Weekend Food For Thought
On Todays Menu: Busiest Ports in The World, China's Exports, “Your Power Tools Got Worse On Purpose”, A New Frontier of Investment Opportunities, The Empire of Wuxi, and much more...
Hello from Lisboa,
I hope you had an interesting and productive week.
Claude Sonnet states that:”A threshold effect occurs when a system absorbs change gradually — then suddenly shifts to a fundamentally different state once a critical boundary is crossed.”
Take Note…look out for the gradual shifts that has the potential for massive change once a threshold is met
1 Getting Visual
2 If You Read One Thing Today - Make Sure it is This
3 Consequential Thinking about Consequential Matters
4 Big Ideas
5 Big thinking
6 Treat Possibilities as Probabilities
1 Getting Visual
Container Kings: Show me global manufacturing in one visual - Seaports are the backbone of global trade, moving the containers that carry everything from electronics to clothing to industrial goods. In 2024, the world’s ports processed the equivalent of 743 million shipping containers, and more than half of that traffic passed through just 20 ports. China alone accounts for over 40% of global container traffic, reflecting its central role in global manufacturing and export supply chains. Of the six busiest ports worldwide, five are in China, led by the Port of Shanghai. Shanghai processed over 51.5 million TEUs in 2024, making it by far the busiest port worldwide. Singapore, the runner-up, processed more than 10 million fewer TEUs that year. China’s port strength extends well beyond Shanghai. Ningbo-Zhoushan (39.3 million), Shenzhen (33.4 million), Qingdao (30.9 million), and Guangzhou (26.1 million) are all busier than any port outside China except Singapore. Their prominence today reflects not only China’s economic ascent into its role as “the factory floor of the world,” but also historical tradition. Between 1757 and 1842, Guangzhou served as the sole port where Western merchants could trade with China, back when the city was still known as Canton. China’s ports may dominate the rankings, but they are not the only major shipping hubs in Asia. The continent is home to three-quarters of the world’s busiest ports, including Singapore, Busan (24.4 million), and Port Klang (14.6 million). Many of these ports lie at strategic chokepoints. Singapore and Port Klang both sit on the Strait of Malacca, the busiest shipping strait in the world, through which roughly 25-30% of globally traded goods travel. Further west, Jebel Ali in Dubai (15.5 million) is the busiest port outside East Asia. Located in the Persian Gulf, it is the world’s largest manmade harbor and is owned by Emirati logistics giant DP World. Europe and the U.S. remain major trading hubs, but their busiest ports now trail far behind Asia’s largest container gateways. Rotterdam (13.8 million) is the busiest non-Asian port worldwide, followed by nearby Antwerp-Bruges (13.5 million). Notably, only one African port appears in the world’s top 20: Morocco’s Tanger-Med (10.2 million). No ports from Central or South America made the ranking.
Global Trade - One constant with shifting dynamics below: China’s share of global exports rising, the trading partners shifting…
Exponential Exports of Exponential Technologies - Exhibit 1: China’s exports of photovoltaic systems…
Exponential Exports of Exponential Technologies - Exhibit 2: China’s exports of Batteries…
Exponential Exports of Exponential Technologies - Exhibit 3: China’s exports of EVs…
Seeking Sovereign Power Sources - Europe has saved €100 million a day on energy imports since the Iran war started, from solar capacity already on the grid…Asia has also been steeping up their imports of EVs and Solar+Batteries since the Russia/Ukraine conflict and then again after the US/Iran conflict…Threshold effect underway…
GPTs & The Way of the Productivity J-Curve: “Whether it is the advent of electricity, the internal combustion engine, or the Internet, every technological revolution in history has delivered much the same experience for investors: short‑term pain followed by long‑term gain. The costs and benefits invariably follow a J‑curve. First come the costs. Heavy capital investment in the new technology marks the initial developmental phase. This is typically followed by a temporary decline in productivity and corporate profitability. Then comes a turbulent period of societal adjustment, when some jobs are destroyed by the emerging technology and new ones are created. Only after this do the benefits of the new technology’s diffusion begin to materialise — higher output, revenue and employment, alongside other gains for society, such as more leisure time. Take electrification in the late 19th century. Companies first spent heavily to rewire their factories, disrupting existing processes but with little to show for it in terms of productivity or profit. Only after they completely reorganised their operations around the new technology — changing factory layouts and introducing more continuous production — did the real pay‑off arrive, in the form of higher output per worker and lower costs. It was at this point, long after the initial spending wave, that rewards finally began to flow to investors. There is a useful lesson here for those trying to understand how the latest general‑purpose technology — artificial intelligence — might evolve. AI appears to be at its very early stages of development. At this juncture, many industries are experiencing a productivity dip: significant investments in data infrastructure, organisational restructuring and workforce retraining are temporarily outweighing measurable returns. Put another way, cognitive work processes are being redesigned more quickly than employees can adapt, causing productivity declines despite considerable technological advancements Against this backdrop, investors need to factor in the social and fiscal adjustment costs ahead, recognising that AI’s long‑term rewards are likely to come with a deeper, more uneven and more fiscally burdensome transition than many currently assume. However, this is not to say that the pain phase of the transition – with all the fiscal strain and social tensions it could bring – is not worth it. This is because of the Jevons Paradox, a phenomenon observed during almost every major technological shift in history: when technology makes a resource or task much more efficient, its effective price falls, so people and businesses find many more uses for it and total consumption can actually rise. If AI follows this historical pattern – which we think it will – more abundant AI inputs will ultimately expand the range and volume of tasks we perform, supporting new forms of work, novel industries and business models as well as higher living standards, rather than permanently displacing humans. Investors should view AI as a long build-out, rather than an imminent economic and investment game changer. While global AI spending is rising fast – projected to reach USD 2.5 trillion in 2026, up 44% year on year – it remains modest relative to the size of the economy. US tech firms’ AI spending is estimated to be about 1% of GDP, smaller than recent investment booms including the US shale craze of the mid-2010s and the dot-com bubble of the 1990s. It will create a future well worth its longer-lasting J-curve.” - Pictet Research
Changing Habits (Me, Myself and My AI?) - Hybrid Work’s Permanent Shadow on Transit - Transit ridership in NYC and London has stabilized at roughly 80% of pre-pandemic levels, as hybrid and remote work have permanently reshaped how often people ride the rails.
2 If You Read One Thing Today - Make Sure it is This
Keyana Sapp tells the story of why “Your Power Tools Got Worse On Purpose” by looking at “How TTI and Stanley Black & Decker took the same playbook in opposite directions.” It’s an illustrative journey of the power of entrepreneurship and the inherent fragility of so-called; “Financial Engineering” that has powered the PE wave and the MBA mind melt…Read it here in full:
https://www.worseonpurpose.com/p/your-power-tools-got-worse-on-purpose
Some Takeaways
The two conglomerates
Techtronic Industries (TTI) is a Hong Kong company founded in 1985. They own Milwaukee, Ryobi, and manufacture Ridgid power tools under license from Emerson Electric. They bought Milwaukee from Atlas Copco in 2005 for about $626 million.
At the time, Milwaukee was a respected but mid-tier brand known mostly for Sawzalls and hole hawgs. TTI also owns the floor care brands Hoover, Dirt Devil, and Oreck, though those haven’t gotten the same love.
Stanley Black & Decker (SBD) is the result of Stanley Works merging with Black & Decker in 2010. That merger created a company that already owned DeWalt, and from there they went on a tear. Craftsman from Sears for $900 million. Irwin and Lenox from Newell Brands for $1.95 billion. They absorbed Porter-Cable, Bostitch, MAC Tools, Proto Industrial, and Vidmar.
Over $6 billion in acquisitions since 2002. At one point they owned so many brands in the same category that their own products were cannibalizing each other on the same store shelves.
Both companies bought up everything. What they did next is where the story splits.”
TTI: Buy it, invest in it, leave it alone
TTI bought Milwaukee and basically let it run itself. Kept the R&D operation in Brookfield, WI. Kept the engineering team intact. Dumped $206 million into R&D in a single year. More than 4.4% of total sales going straight back into product development, every year.
The results showed up fast. M12 and M18 launched within two years of the acquisition. Then FUEL brushless motors. Then ONE-KEY, the first digital platform for tools and equipment that lets you track inventory, customize torque settings, and lock a tool remotely if it gets stolen. Then PACKOUT modular storage, which turned a plastic box into an ecosystem. Then MX FUEL, pushing cordless into concrete saws and breakers that used to require a gas engine or a generator.
Milwaukee’s Brookfield campus went from 190,000 square feet to over 500,000. They opened a $100 million campus in Menomonee Falls. A manufacturing plant in West Bend. Facilities in Greenwood and Grenada, MS. In the last five years alone, Milwaukee has invested $368 million in domestic expansion.
Wisconsin headcount went from around 300 in 2011 to over 2,000. Across the US, Milwaukee now employs over 5,900 people. Globally, the company now has over 10,000 employees and generates roughly $8 billion in revenue on its own.
Their portfolio strategy is clean. Ryobi handles DIY at Home Depot. Milwaukee handles pros. The two brands don’t eat each other. They serve different people at different price points with different expectations, and TTI lets each one keep its own identity, its own engineering, its own product roadmap.
TTI did $14.6 billion in revenue last year with $44 million in net debt. The founders still own roughly 24% of the company. Milwaukee grew 11.6% SBD:
Buy it, merge it, cut it
Stanley Black & Decker took the opposite approach.
The 2010 merger of Stanley Works and Black & Decker created a company that already owned DeWalt. From there they went on an acquisition spree that should have built an empire. Instead it built a bloated holding company drowning in debt and leadership turnover.
They bought so many brands they were competing with themselves on the same store shelves, then starved the weaker ones to feed DeWalt.
The tools division has been a revolving door at the top. After the previous head left, two executives served as acting co-presidents before Chris Nelson was brought in from Carrier Corporation in June 2023. Nelson had zero tool industry background. He’d been running an HVAC division. Before that, McKinsey and Johnson & Johnson.
The CEO who hired him didn’t last much longer. Donald Allan Jr. stepped aside in October 2025 after three years in the role, leaving behind a stock price that had dropped roughly 50% from its 2021 peak. Three years, two billion dollars in cost cuts, 7,000 jobs eliminated, and the guy who ordered all of it still couldn’t make the numbers work.
The cost-cutting has been relentless. SBD launched a $2 billion “cost reduction and operational simplification” program. Since late 2023, they’ve cut roughly 7,000 employees globally. Closed plants in South Carolina and Texas. Sold off their aerospace fastener business to Howmet for $1.8 billion in cash. The total workforce dropped from about 48,500 to 43,500 in a single year. Annual filings show $141 million in restructuring charges in 2022 and another $39 million in 2023.
SBD is carrying $6.1 billion in long-term debt.”
Porter-Cable: The saddest story in tools
If the Craftsman failure is the most public, Porter-Cable is the most painful.
Porter-Cable was founded in 1906. They invented the portable belt sander. The helical-drive circular saw. The Speedmatic router that professional woodworkers built entire shops around. In 1996, the Smithsonian Institution collected their company history as part of the American manufacturing record. They were that important.
By the early 2000s, Porter-Cable was widely considered the highest quality line of woodworking power tools available in America. When distributors worked trade shows, a significant chunk of their booth time was spent listening to customers talk about the Porter-Cable tools in their garages. How long they’d had them. How they still ran perfectly. The service centers were staffed by people who could recite 14-digit part numbers for bearings off the top of their heads.
Porter-Cable once discontinued the 126 door plane because the tooling to build them had worn out. After an outcry from the contractor community, they retooled the machines and brought the plane back. It was simply the best tool in the world for installing doors. That’s the kind of company it was. Professional, passionate, invested.
SBD bought them in 2004. The cheapening of internal components started immediately.
According to a former tool industry representative who spent 30 years in the business, the plan was clear from day one. Cheapen the internals to build more profit margin into each unit. Discontinue large portions of the product line, including iconic legacy tools the brand was built on. The service centers closed within roughly six months of the acquisition.
The reps who built their careers on the Porter-Cable name were fired shortly after. One of the best in the industry saw what was happening early, left for Metabo, and stayed there for almost two decades. That tells you the caliber of people SBD pushed out the door.
Router line discontinued. Social media went dark for years. No new product development. You can still find some Porter-Cable stuff at Tractor Supply, but the brand is functionally dead.
A 118-year-old company. Important enough for the Smithsonian to preserve. Reduced to clearance bin filler.”
The ones that never sold
A few companies watched all of this happen and said no.
Klein has been family-owned since 1857. Sixth generation, still private. Their core lineman’s pliers remain the industry standard. Though it’s worth noting that they’ve started slapping the Klein name on a wider range of white-label products, and the quality on some of those newer items doesn’t match the reputation. The legacy tools are still excellent. The expansion products are a gamble.
Makita has been independent since 1915. They’re owned by Makita. They make Makita. No parent company, no conglomerate, no PE firm in the background. Multiple tradespeople swear by their ergonomics, vibration resistance, and longevity. Their batteries are overpriced and the US market gets a fraction of the product line that Japan gets, which is its own frustration. But nobody’s gutting them from the inside.
Knipex is family-owned out of Germany. They make what many consider the best pliers on the planet. Part of a larger group (Knipex Group) but not publicly traded, not for sale.
Channellock is still owned and operated by the founder’s descendants. US-made hand tools. They’ve quietly stayed private while everyone around them got acquired.
Hilti is owned by a family trust and still operates as a family company. Their customer service is legendary. One commenter shared that he bought bolts from them once in 2010 and they still have a dedicated service rep who calls him twice a year. He buys bolts out of guilt now.
Bosch is a private company owned by the Robert Bosch Foundation. Their power tools division is basically a rounding error in their overall business. They make everything from fuel injection systems to dishwashers to automotive sensors. Different beast entirely.
These companies prove the same thing from the opposite direction. You don’t have to get acquired.
You don’t have to take the PE money.
You can just keep making good products and telling everyone else to go to hell.
It’s harder, and it’s slower, and the growth chart won’t impress a Wall Street analyst. But the tools last. And the brand means something 50 years from now instead of ending up in a clearance bin.”
The pattern
This isn’t a tools story.
This is what happens in every industry once the conglomerates and private equity firms show up. Acquire the brand. Consolidate the operations. Cut costs. Extract value. Move to the next one.
The names change. The industries change. The strategy doesn’t.”
3 Consequential Thinking about Consequential Matters
The Pictet Research Institute looks into; “A new frontier of investment opportunities at the crossroads of population change and technological innovation” It’s an interesting exploration worth pondering - go do it in full here:
https://am.pictet.com/content/dam/am-pictet/media/global/investment-research/demographics-and-technology/PictetResearchInstitute_Demographics-and-Technology.pdf
Some Takeaways
“Ageing economies face a stark choice: do nothing and decline or transform and continue to grow. Fortunately, automation and AI can go a long way to counterbalance the shrinkage of the labour force, and the timing of their evolution is opportune.”
“While the drivers of population change are multiple and uneven, one constant holds true: by 2050 the world’s leading economies are all expected to experience a significant increase in dependency, or the ratio between those not working and working. Unlike other economic forecasts that are estimated with a significant level of uncertainty, the probability that projected demographic changes will take place is very high.
In addition, there is not much a country can do to materially alter its demographic course, especially in the span of a couple of decades.”
“The interplay between demographic pressures and technological advancements is expected to reshape the global economic landscape, presenting both challenges and opportunities for governments, businesses and individuals to navigate.
From an investing perspective, it will reshape the investment opportunities across countries and industries, generating new winners and losers.
The sequencing of automation typically adheres to a pattern in ageing economies. First comes the deployment of substitution robots, which directly take the place of workers who are becoming scarce or expensive to employ. Then comes the use of productivity robots, which improve hourly output. In an ageing society, both types of robots have a role to play.
Substitution robots simply sustain production levels with fewer workers, whereas productivity robots have the potential to increase output and generate competitive advantages. Yet worker-replacement robots can be implemented swiftly, while productivity-enhancing systems necessitate additional investments in training, data systems and organisational redesign, which typically take years to yield returns.
The economies that successfully navigate both automation phases are likely to counteract labour shortages with productivity gains, whereas those that remain only in the substitution phase may run the risk of eventual declining competitiveness.”
“For all countries undergoing demographic transitions, the key to the successful adoption of productivity-enhancing technologies fundamentally lies in developing the infrastructure and technology diffusion capabilities required to turn automation and AI into productivity-improving engines.
In the context of industrial automation, advancements in robotics and AI are enabling machines to perform an expanding range of tasks. We estimate that AI will achieve peak productivity gains in the 2030s, well in time to counteract some of the most significant demographic challenges faced by ageing economies.
The timing of AI adoption in each country, in relation to its demographic pressures, can significantly alter its productivity outlook. Countries may face temporary declines in productivity during the early stages of AI adoption, corresponding to the lower part of the AI J-curve, before achieving significant gains.”
“It is effectively an exercise and a venture in turning the developing demographic challenges into a durable competitive advantage.”
“The economic impact of automation relies not only on cost savings but also on demand patterns.
If ageing populations’ consumption patterns favour sectors that can be automated and deliver efficiency gains, this would have a broader positive effect on the country’s productivity and growth dynamics.
If, on the other hand, ageing populations’ consumption preferences favour sectors that cannot be automated, the economic outlook for the corresponding economies could be bleak.
Understanding how consumption patterns evolve as populations age is crucial for identifying investment opportunities for the years to come.”
“As labour becomes scarcer and more expensive in ageing societies, businesses have greater incentives to invest in technologies that can substitute human labour and/or enhance productivity.”
“China’s situation is particularly arresting. It is projected to be the country most affected by ageing, with the population expected to halve by the end of this century due to its sustained low fertility rate. What’s more, China’s elderly dependency ratio is projected to surpass 100% by 2080, meaning there will be more people aged over 65 than those aged 15 to 65.
Other countries, led by Canada, are set to see their populations rise through 2050 thanks to immigration, assuming past immigration trends continue.
Technology can counterbalance the economic consequences of these demographic changes. Indeed, as labour becomes scarcer and more expensive in ageing societies, businesses have greater incentives to invest in technologies that can substitute human labour and/or enhance productivity. This innovation dynamic is particularly relevant in the context of industrial automation, where advancements in robotics and AI are enabling machines to perform an expanding range of tasks and render existing labour more productive.
The experiences of Germany and Japan show how this dynamic is already playing out in different ways in different countries. Germany’s rapidly ageing population requires immediate worker replacement, whereas Japan’s earlier demographic shift has enabled a more mature emphasis on enhancing productivity.”
“The key to success will not lie in merely deploying the most robots, but in developing the organisational capabilities that turn individual machines into significant efficiency improvements.
As AI takes automation beyond manufacturing to cognitive tasks, grasping these sequential patterns will be essential for navigating the broader economic transformation and investment opportunities on the horizon.
What is more, we believe that rather than leading to stagnation, demographic shifts coupled with automation technologies may serve as a catalyst for productivity growth.”
“…identified demographic change as a key “headwind” to productivity and labour force participation. In contrast, Acemoglu and Restrepo (2017) challenge these theories, revealing that ageing does not necessarily correlate with declining GDP per capita. Instead, the countries hit hardest by ageing are the ones leading the automation adoption process, suggesting that technological adaptation is a critical factor in offsetting the economic pressures of demographic shifts.”
“Using data from the International Federation of Robotics (IFR), Acemoglu and Restrepo (2017) demonstrate that between the early 1990s and 2015, countries experiencing faster demographic ageing – as measured by the growth in the ratio of individuals aged 50 and older to those aged 20-49 – adopted industrial robots at significantly higher rates.
For example, Germany, Japan and South Korea, which are among the most rapidly ageing societies, are leaders in robot adoption. Even within the OECD, a strong correlation exists between ageing and robot adoption. The link between ageing and automation is not merely coincidental; it is underpinned by the principles of directed technological change, as discussed in Acemoglu and Restrepo (2022). As labour becomes scarcer and more expensive in ageing societies, companies have stronger incentives to invest in technologies that can replace workers. This dynamic is particularly relevant in the context of industrial automation, where advancements in robotics and AI are enabling machines to perform an expanding range of tasks.
Acemoglu and Restrepo’s (2022) model shows that labour scarcity can drive innovation in automation technologies, leading to productivity gains that offset the negative effects of demographic change. In this framework, the adoption of automation is not only a response to current labour shortages but also a forward-looking strategy to address anticipated demographic trends. Importantly, Acemoglu and Restrepo (2022) highlight that the economic impact of automation depends on the relative abundance of capital. In capital-abundant economies, where the cost of capital is low, the adoption of automation technologies is more likely to lead to productivity gains and increased output.
Abeliansky and Prettner (2023) propose an alternative model to tackle the same question and reach similar conclusions. Empirical evidence shows that countries with greater capital availability and higher levels of ageing have been more successful in integrating automation into their economies. For example, Germany’s leadership in both robot production and adoption reflects its ability to leverage technological innovation to counteract demographic pressures. This is particularly true in industries with high automation potential, such as motor vehicles, electronics and chemicals, where robots are increasingly performing tasks that were once labour-intensive.”
“Historical trends in technology adoption offer valuable insights to determine where robotics and AI are in their diffusion cycles and when we can anticipate their maximum economic impact.
The global economic order has consistently been shaped by a few general-purpose technologies that are so transformative in scope and have such cascading effects that they fundamentally reshape markets, competencies and even demographic patterns.
From electricity and the internet to industrial robotics and today’s AI platforms, these general-purpose technologies differ markedly but all share a common trajectory: each begins as an elusive frontier technology before institutional alignment, cost structures and skill acquisition converge to enable diffusion across sectors and borders, ultimately becoming the foundational infrastructure of production (Comin & Mestieri, 2014; Stokey, 2021).
Companies and governments adopt general-purpose technologies for diverse, sometimes conflicting objectives such as cost optimisation, strategic autonomy or social inclusion. Yet, once diffusion reaches critical mass, the technology unifies these diverse motives into a cohesive growth trajectory.”
“The narrative of each technological wave is less about sudden creative destruction (the process in which new innovations replace and make obsolete older innovations) and more about the gradual coordination of multiple stakeholders – engineers, capital providers, regulatory authorities and end-users – whose economic incentives progressively converge around shared adoption frameworks:
The electric dynamo initially illuminated urban transport networks before expanding to dispersed agricultural operations.
Packet-switched communication protocols emerged from military research infrastructure and later became the backbone of modern digital payment systems.
Industrial robotics first gained momentum within Japan’s demographically constrained motor vehicle manufacturing sector before spreading across mid- tier European production facilities.
Today, the deployment of AI is concentrated among well-capitalised multinational corporations while spreading to critical applications in public health diagnostics and smallholder agricultural risk management.
This pattern repeats with remarkable consistency across decades and technological domains, offering a roadmap for understanding how today’s emerging technologies might be diffused.”
“Timing disconnect – where initial investments temporarily depress measured efficiency before generating substantial productivity gains – would become a defining feature of all subsequent technological waves.”
“The profound economic impact of the internet arose not from its technical prowess but from how quickly and widely it was able to spread thanks to self-reinforcing network effects, where more users attract more users.”
“Industrial robotics has progressed through various developmental phases and is currently transitioning from Market Scaling & Augmentation to System Integration & Transformation. The technology began its early development phase in the 1980s and 1990s with applications in automotive welding. It then achieved market scaling through collaborative robotic systems and is now nearing full system integration with just-in-time production systems.
“In robotics, scaling has accelerated due to the formation of regional ecosystems around key manufacturing facilities. Robot deployment densities in Japan, South Korea and China now exceed 40 per 1,000 workers, nearly quadrupling the global average (IFR, 2024a). Productivity improvements have followed the established trajectory, with cross-country manufacturing data indicating that increased robot density contributes approximately 0.4% to annual GDP growth.
Notably, robotics seems to have avoided the significant J-curve dip experienced by electricity and the internet thanks to the knowledge and experience gained from previous waves. This framework offers a structured method for identifying where various technologies are within their productivity cycles and determining when coordination benefits – and therefore the biggest growth opportunities – may arise.
Unlike previous technologies, the productivity dip associated with AI arises from significant intangible investments in data infrastructure, organisational restructuring and workforce retraining, which temporarily outweigh measurable returns. Companies are currently incurring substantial costs while facing integration challenges and learning curves. This has resulted in a brief but intense disruption period where cognitive work processes are being redesigned more quickly than workers may be able to adapt to, leading to temporary potential productivity declines despite considerable technological advancements.
However, the recovery phase is expected to be of similar steepness as that of robotics, due to AI’s network effects and the knowledge and experience accumulated during the disruption period and the diffusion of previous related technologies.
AI is currently entering the early scaling phase. The infrastructure required for the augmentation phase has only emerged in the last decade: affordable cloud computing capacity, extensive datasets classified for use by AI (labelled datasets) and computational capacity (GPU-accelerated training architectures).”
“Technology-intensive sectors, such as Motor and Computer & Electrical, use robots mainly for precise welding and assembly, where human error can lead to significant defects. They have a strong focus on productivity enhancements underscoring the complementarity between labour and technology, regardless of demographic developments.
Labour-intensive sectors, such as Textiles, Wood and Food, show considerable variation across countries and typically favour substitution applications. Robots in these industries do heavy lifting and repetitive tasks, addressing pressures from ageing workforces. Countries facing labour shortages have a greater tendency towards automation, even in traditionally manual sectors.
Hybrid sectors, including Chemicals and Pharmaceuticals or Basic Metals & Machinery, use varying levels of automation. The reasons may be country-specific or related to the segment within the industry they are concentrated on.
The above taxonomy suggests that while there may be room for further automation and productivity gains across most industries, the biggest incremental productivity gains may be made with the use of automation in hybrid industries that have underinvested in technology. As a rule, the more precise the work and the higher the cost of errors, the higher the incentive for companies to invest in automation to control costs and remain competitive. In addition, the demographic factor adds further emphasis on automation by inducing companies to use the ever scarcer labour resources more productively.”
Lessons from Japan
Japan’s earlier struggle with deteriorating demographics led it to embrace automation from the late 1990s, when labour shortages in the Furniture and Transportation sectors became acute, as Kushida (2024) discusses.
Companies needed machines to fill positions they were unable to staff. By 2010, most straightforward substitutions had been automated. Japan then transitioned to a second phase: complex electronics manufacturing, where collaborative robots and AI-powered vision systems assist ageing workers in enhancing their performance rather than replacing them entirely.
While Japan has continued to automate, reaching 100% of its potential in sectors such as Computer & Electrical, other countries are still in earlier stages of automating their production. Using the highest level of automation already achieved in Japan, US, Germany and China for the four most automation-conducive industries – Motor, Computer & Electrical, Basic Metals & Machinery and Chemicals & Pharma. It is instructive to look at these industries as not only the most conducive to automation, but also as the three largest contributors to global GDP, with the Motor industry being the fifth.
“In China, average 10-year productivity growth stood at around 6.5% annually while its GDP growth was just shy of 6%. This shows that demographic and employment factors affected Chinese GDP growth negatively. Even though the impact of demographic change is expected to be around -1% annually, productivity growth in China is still very high and may grow further if automation is more widely used, showing that China still has a reasonable buffer to sustain positive GDP growth in the years ahead, albeit likely at lower levels than previously.”
“We do not yet know the full effects that AI could have on the services sector – a significant component of GDP, especially in developed economies. What we do learn from this study, however, is that the more automation a sector uses and the more it caters to an ageing population, the greater its growth and productivity potential, and therefore its profitability, so long as the specific economy has the infrastructure in place for a particular technology to be adopted and achieve its potential productivity gains.
Furthermore, taking into account country differences in consumption preferences among ageing populations may provide useful guidance as to the geographies where a particular investment idea may find most fertile ground.”
“The more automation a sector uses and the more it caters to an ageing population, the greater its growth and productivity potential, and therefore its profitability.”
“We identify three factors that need to converge for a promising investment opportunity to arise. Specifically, it needs to:
1. cater to the demographic shifts in a particular economy or geography;
2. be in an industry that is conducive to automation and AI technologies; and
3. be developed in an economy with the necessary infrastructure for those technologies to achieve their full productivity potential.
These factors provide a new lens for evaluating investment themes and opportunities and reinforce once more how the changing world around us requires a changing approach to investing. In this transformational environment, we need to adjust our investment framework away from broad country or sector bets and towards opportunities where demographics, innovation and infrastructure forces align.
Adapting to morphing population dynamics will require countries and industries to plan strategically and take well-timed policy and investment decisions.”
“In this transformational environment, we need to adjust our investment framework away from broad country or sector bets and towards opportunities where demographics, innovation and infrastructure forces align.”
4 Big Ideas
ChinaTalk takes a loom at ‘The Empire of Wuxi’ and the evolving model of the global pharmaceutical supply chain - a key input in our economies and lives - go explore it here in full…
Some Takeaways
When we say “WuXi,” we don’t just mean WuXi AppTec. Although this family of companies is often spoken about as if it were a single company, in reality, it is a group of companies comprised of WuXi AppTec (药明康德), WuXi Biologics (药明生物), and a set of tightly integrated businesses, all more or less under the same leadership but dispersed throughout the industry. Together, they are stronger than the sum of their parts, and form what we envision as the Empire of WuXi (hereafter just “Wuxi”).
The TSMC analogy is tempting, since just as TSMC manufactures chips for companies like NVIDIA and AMD, WuXi, instead of discovering and commercializing its own blockbuster drugs, it provides the services (chemistry, testing, manufacturing) that allow others to do so. And both have the ability to gut-punch the global economy if their employees stop coming to work.
But AI analogies, tempting as they are, can do more harm than good. TSMC sits at a true chokepoint, with essentially no major rivals. If you want cutting-edge chips, you go through Taiwan. But WuXi does not monopolize a single irreplaceable step in the biotech supply chain. In fact, it has strong competitors both in China and globally.
WuXi AppTec and WuXi Biologics are the third- and fifth- largest contract development and manufacturing organizations (CDMOs) in the world by revenue. The remainder of the top ten are all based in U.S. partner nations, including the top two of Lonza (a Swiss company) and Catalent (a U.S. company). So, if there are plenty of alternative companies in U.S.-aligned nations, why is WuXi such a bogeyman for the U.S.?
In the same way that China’s rare earth stranglehold matters because of where those minerals sit in critical supply chains, WuXi, with its unique corporate structure, is embedded at many layers of the biostack. It has accumulated a structural indispensability that is harder to replace than a single dominant manufacturer would be.
A 2024 survey by the Biotechnology Innovation Organization estimates that 79% of US biopharma companies have at least one contract with a Chinese CDMO or CMO. WuXi AppTec alone is estimated to be involved in roughly a quarter of all drugs used in the United States (according to WuXi). And an estimated 65% of WuXi AppTec’s total revenue comes from U.S.-based clients.
Even if the U.S. and its allies lead in certain sectors of biotech, the growing recognition that WuXi has embedded itself throughout the supply chain has raised concern about systemic dependency and the leverage that comes with it.
The U.S doesn’t have an easy way to address this. China’s specific advantages in biotech look less like control over a single node and more like what it achieved with its manufacturing sector. It is about process expertise, cost efficiency, labor and talent, and deep integration into global supply chains — perhaps more like BYD’s success in the EV sector. These are not easily reducible to export-controllable chokepoints.
A key inflection point for WuXi is the 2015 reform of China’s drug review and approval system. By decoupling drug approval from manufacturing and encouraging outsourced production, the reforms accelerated a feedback loop: more innovative drugs → more R&D → more outsourcing → more innovative drugs, and so on. WuXi expanded aggressively to meet that demand and become the titan it is today, including earlier moves like its 2008 acquisition of a U.S.-based AppTec business, which gave it both new capabilities and a physical foothold in the American market (and the name of its most famous company, WuXi AppTec).
For years, this dual positioning was an asset. The intertwinement of the U.S. and Chinese biotech systems was not accidental but foundational to WuXi’s rise. Western pharma outsourced to China for cost and scale; Chinese firms like WuXi grew by serving those needs. You could argue that this was exactly the outcome the U.S. wanted before it realized how powerful China would become.
Li’s vision for WuXi’s role in the pharma business ecosystem was explicit from early on. WuXi was not meant to be a traditional drug company, but an enabling platform for global innovators. Rather than designing drugs, they would build the infrastructure needed to quickly find and develop them. The novelty of this business model was not simply exploiting wage arbitrage — U.S. and European pharmaceutical companies already knew how to outsource chemistry. Instead, Li’s key insight was to reframe the role of contract R&D in the drug development process.
Traditionally, outsourcing drug companies would partner with different contractors for each step of drug development. WuXi provided an enticing alternative. Instead of contracting one company to test the initial drug, another to optimize its potency, and another to manufacture it at commercial scale, drug companies could work with WuXi through the entire pipeline.
Li would later define this approach as an “open-access platform” (开放式平台). Unlike more siloed competitors, WuXi was committed to “following the molecule” as it progressed from the research laboratory to regulatory approval and commercialization. This business model would later be codified as a “contract research, development, and manufacturing organization” (CRDMO) and copied by other companies.
This approach is a win-win for both parties. For the drug developer, it minimizes the need to switch between different corporate ecosystems, eliminating the inefficiency of juggling multiple contracts and ensuring each partner is up-to-speed. For WuXi, it incentivizes customers to stay “stuck” to their services for years, leading to predictable business and access to the revenue scaling that occurs as the drug progresses towards commercialization. Given the immense uncertainty involved in pharmaceutical development, this level of stability for provider and customer is extremely attractive.
WuXi doubles down on this model by targeting a “long tail” of biotech customers. Rather than limiting themselves to massive deals with the pharmaceutical giants, they target many small- and medium-sized firms. With more limited resources, these small companies benefit particularly from the cost efficiency of WuXi’s end-to-end services, which then locks them into the pipeline. Their sheer number and diversity also diffuse the risk of major damage from any one customer pulling out. Furthermore, research by consultancy firms has shown that these smaller companies tend to produce more innovative drug leads than their big pharma counterparts. WuXi is therefore able to link itself to these disruptive — and therefore lucrative — products early on. These strategic decisions have given WuXi a “strong, diverse, and sticky customer base.”
Does WuXi have a technical moat?
Importantly, however, these technologies didn’t originate from WuXi labs. So, unlike the TSMC analogy, there is not a WuXi-specific technological moat around their services. Instead, WuXi’s biggest competitive advantage lies in their integration across the technology stack.
Indeed, a quick scan of their advertised capabilities reads like a catalog of the hottest frontier capabilities in drug development. A company can use WuXi’s DNA-encoded libraries to quickly scan for usefully potent molecules, including with options to avoid sharing IP. Biomanufacturing for complex biologics has been standardized and optimized, with new methods being deployed to further boost productivity at scale. In-house expertise in finicky drug types like peptides (including GLP-1s), antibody-drug conjugates (an expanding class of mainly anticancer drugs), and monoclonal antibodies (of COVID-19 treatment fame) expands the customer base they can serve. And, of course, AI and automation are being deployed throughout the pipeline.
Most biotech and pharmaceutical firms lack the resources and expertise to deploy these advanced biotechnologies in-house. But WuXi’s comprehensive and integrated platform offers them the access and support needed to compete at the technological frontier. A positive feedback loop is born as WuXi aggressively invests in further optimization and expansion, and the platform becomes even more attractive to the next wave of ambitious firms.
An excellent example of WuXi’s ability to adopt and deploy new technologies is their development of the “scale out” paradigm for manufacturing biologic drugs.
The most critical advantage is the Chinese workforce. Chinese universities produce dramatically more STEM Ph.D. graduates than their U.S. counterparts. WuXi capitalizes on this geographic concentration with targeted training programs that attract top candidates and develop company-specific skills. WuXi also invests in training workers at every level of the production process, including the technicians and operators running factory floors. This is precisely the kind of vocational and technical workforce development that the U.S. has chronically underfunded and undervalued. Because this highly skilled Chinese talent is often half the cost or less than Western equivalents, companies like WuXi can deploy larger teams to shorten timelines and overcome obstacles.
WuXi also benefits from China’s established excellence in advanced manufacturing. Because China largely controls global production of raw materials and active ingredients for small-molecule pharmaceuticals and is rapidly domesticating the supply chain for biologics, domestic companies benefit from easier sourcing and more resilient supply chains. This colocalization directly translates into accelerated procurement and lower overhead costs.
These advantages are compounded by the central government’s aggressive championing of biomanufacturing, such as labeling biomanufacturing a national priority and doling out subsidies.
The U.S. is quite concerned that China is “catching up” in biotech despite spending far less on relevant R&D:
No single country is waiting to absorb WuXi and China’s cheap and diffuse biotech role. India has a large base of FDA-approved facilities, competitive costs, salaries about half of China’s, and a large and growing Ph.D. pipeline. It has thus received a surge of inquiries from U.S. pharma eager to diversify away from China. However, most of India’s strength is concentrated in small molecule generics, a very different skill set from the complex biologics manufacturing that makes up so much of WuXi’s value. South Korea’s Samsung Biologics is strong on biologics (rivaling WuXi Biologics), but weaker on the small molecule CRO and chemistry services where WuXi AppTec has built its deepest moat. No single country or company can replace all of the different roles WuXi plays, but if the U.S. leveraged its multilateral relationships to build a coordinated alternative across trusted partners, that would be its best shot, something Trump 2.0 has moved against.
The uncomfortable truth is that a U.S. biotech industry fully decoupled from China would be a slower and more expensive one.
5 Big thinking
Voltaire is often listed among the leading Enlightenment thinkers, alongside Montesquieu and Locke. Or he is mentioned among famous playwrights, such as Molière and Racine. But it’s unlikely to see Voltaire listed among successful entrepreneurs. Yet at 76, he founded a startup and turned it into a successful international business. Ivo Velitchkov of the ‘Link & Think’ Substack explores the story and it’s lessons here:
Some Takeaways
“Voltaire wasn’t planning to start a business, nor did he need to. His interests were in writing, theatre, and political causes such as penal reform. He had plenty of revenue streams; he didn’t need another. However, his experience in financial management and the circumstances under which the idea was born might make the endeavour less surprising than it first appears.”
“The watchmaking wasn’t Voltaire’s first manufacturing startup. A few years previously, he set up a small silk business. Since he was previously engaged in agriculture, literally cultivating his garden, a statement he famously ended Candide with, it seems he just moved to the adjacent possible. (…) Voltaire’s silk business started with raising silkworms, but did not stop at producing silk. He also produced silk stockings.”
“It so happened that the French government was facing financial difficulties and couldn’t begin the promised investment in Versoix.
Voltaire realised that many of the emigrants were skilled craftsmen. So, while waiting for the Versoix to be built, he can help them start a business as independent watchmakers.
That’s how the Ferney startup took off as something between a social enterprise and a business incubator.”
“Voltaire established the watchmaking business at Ferney as a social enterprise. Without losing its role as such, the startup grew into a successful business selling watches in more than eight markets and, at its peak in 1776, generated revenue of 600K livres annually, equivalent to ten million euros today.”
“These extraordinary results were almost entirely due to Voltaire’s personal efforts, for he had reinvented himself in a protean variety of roles: not just the overall manager, co-ordinator and organiser but also the financier, the virtual bank manager, the sponsor, the builder of homes and factory space, the buyer of precious metals and other raw materials and the international sales manager.”
“Voltaire realised that businesses are most vulnerable at the outset and created the most favourable conditions for production: housing for watchmakers, tax exemptions, and interest-free financing, which he supplied himself.
He turned the theater in his house into a watchmaking workshop. In a letter to the marquis de Jaucourt, Voltaire wrote:
Our theatre auditorium, which you remember, has been transformed into workshops. There, where we once recited verse, we are now melting gold and polishing cogs. We must build new houses for the emigrants. All the workers of Geneva would come here if we were in a position to house them. We must remember that everyone nowadays wants a gold watch, from Peking to Martinique, and that there used to be only three great manufacturing centres, London, Paris and Geneva.”
“If Voltaire has to be listed, not only among prominent Enlightenment figures and famous playwrights, but also among successful pre-industrial entrepreneurs, who else will belong to that list?
I would nominate Jakub Fugger and Josiah Wedgwood. Although the three of them differ in the size of their businesses, wealth, and business models, they are comparable in how they contributed to the development and operation of the modernity machine.”
6 Treat Possibilities as Probabilities
Have a Great weekend when You get to that stage,
Sune