection Agency's Superfund program has spent over $60 billion since 1980 cleaning up contaminated industrial sites, and the pipeline stretches for decades more.

Every nation that industrialized on the linear model spent the next 50 years paying to undo it.

There is an argument (you will hear it made with great confidence in development economics) that this is simply the price of development. The Environmental Kuznets Curve, as it is called, proposes that environmental degradation increases as countries industrialize and then decreases as they get wealthier and can afford to clean up. Grow first, clean later. The evidence for this theory is, to put it charitably, mixed. For some pollutants, in some countries, the curve holds. For carbon dioxide, ecosystem services, and biodiversity loss, the empirical record is devastating. More importantly, even where countries did eventually clean up, they did so using technology, institutional frameworks, and capital instruments that did not exist when they first industrialized.

Vietnam is industrializing now. The technology, institutional frameworks, and capital instruments for a different model exist now. The EKC argument asks Vietnam to accept 50 years of damage that was unavoidable in 1960 but is entirely avoidable in 2025. This is not economics. It is path dependency dressed as inevitability.

Vietnam's industrial zone program is, in scale terms, extraordinary. From 1991, when the first KCN (Khu Cong Nghiep, industrial zone) was established in Tan Thuan, Ho Chi Minh City, the country has expanded to approximately 410 operational zones by 2024, with a further 100+ in planning or construction. Supported by a legal framework that makes zone establishment relatively fast, incentivized by competitive land and labor costs, and turbo-charged by the post-China+1 FDI shift, Vietnam's industrial zone buildout is one of the most rapid infrastructure deployments in Asia's recent economic history.

Each zone is essentially the same model: concentrated manufacturing clusters, served by shared road infrastructure, often with shared utilities (electricity, water), and operating under central management. Some zones have begun adding waste management facilities. A small number have been designated "eco-industrial parks" and have received support from UNIDO's global EIP program to introduce additional efficiency measures.

The object-level metrics are improving. Water recycling is up. Energy efficiency is better. A handful of industrial symbiosis connections (where one factory's waste becomes another's input) have been established in UNIDO's pilot zones. The Ministry of Planning and Investment has published guidelines for EIP development. There is genuine momentum.

And yet.

If you look at Vietnam's industrial zones not as collections of factories but as systems (asking what flows into them, what flows out, how those flows connect to each other, and what the design of those connections enables or forecloses), a different picture emerges.

The zones are designed as sectoral silos. An electronics zone hosts electronics manufacturers. A textile zone hosts textile manufacturers. This is logical from an FDI attraction standpoint: anchor tenants attract suppliers, suppliers attract complementary manufacturers, the cluster builds competitive density. But it is exactly wrong from an industrial symbiosis standpoint. Symbiosis between industries requires diverse material and energy flows. An electronics-only zone produces electronic waste, heat, and specific chemicals. A textile zone produces fiber waste, water effluent, and heat. They could exchange. They almost never do, because they are not in the same zone, there is no infrastructure for exchange, and no policy framework makes the exchange economically attractive.

LEGO's Binh Duong factory is the clearest example of this problem at the object level. It is genuinely impressive: LEGO's ambition to run on 100% renewable electricity, treat wastewater on-site, and zero-waste-to-landfill is serious and measurable. But that factory's output includes thermal waste, production off-cuts, and logistical packaging materials that flow nowhere productive within the zone. They are managed as waste rather than designed as inputs. The factory is optimized as an object. The zone it sits in is not optimized as a system.

This is not LEGO's fault. LEGO is doing what it can within the infrastructure it is given. The problem is that the infrastructure was designed without system-level thinking.

According to research by Le Thi Thanh Mai and colleagues published in 2025, Vietnam's CO2 emissions grew at an annual rate of 7.26% between 1986 and 2014, with trade openness and energy use as primary drivers. The environmental Kuznets curve, they found, holds for income: Vietnam will get cleaner as it gets richer. But trade and energy intensity effects work in the opposite direction. Which means: as Vietnam's industrial zones grow more integrated into global trade networks, as they consume more energy to serve export markets, emissions from that sector will continue rising even as the country's overall wealth increases. The net-zero pledge has to fight against the structural incentives of the very model being deployed.

The UNIDO eco-industrial park program deserves credit for being the most serious systemic intervention in Vietnam's industrial zone landscape. It has converted three existing zones (in Can Tho, Da Nang, and Ninh Binh) to EIP status, introducing waste exchange agreements, energy optimization, and shared service infrastructure. These are real improvements. The methodology, derived from the Reidy-Roberts framework for industrial symbiosis, is evidence-based and rigorous.

But the UNIDO work operates within the constraints of existing zone design. It retrofits symbiotic connections into infrastructure that was not designed for them. It does not change the sectoral silo structure. It does not integrate with regional material flow planning. It does not affect the planning of new zones, which continue to be designed on the same conventional template. The pilot zones are demonstrations of what is possible at the object level. They are not changing the system-level design criteria for Vietnam's industrial infrastructure program.

In 1972, a network of companies in Kalundborg, Denmark began exchanging waste and by-products between themselves. An oil refinery shared excess gas with a power plant. The power plant shared waste heat with a fish farm and a pharmaceutical company. Gypsum from the power plant's sulfur scrubbers became wallboard. Fly ash became cement. Steam became district heating.

None of this was planned top-down. It emerged from a network of bilateral agreements between companies that discovered each other's waste streams were useful inputs. Over 50 years, the Kalundborg Symbiosis has involved 12 companies in over 30 resource exchanges, reducing annual CO2 emissions by approximately 635,000 tonnes, annual water consumption by 3 million cubic meters, and generating significant cost savings for all participants.

Kalundborg is now a 50-year-old proof of concept for industrial symbiosis. The core insight is that waste is an artifact of linear design. In a system designed for resource cycling, waste from one process is an input to another. This is not idealism. It is how natural ecosystems work, and it is how Kalundborg's participating companies have been making money for five decades.

The question for Vietnam is not whether industrial symbiosis works. That question was answered in Denmark in 1989. The question is whether Vietnam's industrial zone design makes symbiosis structurally possible.

Currently, it mostly does not. The barriers are systemic:

Zone design: Sectoral clusters prevent cross-industry symbiosis. Mixed-industry zones that integrate manufacturing with food processing, with logistics, with materials recovery, with energy generation create the diversity of flows that symbiosis requires. Vietnam's zone design does not prioritize this diversity.

Infrastructure configuration: Resource exchange between companies requires physical infrastructure: pipelines for steam, conveyor systems for material flows, shared storage for by-products. This infrastructure is far cheaper to build into a zone at design stage than to retrofit afterward. New zones going up now will not have it, because the design criteria do not require it.

Policy framework: Industrial symbiosis requires that waste has value. In Vietnam's current framework, industrial waste is primarily a cost to be managed and disposed of. A carbon price, a resource extraction tax, or a mandatory material disclosure system would change the calculation: making linear waste disposal expensive and resource exchange attractive. None of these are currently in Vietnam's industrial policy framework.

Planning integration: Vietnam's industrial zones are planned and managed by the Ministry of Planning and Investment, while environmental regulation sits with the Ministry of Natural Resources and Environment, and energy policy sits with the Ministry of Industry and Trade. This fragmentation means that zone design decisions are made without integrated analysis of material flows, energy networks, or waste stream potential. System-level design requires system-level governance.

At Except, we have worked on industrial symbiosis programs in Vietnam and across Southeast Asia for over a decade. We ran sustainability leadership masterclasses in Ho Chi Minh City in 2023, working with entrepreneurs, business leaders, and investors who are deeply aware of these challenges and are actively looking for better frameworks. The desire for systemic solutions is real. What is often missing is the institutional vehicle to act on that desire at the scale of zone design and infrastructure planning.

Here is the argument that the EKC theorists miss: the economic case for industrial symbiosis is not primarily environmental. It is competitive.

Industrial symbiosis reduces costs. When your steam waste heats your neighbor's facility instead of being vented to atmosphere, you both save money. When your packaging waste becomes your neighbor's raw material instead of going to landfill, you both save money. Kalundborg's participants have documented over $160 million in cumulative cost savings over five decades. The business case compounds: every resource cycle closed reduces exposure to commodity price volatility, because inputs come from within the system rather than from global markets.

Industrial symbiosis also builds supply chain resilience. The COVID-19 pandemic demonstrated with brutal clarity what happens when global supply chains break. Zones designed with densified local resource cycling are less exposed to global supply disruption, because they have built-in redundancy through their symbiotic networks.

And here is the FDI argument that Vietnam's investment promotion agencies should pay attention to: the global capital market is changing. ESG-aligned FDI is growing rapidly. Premium brands (LEGO is the example already in this country, but there are hundreds of others) are actively looking for industrial locations that support their sustainability commitments, not just their cost requirements. A Vietnam industrial zone that can credibly offer industrial symbiosis infrastructure, circular supply chain integration, and documented emissions performance will attract a category of FDI that conventional zones cannot. The 2030s FDI landscape will reward circular zones.

Vietnam's current industrial zone program is building for the 2020s FDI landscape. It is competing on cost, speed, and basic infrastructure quality. These are the right metrics now. In ten years, they will be table stakes, and the competition will have moved to sustainability performance.

The infrastructure being built today has a 30-to-40-year useful life. The industrial zones opening in 2025 will still be operating in 2060, five years past Vietnam's net-zero target. If those zones are designed on the linear, sectoral-silo model currently being deployed, they will be generating emissions, material waste, and resource inefficiency for the next four decades. The transition cost to retrofit them for circular operation will be enormous. Germany and Japan already demonstrated that.

Vietnam does not have to pay that cost. But the window to avoid it is not unlimited.

The deepest problem is not a technology gap or a capital gap. Those can be closed. The deepest problem is a framework gap: Vietnam's industrial policy literally cannot see the system level.

The planning tools in current use count zones, square meters, jobs, export revenues, and FDI inflows. These are object-level metrics. They measure the performance of individual components. They do not measure the performance of the system those components constitute: the material flow efficiency of the zone network, the resource cycling rate of the industrial sector, the emissions intensity per unit of economic output as a function of industrial design rather than just production volume.

Without system-level metrics, system-level decisions cannot be made. Planners who can only see the object level will optimize for object-level performance. They will count LEED certifications and installed solar capacity and individual waste management facilities. They will celebrate LEGO's factory as a win. And they will continue building sectoral-silo zones that prevent the systemic resource cycling that would make those wins multiply.

This is not a Vietnamese problem. It is a problem Tom sees in industrial policy everywhere. Object-level thinking is the default. It is easier to count. It produces visible outputs. It satisfies the metrics that funders, investors, and governments have agreed to measure. System-level thinking requires a different analytical infrastructure: material flow analysis at zone and regional level, cross-sector symbiosis mapping, infrastructure configuration modeling, and governance frameworks that span the ministries currently operating in silos.

The UNIDO EIP program is beginning to build that analytical infrastructure. The Vietnamese government's own EIP guidelines are a start. But they need to become the design criteria for new zones, not the retrofit criteria for existing ones.

Here is the practical implication. Every industrial zone that breaks ground in Vietnam in the next five years is a 40-year infrastructure decision. The question to ask about each of those zones is not "does it meet current environmental standards?" It is "does its design make circular resource flows structurally possible at the scale of the zone, the region, and the industrial sector?"

That is a different question. It requires a different analytical framework. It requires system-level metrics. It requires integrated governance across planning, environment, and energy ministries. It requires zone design guidelines that build in mixed-industry diversity, resource exchange infrastructure, and material flow connectivity from the beginning: not as retrofits, not as aspirational guidelines, but as design criteria that determine approval.

Vietnam's opportunity is real and it is time-limited.

The country has approximately a decade before the current infrastructure buildout locks in the industrial configuration that will persist to mid-century. Ten years in which the zones not yet built can be designed differently. Ten years in which the policy framework can be updated to require system-level thinking at zone design stage. Ten years in which the JETP financing mechanism can be used to capitalize the incremental cost of circular industrial infrastructure rather than object-level green upgrades.

We have worked in Vietnam long enough to know that the people developing industrial policy are not indifferent to these arguments. They are working within a system that selects for speed, FDI volume, and cost competitiveness, because those are the metrics by which success is measured. Changing those metrics is a governance challenge, not a technical one.

But the governance challenge is tractable. Vietnam has demonstrated, with its COP26 commitment, with its JETP agreement, and with its EIP program, that it can make systemic policy decisions when the political will exists. The question is whether that will can be mobilized at the level of industrial zone design criteria before the window closes.

The 410 zones already built are what they are. The 100 more in the pipeline are the ones that matter. Design them as systems. Build in the diversity of flows that symbiosis requires. Require resource exchange infrastructure from the first shovelful. Make the circular configuration the default, not the premium option.

The LEGO factory in Binh Duong is impressive. It should be the starting point for what the zone around it does with everything that factory produces. That connection between the object-level ambition and the system-level infrastructure is the gap Vietnam has a chance to close right now.

After 2035, closing it will be much more expensive.