The coming electro-tech revolution

This post tries to document in brief and graphically the emerging electro-industrial stack that has the potential to transform the physical world just as digital technologies have done elsewhere. 

This post draws mainly from four sources - Noah Smith’s introduction to the Electric Age, a succinct primer on the electro-industrial stack by Ryan McEntush on a16z, a brilliant 40,000-word comprehensive essay by Packy McCormick and Sam D’Amico, and an excellent graphical distillation of the electrotech revolution by Ember. 

McEntush describes these technologies as the “bridge between software and the physical world… enabling machines to behave like software: minerals and metals processed into advanced components, energy stored in batteries, electrons channeled by power electronics, force delivered by motors and actuators, all orchestrated by software running on high-performance compute.”

Noah Smith has a good summary of what’s driving the electrotech revolution:

Basically, these three things allow electric motors to replace combustion engines (and steam boilers) over a wide variety of applications. Batteries make it possible to store and transport electrical energy very compactly and extract that energy very quickly. Rare-earth motors make it possible to use electrical energy to create very strong torques — for example, the torque that turns the axles of a Tesla. And power electronics make it possible to exert fine control over large amounts of electric power — stopping and starting it, rerouting it, repurposing it for different uses, and so on. With these three technologies, combustion’s main advantages vanish in many domains. Whether it’s cars, drones, robots, or household appliances, electric technology now has both the power and the portability that only combustion technology used to enjoy.

Thanks to low-cost, high-energy density batteries, we can now store, move, modulate, and deliver electricity efficiently, powering everything from data centers to drones. There are four parts to what McCormick and D’Amico call the electrotech stack - batteries, power electronics, motors and actuators, and a compute layer. McEntush describes the importance of power electronics.

Power electronics are the hidden nervous system of modern machines. At their core are power semiconductors which, unlike logic chips that process information, manage energy itself — converting, inverting, and regulating flows between sources and loads. Historically, power systems relied on slow silicon switches, steel-core transformers, and bulky analog controls. At high voltage and frequency those approaches are at their limits. Wide-bandgap devices — like silicon carbide (SiC) and gallium nitride (GaN) — switch faster, withstand higher temperatures, and enable precise digital control. This software-driven, solid-state (no moving parts) foundation stitches together the electro-industrial stack… Scaling WBG power electronics is also critical to easing the grid’s growing infrastructure bottlenecks… the way forward is solid-state transformers built with domestically produced WBG power electronics… Future systems will require vast amounts of precisely managed power; delivering it will depend on solid-state electronics under digital control.

And that of motors and actuators

Motors and actuators convert electrical energy into mechanical motion, like in a drone motor or an industrial robot arm. Today’s performance leader is the brushless permanent-magnet synchronous motor (PMSM) using NdFeB magnets, prized for torque density, efficiency, and compactness. But that advantage comes with a strategic cost: dependence on rare earths. Alternatives for motion systems span both motor design and actuation type, each with trade-offs… Actuation choices are also evolving. Flight surfaces, reclining seats, missile fins, landing gear, and industrial end-effectors are shifting from old-school hydraulics to electromechanical systems for lower weight, higher reliability, and precise digital control… The market is testing these across applications… Small gains in motor and actuator efficiency compound across the stack, and as general-purpose robotics scale, this industrial “muscle” will move larger fractions of GDP.

Finally, the compute layer is no less demanding.

The compute layer converts electrical energy into intelligence, controlling everything from autonomous vehicles to advanced weapon systems and industrial robots. Today’s performance leader is 2 to 4 nm-class logic — most often, GPUs designed by Nvidia and fabricated by TSMC… Compute extends beyond the GPU alone; advanced packaging now sets performance limits as much as transistor scaling. System design has expanded from single chips to whole machines, like co-optimizing die, memory, and interconnects alongside rack-level power delivery and cooling. Just as important, software frameworks, compilers, kernels, and drivers map models onto this topology, manage communication and memory, and orchestrate performance across infrastructure. Chinese firms lead in scale for mature nodes and low-cost packaging; the West leads in EDA, lithography, and software ecosystems, but still lacks true leading-edge manufacturing capacity outside Taiwan.

This graphic captures the composition and evolution of the four parts of the electro-tech stack.

The revolution in electrotechnologies has been driven by innovations that have dramatically increased performance outcomes… 

… while also equally dramatically lowering costs…

… and become competitive enough with existing technologies. 

The result of all this is that electricity has become the largest source of energy…

… across sectors…

… and is becoming the major share of incremental demand.

These technologies build on the foundations of an emerging category of critical minerals, which have unique properties, have limited substitutes, and have fragile supply chains, and are “becoming as strategically contested as oil was in the 20th century”. This is a good graphic.

China dominates the midstream of these minerals - the chemistry, metallurgy, and finishing that creates chemicals, alloys, foils, laminations, and powders. It involves the highly specialised and high margin operations of chemical processing (where raw concentrates — from ore, brine, or slurry — are refined into high-purity compounds and metals) and manufacturing of advanced materials with high purity, unique chemical, structural, and magnetic properties (battery precursors and active materials, magnets, and specialty alloys), that feed directly into end-use industries. Manufacturing is more sophisticated and challenging to master.

This isn’t assembly-line work, but precision materials engineering. Processes like heat treatment, doping, sintering, and nanostructuring push performance limits. Minor deviations in particle size or crystal structure can affect battery cycle life or magnet strength. Specs are tight, recipes proprietary, and tolerances unforgiving. Manufacturing follows a different logic than mining or refining: it prioritizes consistency, qualification, and yield over sheer volume. This is especially true for defense-critical components like high-temperature magnets — strategically vital but smaller markets with strict regulatory regimes, long product cycles, and little tolerance for disruption. OEMs tend to value reliability and traceability over novelty. Incumbents like Umicore (cathodes) and Vacuumschmelze (magnets) have spent decades mastering complex processes and earning deep customer trust. Their edge in materials science, process control, and compliance makes it difficult for new entrants to compete.

This is a great summary by McCormick and D’Amico on how the West gifted its leadership in all four parts of the electrotech stack to China.

The four key Electric Stack technologies were invented at various points between the 1960s and 1990s in America, Japan, and the UK, and reached critical maturity around the same time in the 1990s. Then, in many cases, we sold the future. GM sold its neo magnets division, Magnequench, to China for $70 million. A123 Systems, which invented the Lithium Iron Phosphate (LFP) battery, went bankrupt and sold to Wanxiang for $257 million in 2013… By controlling these four technologies, China has become the world leader in everything from EVs to drones to electric bikes to robots.

A giant piece of this is that mastery of this stack applies across domains, allowing market leaders like BYD to make everything from cars, to home energy products, to iPads, to much of the world’s drones. Within the whole sector – the components, software, and expertise largely transfer – meaning mastery of one product of the stack allows success in scaling others. Advantages compound. The result has been China getting the best “LEGO set” in the world, with regards to this stack. 

Three imperatives are driving the electrotech revolution - physics, economics, and geopolitics. Let’s briefly examine each. Electrotech is three times more efficient than fossil fuels in sectors making up two-thirds of fossil fuel demand (electricity, road transport, and low-temperature heat), and is more efficient than other solutions like CCS, biomass or hydrogen.

And it requires over 50 times less material than fossil fuel systems. 

Electrotech has enormous opportunities to innovate and improve efficiencies and drive down costs. 

And it is small and modular, thereby allowing for greater scope and span for experimentation.

Furthermore, unlike fossil fuels, electrotech fuels (such as solar and wind energy) are available globally, thereby increasing energy security.

How, China being the runaway leader on electrotech, brings serious security risks.

Finally, a matrix of policies aimed at the promotion of electrotech technologies. 

Restoring the balance in politics, economics, and beyond

A recurring underlying aspect of many problems that we see today in the economy, polity, and society is that of imbalance in the pursuit of ideas and ideologies. 

In public discourses, we see it in the prioritisation of individual rights over the social and the collective. This manifests in the dominance of liberalism on social issues and free-market principles on economic issues. 

In recent years, especially in the US, woke liberalism has been squeezing out conservative views. The resultant tensions manifest in the political realm on issues like family values, transgender rights, and most prominently in debates on immigration. It has resulted in the near-complete polarisation between the liberal and conservative political camps, with almost no meeting ground. The political centre stands egregiously vacant. 

In economics, it’s about the pursuit of free market and efficiency-maximising ideologies gone too far. I have blogged hereabout 25 economic orthodoxies that conflict with reality. Generally, in Western economies (again, especially the US), it is about consumption marginalising production, and the elevation of virtual innovation (software or digital) over physical innovation. The near deification of AI and the complete neglect of the (perhaps more important) emerging electro-industrial-tech stack is an illustration. 

In international trade, trends like tariff reduction, trade liberalisation, globalisation of value chains, and offshoring have clearly gone too far. The most striking manifestation of this trend is China’s stranglehold on the world economy in manufacturing, and the associated destruction of manufacturing bases and loss of manufacturing jobs across countries.

Nowhere is the loss of balance as salient as in finance. It manifests in the disproportionate and growing importance of private (venture capital and private equity) over public markets (banks and capital markets) in financial intermediation. I have written here about the problems and consequences of financialisation going too far. It is evident even in the preference among youth for liberal arts education over STEM courses in colleges. 

In important areas of global concern like climate change, this imbalance has led to a headlong plunge into renewable energy sources and electric vehicles and wholesale abandoning of fossil fuels and traditional industries. It has swept aside daunting transition challenges like financing sources, sunk costs of legacy systems, unsustainable mitigation and adaptation costs for developing countries, technical problems of integrating energy systems, and so on. 

This imbalance is also stark in our engagement with emerging technologies like Artificial Intelligence (AI). The agenda on automation and the application of AI is almost exclusively framed and driven by Big Tech companies. The public narrative is framed in terms of innovation and human progress, the most desirable of all objectives. But the driving force behind the race to adopt these technologies is efficiency maximisation and cost reduction, which enhance business competitiveness and increase profits. Its larger consequences are never a consideration, and adoption is done without any public debate.

On this issue, as I blogged here, Daron Acemoglu and Simon Johnson have shown how ideas and technologies have deep political significance and how agenda-setting shapes the nature of the “progress” arising from these ideas and technologies. The agenda framing makes certain aspects of the issue salient while obscuring certain others. This process is deeply political. The political power balance determines what’s made salient and what’s obscured.

Another imbalance surrounds the marginalisation of the role and importance of governments and the elevation of the private sector. For example, the ever-expanding use of consultants and outsourcing of services has enfeebled government capabilities and left the state open for capture by vested interests (see this and this). It has amplified the self-fulfilling dynamic of governments are inefficient and therefore should deregulate and exit. There’s a real risk that the new movement of deregulation will add to this enfeeblement without achieving anything substantial in its original objective. 

In development, it is about neglecting plumbing issues, such as state capability, in favour of innovation, management theories, and the application of IT solutions. For example, in school education, the fundamental issue of improving classroom instruction quality (and therefore teacher capabilities, motivation, pedagogical techniques, and teacher-student engagement, among others) is often overshadowed by the pursuit of smart classrooms, digital content, and blended learning, among other initiatives. I have blogged here on ten things in development orthodoxy that deviate from reality. 

The common thread in all these examples is that of an idea or ideology being taken to its extremities, sweeping aside counter-views. The resultant common deficiency is that of balance. Any idea or ideology unrestrained by countervailing views and unmoderated by reason and prudence becomes unbalanced and verges towards fanaticism. 

The Greeks had a word for balance, meson, or the middle. At a philosophical level, the Bhagavad Gita refers to the highest state of balance, or equipoise. In physical and social systems, this balance is achieved in a state of equilibrium (it is a different matter that there might be multiple equilibria). The essence of stability in any system is this balance. 

However, the innate dynamic of ideas and phenomena generates a gravitation or swing to the extreme. This is just as true of social systems as it is of physical systems. Any trend - capitalism, socialism, statism, globalisation, liberalisation, privatisation, deregulation, financialisation, automation, etc - if left to itself, follows a self-reinforcing feedback loop that ends up destroying countervailing forces and spawns its excesses. 

I blogged here about a universal dynamic to how ideas evolve and play out. They trigger interest and get gradually adopted, with their degree of adoption increasing over time. This, in turn, creates distortions that cause a backlash against the idea. The backlash strengthens over time and results in a correction of the excesses that had seeped into the idea. 

A simple framework to explain this is the Hegelian dialectic, wherein as a thesis (idea) evolves, it conflicts with its emerging antithesis to generate a synthesis, often a better state of affairs. As Hegel wrote, thesis begets anti-thesis, both of which undergo a struggle to generate a synthesis, and so on it goes.

In their highly influential book, The Fourth Turning, Neil Howe and William Strauss describe a cyclical trend in history. Their century-long cycle encompasses four phases, or turnings as they call it - High, Awakening, Unravelling, and Crisis. Each turning lasts a social generation of about 20-25 years. 

The work of Howe and Strauss has resonance in other similar interpretations of history. In this essay from 1976, Sir John Bagot Glubb, the former Commander of the Arab Legion, describes history in terms of cycles of around 250 years, or 10 generations of 25 years each. Peter Turchin, an expert in cliodynamics, uses maths to model historical changes and find historical cycles.

Be that as it may, this imbalance has inevitably forced backlashes across fields - politics, economic policies, trade, financial markets, public systems, etc. Across them, orthodoxy is on the retreat. Populist politics, anti-immigrant sentiments, protectionism, revival of manufacturing, retreat of globalisation and offshoring, support for fossil fuels, etc., are a result of this backlash. 

In the circumstances, the challenge for us is to identify and acknowledge the imbalance within systems, and then figure out ways to deal with the problem. There’s a need to consciously cultivate or encourage countervailing forces to achieve a dialectical balance. Only open systems can engage meaningfully through such a process to achieve balance. 

In this context, it is also useful to draw from a concept formulated by Aristotle, phronesis, or practical wisdom. It is the ability to exercise good practical judgement, as the highest intellectual virtue. Unlike theoretical knowledge or technical skill, it is about knowing how to act rightly in specific situations by balancing general rules with context, ethics, and experience to achieve good outcomes. Its critical value is underlined by Albert Hirschman, describing the ability to exercise good judgment as the binding constraint in development.

This is important since a related theme associated with the imbalance is the supremacy of expertise and technocracy, and the marginalisation of prudence and politics. This trend must be acknowledged and reversed for any meaningful effort to restore balance in these realms.

Weekend reading links

 1. US equity market exceptionalism may have peaked, says Ruchir Sharma. 

A weakening dollar helped the rest of the world outperform the US in 2025, reducing the country’s share of the global stock market index from a high point of 66 per cent at the end of 2024 to 64 per cent now. While “American exceptionalism” may have peaked, this shift has ample room left to run given the still wide disconnect between America’s market cap and its 26 per cent share of the global economy.

I agree with this assessment of the Chinese economy. 

China’s domestic economy is hardly growing, weighed down by a busted property market, too much debt and a shrinking population. But strength defines the export sector, which is expanding its share of global markets and propping up the overall economy. Without the export surge, nominal GDP growth would be barely 3 per cent, significantly lower than the officially reported rate of about 4 per cent. Wall Street analysts and economists keep urging China to unleash new stimulus, but their Keynesian bias is blinding them to the underlying problems. China’s total debt including households and corporations is already above 300 per cent of GDP; its augmented fiscal deficit (which includes its influential local governments) is above 11 per cent of GDP. Lacking the money for stimulus, Beijing will be hard pressed to spend more, and the domestic economy will keep disappointing... For the last two years, China has engineered a dramatic increase in its export volumes by slashing prices and holding down the value of the renminbi. As a result, China continues to gain global market share at the expense of rival exporters.

2. Adam Tooze points to an important consequence of Trump's Venezuela invasion: the stabilisation of Guyana's oil market. 

3. This about the invasion of Venezuela and the kidnapping of Nicolas Maduro, and the emerging Donroe Doctrine generally is important. 

Competition with China over resources in America’s “backyard” will only get fiercer, predicts Stephanie Junger-Moat, chief executive of Karcsi Global, a corporate consulting firm focused on Latin America. Chinese companies have extensive interests in both oil and mining in Latin America. They have invested in the “Lithium Triangle” of Argentina, Chile and Bolivia to supply their battery industry and have significant stakes in Chilean copper and Peruvian iron ore. Junger-Moat says competition could intensify if Trump moves to exert more control over the Panama Canal, which handles 5 per cent of maritime trade, and puts pressure on Latin American countries to limit Chinese trade and investment. “The clear short-term losers would be the countries stuck in the middle of this that are resource rich but with little negotiating power,” she says.

Given China's significant investments in mineral extraction in Latin American countries, the US invasion raises the prospects of a face-off between the two countries in America's backyard.

China has gone from doing nearly no business in the region two decades ago to bilateral trade worth more than $500 billion in 2024. Chinese mining companies extract copper from Peru and lithium from Argentina. China’s agricultural conglomerates import lifeline commodities like soybeans from Brazil. Chinese utilities power entire cities. China controls much of the shipping infrastructure and the ports that transit goods across the Pacific. Latin America’s 670 million consumers are also buying Chinese brands. In Mexico, dealerships sell gasoline-powered Chery cars and MG sedans. In Brazil, the fast-food chain Mixue sells ice cream, the e-commerce platform Meituan delivers food and the ride-hailing service Didi ferries people around. In Peru, Xiaomi smartphones are popular...

China’s pursuit of deeper ties in Latin America began two decades ago. At the time, Chinese companies were scouring the world to secure copper, oil and iron ore to power China’s breathtaking economic growth. Chinese banks extended ever-larger loans to countries across Latin America in exchange for oil and critical minerals, and along the way China began building railways and highways and selling its goods. Since then, China has economically displaced the United States in 10 of 12 countries in South America alone, according to research by Francisco Urdinez, an associate professor of political science at the Pontifical Catholic University of Chile. China now engages in more trade, investment and development financing than the United States in most of the region, including Central America. China is Latin America’s largest official source of aid and credit, offering an estimated $303 billion in financing across the region between 2000 and 2023, according to AidData, a research institute at the College of William and Mary in Williamsburg, Va. Between 2014 and 2023, for every $1 lent or given in aid by the United States in Latin America and the Caribbean, China provided $3, said Brad Parks, the executive director of AidData. These investments in many cases have left the countries saddled with debt and obligations to fulfill contracts for commodities like oil.

Venezuela itself may well become the first flash point. 

China is still owed, by one estimate, about $10 billion that Venezuela is paying off through oil shipments. In 2024, more than half of its crude exports — or 768,000 barrels — went to China, according to Kpler, a global oil monitoring service.

4. Some facts about India's Electronics Manufacturing Service (EMS) providers.

Dixon, a 30-year-old company, is on track to produce around 40 million smartphones in FY26, capturing over 50% of India’s mobile outsourcing market. Yet, it spends less than 1% of its revenue on R&D, according to its annual reports. In contrast, Chinese ODM Huaqin Technology, which also started off as a contract manufacturer, spends around 5% of its revenue on R&D... India’s largest listed EMS players with foreign-partner JVs—the likes of Dixon, Syrma SGS, and Amber Enterprises—spend 0.2–0.9% of their revenue on R&D. Their patents reflect this: Dixon’s are valued at a mere Rs 2 crore, while Syrma’s intangible assets, valued at over Rs 1.5 crore in FY25, grew through acquisitions rather than internal R&D.

They have also not been able to capitalise on their JVs with foreign design companies (ODMs). 

Indian EMS players have chased this be-all-end-all through joint ventures (JV) with Chinese, American, and Taiwanese original design manufacturers (ODM). They’ve brought home some of the biggest ODMs through JVs such as Dixon-Longcheer, Bhagwati-Huaqin, and Syrma SGS-Shinhyup Electronic... On the face of it, Dixon and its ilk’s JVs with foreign partners seem a win-win Indian companies learn to manufacture complex products such as smartphones and TVs, and foreign players get access to one of the largest consumer-electronics markets in the world. But there’s often an implicit power imbalance in the way they’re structured. 

“Such partnerships transfer process discipline, quality systems, and operational know-how, but core design authority and roadmap ownership usually stay with the foreign partner,” says Sanchit Vir Gogia, an analyst at tech-research-and-advisory firm Greyhound Research. Another executive at an EMS firm put it more directly. “Even if they tell us how a certain printed circuit board is made, it’s limited to supporting its manufacturing in our factories. We never know what future technologies foreign companies are working on back home,” he says. The result: partnerships meant to facilitate tech-transfer from one partner to another don’t actually end up doing so. The little design that is getting localised in India is that which is tweaking pre-existing products for local use cases... Foreign partners carefully control what—and how much—knowledge flows to their Indian counterparts, which creates a hierarchical power dynamic within the partnership.

On top of it all, the integration of design and manufacturing ecosystems remains broken. “India has research institutions and it has manufacturing lines. What it lacks are applied industrial labs tied directly to clusters—places where engineers solve production problems, test reliability, and feed learning back into design,” says Gogia. Without this shared infrastructure, firms face a beguiling choice: over-invest individually or under-invest collectively. Most end up choosing the latter. Most EMS firms operate on razor-thin margins. “Where capital costs are already high, long-horizon R&D stops being a strategic bet and starts looking like a threat to liquidity,” says Gogia. The only thing that makes it viable is scale.

5. In a reflection of ageing and weak infrastructure, parts of Berlin suffers a power outage that stretches from Saturday to Wednesday, in Germany's longest outage since the Second World War!

6. India's declining tax to GDP ratio

And tax buoyancy

I'll blog more on this separately. But this is intriguing. 

7. On India's public e-bus systems.

Under India’s electric bus programme, cities don’t buy e-buses directly. Procurement runs through central agencies earlier under the Faster Adoption and Manufacturing of Electric Vehicles (FAME) scheme, and now under PM-ebus sewa. These agencies aggregate demand, float tenders, and sign long-term contracts with manufacturers and operators. The goal is scale and cost control. It also shifts control away from cities. Central government agencies such as CESL and NVVN act as buyers of record. They sign contracts with OEMs and hold warranties. Payments flow through them. Meanwhile, state transport undertakings (STUs) operate the buses. They schedule routes, deploy drivers, and handle breakdowns on the ground—paying a per-kilometre fee which goes to the OEMs. “This structure wasn’t accidental,” said a former NVVN official. “The electric bus push came from the Centre, and the subsidy came from the Centre. So control also stayed central. That’s very different from how states normally buy buses.”

... The model works when buses run as expected. When they do not, authority fragments. STUs can log faults and track downtime. They cannot escalate repeated failures directly to manufacturers or demand design fixes. Those decisions sit higher up the chain and move through contract clauses and payment cycles... The state transport units can track downtime and flag violations, but the actual fines are processed by the contract owner—the central aggregator agencies in this case. What follows is less a punishment than a reconciliation exercise. Numbers are logged. Penalties are calculated. And eventually, amounts are just adjusted against future payments, often weeks later... More importantly, those deductions are capped by design. Under the gross cost contracts used for electric buses, penalties apply only after fleet availability falls below a defined threshold, usually around 85–90%. Contracts also set a maximum amount that can be deducted in a day and another cap for the entire month. Once those caps are reached, deductions stop, even if the buses continue to remain idle. Effectively, a bus that breaks down briefly and one that is out of service for several days can end up facing similar financial penalties.

8. Elon Musk tweeted in response to Nvidia's launch of its own autonomous driving software. 

What they will find is that it’s easy to get to 99% and then super hard to solve the long tail of the distribution.

This is the challenge for AI applications

Musk is right that it is the edge cases that have made fully autonomous driving so hard. The real world is way messier than any computer simulation. A good example occurred in San Francisco in December when a power outage knocked out scores of traffic lights, causing problems for the robotaxis operated by Waymo, owned by Alphabet. In spite of its fleet clocking up more than 100mn miles of autonomous driving, Waymo’s cars froze when the lights went dark, clogging the city’s streets. In such unexpected circumstances, remote human interventions are still needed to instruct the cars how to respond. Waymo uses an app called Honk to summon human gig workers to solve other problems too, such as shutting car doors after passengers have left them open.

9. China's EV manufacturing is increasingly dependent on exports for survival.

The China Passenger Car Association, an industry group based in Beijing, has forecast the country’s auto exports will rise by 20 per cent this year, driven by EV sales from Tesla rival BYD... Mexico, Middle East, Russia and parts of Europe are among the top export markets, according to Chinese data. Chinese carmakers are rapidly setting up factories and sales networks around the world to circumvent rising tariffs, except in the US, where they are limited by levies and security controls. Overseas sales, which include exports and cars made by Chinese companies in markets outside China, account for about 20 per cent of the Chinese industry revenue and close to half its earnings, according to UBS. Seven of China’s biggest auto groups — BYD, Great Wall Motor, Chery, SAIC, Changan, GAC and Geely — have 31 factories overseas. Shenzhen-based BYD, UBS analysts noted, was “the most aggressive” with plans to double the number of its European stores from about 1,000 to 2,000 by the end of 2026. The domestic outlook in China remains tough as Beijing scales back tax breaks and subsidies for electric cars. This will put financial pressure on a crowded EV market where more than 100 companies face wafer-thin margins and a regulatory crackdown on unsustainable discounting practices. Goldman Sachs analysts said profits this year would be further diminished with “aggressive” EV competition, with the release of 119 new models — roughly one every three days — and slowing volume growth.

This is an excellent graphic on technology disruption: in the Chinese automobile market.

As China’s economy boomed and car ownership soared, annual ICE sales rose from 3.9mn in 2005 to peak at 23.9mn in 2017, according to data from the China Association of Automobile Manufacturers, another local industry group. Now, having retreated to about 14.5mn in 2025, ICE sales in China are set to sink to below 5mn by 2030, their lowest level in about 25 years, UBS has forecast.

10. Labour market skills in the age of AI

An alternative telling of the boom in software employment and salaries is a boom in demand for people who — while certainly mathematically skilled — are primarily distinguished by their aptitude in using these skills to work closely with others in finding creative solutions to complex and multi-faceted problems. Contra the narrow focus of policymakers on Stem subjects or coding, now more than ever our economy rewards broad skillsets: team players, problem solvers, good communicators and creative thinkers.

And this.

Even within tech and other deeply quantitative fields, roles combining strong coding skills with creativity and collaboration are the ones in which people have thrived. People in mathematical jobs with the lowest emphasis on social skills (actuaries and mathematicians among others) have fared markedly worse both in terms of employment and earnings than those for whom collaboration, creativity and interpersonal interaction play a larger role (software developers among them).

11. Finally, the job market for economists in the US is bleak.