Summary
A Nature commentary argues that national economic accounts should track how supply chains rely on computing, data networks, electricity and digital payments. The proposed measures could help identify bottlenecks before disruptions spread.
National economic accounts could be expanded to show how production depends on computing, data networks, electricity and digital payments, argues a 6 October 2026 Nature commentary by Dabo Guan, a professor at Tsinghua University and University College London. The proposal is to make these dependencies visible in a practical statistical framework, so governments and businesses can identify supply-chain bottlenecks and direct resilience measures towards them.
The argument reflects a change in how goods and services are produced and traded: information now moves alongside physical materials. Companies use digital systems to share product designs, manage inventories and schedules, coordinate logistics and settle payments. When those systems fail, disruption can spread between suppliers and customers even when physical goods and facilities are undamaged.
Why current economic accounts miss digital dependencies
National statistical offices use input–output tables to map the monetary value of goods and services flowing between industries. Guan argues that these tables do not capture the energy, data or computing capacity on which production increasingly relies. Companies, meanwhile, record spending on information technology, but not necessarily the data, computation and electricity required to keep operations running.
Some statistical work already tracks digital activity. Digital supply–use tables, developed with an OECD framework, record the production, trade and consumption of digital goods, services and transactions. Canada and Australia published estimates in 2019, and the OECD issued a compilation handbook in 2023. The commentary says these tables are still based on economic values, rather than the physical and operational constraints that can limit supply.
Those constraints are increasingly connected. More AI use can raise demand for computing; data centres need electricity; and the resulting data must move across networks and be stored. Digital transactions also depend on systems for recording and settling payments. The International Energy Agency projection cited in the commentary puts data-centre electricity consumption at 415 terawatt-hours (TWh) in 2024, potentially rising to around 945 TWh by 2030.
A proposed way to trace points of failure
Guan points to disruptions that reveal dependencies only after they break. A fire at an electricity substation shut London Heathrow Airport for a day in March 2025, according to the commentary, disrupting flights and travel. It says a subsequent official review found that the airport’s reliance on that single supply point was not known to the energy networks. The commentary also cites a 2025 cyberattack that halted Jaguar Land Rover production for five weeks and affected thousands of supply-chain firms.
The proposed statistical tool would extend existing accounting systems to track quantities such as computing resources supplied and used, data generated and transferred, and how those flows connect to electricity constraints and payment settlement. It would also trace dependencies across domestic and cross-border production networks. Guan’s aim is a system that national statistical offices can compile, audit and update, rather than a one-off dataset or a collection of proprietary company dashboards.
Such a map could help identify, for example, a facility or network on which multiple important sectors depend. Governments and companies could then target resilience spending at those dependencies, improve grid connections where computing capacity is constrained, or develop alternative payment-settlement routes. The commentary sets out the measurement need and broad categories to track; it presents a proposal, not a finished statistical standard.