Weekend reading links

1. Indian electronics contract manufacturers are unwilling to move up the value chain.

Indian EMS players have chased scale 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. Courtesy of these, India made up around 3.5% of the world’s total electronics output in 2024. But what about those who’ve already hit a certain scale? For instance, 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...

Today, India is one of the world’s top five electronics manufacturers, and the second largest manufacturer of smartphones. But the creation of a holistic electronics system design and manufacturing (ESDM) ecosystem, the original intention of the PLI scheme, is still a distant dream. That’s because both design and component manufacturing remain in their infancy... 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.

This is a cautionary note for India's electronics components manufacturing scheme.

Essentially, while suppliers in India sell to specification, Chinese suppliers would co-design... China’s manufacturing, R&D, and component-development ecosystems sprang up together rather than sequentially... The government’s electronics-component-manufacturing scheme is certainly a step in the right direction, but its progress will hinge on whether firms build design and testing capability alongside manufacturing; rather than simply replicating the assembly model at the component level.

This is a good read on this topic with several links.  

2. Nvidia's latest chips are finding their way to China despite a Commerce Department review gate to assess for national security risks. 

Within China, there is little evidence the firm actually exists, raising questions about who in China is behind the company. Only scant information can be found online about New Matrix, which was created 17 months ago. At the building that New Matrix lists as its address, a company directory does not include the firm. At the nearby property management agency for the industrial park, a receptionist and a manager both said they had no record of the company. According to Chinese corporate records, New Matrix’s owners have ties to a prominent Chinese private equity fund that the U.S. government placed on a trade blacklist in 2024. Two analysts who examined the corporate records and other Chinese-language information said they believed New Matrix may be a front for firms associated with that fund. The sale of chips to New Matrix appears to be part of a pattern that has concerned technology experts inside and outside the government... Corporate records show that several other Chinese companies approved to purchase H200 chips or assemble them into servers have ties to firms that have been blacklisted by the U.S. government, or that work as military contractors in China... the approvals raised questions about how rigorously the Trump administration is policing who can get their hands on Nvidia’s chips.

3. The numbers behind India's data centre boom.

Operational capacity should touch about 2 gigawatt (Gw) by December, and most forecasts for 2030 sit between 5 and 8 Gw... The building, with its power and cooling, costs ₹50-70 crore per Mw, or $5-7 million, depending on whose estimate you prefer. The chips cost several times as much. By Jensen Huang’s own reckoning, Nvidia’s share of a Gw is about $25 billion for Blackwell and $40 billion for the coming Rubin generation: $25–40 million per Mw, counting only what Nvidia sells. The building will last 25 years. The chips run the risk of looking dated in a few years.

Taking India from 2 Gw to 5-10 Gw means building another 3-8 Gw. At ₹60-70 crore a Mw, that is ₹2-5 trillion by 2030, before anyone builds the power plants to run them. If 60-70 per cent is debt, banks are looking at ₹1.2-3.5 trillion of fresh lending. Put that next to what banks already hold. Bank credit to all of Indian infrastructure was ₹15.1 trillion in June 2026. The buildings alone could add 8-23 per cent to that book in four years, somewhere between a fifth and more than half of what banks now add to infrastructure lending each year. The solar farms and transmission lines behind them come on top... If even one Gw of India’s build-out is filled with chips owned by Indian companies, that is $25-40 billion, or ₹2.4–3.8 trillion.

4. Amidst the stalemate in the Gulf, a little known South Korean shipping company, Sinokor, has become the world's largest owner of supertankers.

Sinokor, has accounted for about a half of the tanker transits through the strait since the beginning of September... In an exceptionally well-timed bet, Sinokor, led by a father and son who are little known even in South Korea, began a buying spree in the two months before the United States and Israel attacked Iran at the end of February. It is now, by some counts, the largest operator of supertankers in the world... Sinokor owns 129 very large crude carriers — the industry name for tankers capable of carrying two million barrels of oil — up from 42 in December, according to BRS Shipbrokers... 

Last year, the top three supertanker owners controlled 16 percent of the market, compared with 27 percent today, said John Cotzias, a co-founder of Xclusiv Shipbrokers, a Greek firm. Veson Nautical, a maritime analysis firm, estimates that Sinokor spent $5.6 billion buying 70 tankers since the start of this year. Sinokor now owns 14 percent of the roughly 940 supertankers in operation. The next-largest owner and manager of very large crude carriers, China Merchants Group, controls about 5 percent of the market. All along, shipping rates are rising. Tankers chartered for the 15-day run between the Persian Gulf and the Gulf of Oman have surged past $1 million per day, up from about $120,000 a day last year, Mr. Cotzias said.

5. Excellent primer on the automobile industry.  

6. Vokswagen taught Chinese automakers, and is now being eased out of that market.

In 2020, global brands had close to 75 per cent of the Chinese market of 25.2mn cars. ICE vehicles, which accounted for more than 95 per cent of sales, are now barely a third of the market. This is set to fall to a quarter of the market of around 20mn cars. While Chinese carmakers’ China sales volumes were flat in the first nine months of 2026, those of German groups dropped 30 per cent while Japanese groups fell 25 per cent, according to data from Automobility, a Shanghai consultancy.

Chinese car exports are now predominantly EVs.

7. Pricing power of chipmakers

8. The Lenskart story in numbers.

Lenskart opened some 600 stores in India and roughly 80 abroad, close to two a day between 30 September 2025 and 30 September 2026. It now runs 2,868 stores in India and 756 overseas, a total of 3,624. That puts the company ahead of Specsavers, the British chain that has long called itself the world’s second-largest optical retailer and has 2,987 stores. Only Essilorluxottica, with around 18,000 stores, is bigger... Over the same period, its closest rival in India, Titan Eyewear, closed 33 stores... Lenskart’s oldest backers, including Softbank, Alpha Wave, Temasek, and others, have sold Lenskart shares worth Rs 18,500 crore, making between 2.5X and 7X on their investments. SoftBank alone has sold more than Rs 7,500 crore, Alpha Wave Rs 5,000 crore and Adia Rs 4,000 crore... Partial exits worth Rs 18,500 crore by key investors could have set off panic in other stocks, as they did earlier in the cases of Delhivery, Paytm, and Nykaa. Lenskart shares, however, are trading nearly 85% above the listing price.

9. Good description of how private capital built India's infrastructure in the 19th and early 20th centuries.  

On a spring afternoon in 1853, a 14-carriage train pulled out of Bombay, carrying 400 passengers to Thane, 21 miles away. It is remembered as the birth of the Indian railways... that train was designed, financed and run by a private company... In 1849, the government offered private companies a guaranteed 5 per cent annual return on the capital they invested to build the railways, backed by free land and licences running up to 100 years. It was, in effect, an early and appropriate instrument for derisking a market that London investors knew nothing about. The confidence this created was remarkable. British capital flowed into the East Indian Railway (1849), the Great Indian Peninsula Railway (1849), and the Madras Railway (1853). Oversight in this era came from contracts. Each company’s terms were negotiated separately, with the government approving plans and regulating fares... The results were remarkable. From a single 21-mile line in 1853, the network grew to roughly 9,000 miles by 1880. By 1912, the capital invested in the Indian railways stood at roughly ₹477 crore... if compounded at 7.4 per cent a year (current 10-year G-Sec rate) over 114 years, would be about ₹16.3 trillion in today’s terms.

10. The Nvidia ecosystem that Jensen Huang built.

Lessons from China on administrative reforms for India

A negative feature of bureaucracies is the tendency to pass the buck, avoid decision-making, and indulge in formalism at the cost of outcomes. The silence of counterfactuals makes it difficult to measure, evaluate, and create accountability. China is now waging a war on this, and it has lessons for countries like India. 

The FT has a long read on President Xi Jinping’s war on corruption that has seen large numbers of officials being purged and punished. The numbers are staggering.

Beijing tends to measure the progress of the battle against corruption by the number of officials disciplined. By that score it is a raging success. Nothing like it has been seen since Mao’s rule, according to Minxin Pei, professor of government at Claremont McKenna College in California. About 7.2mn Communist Party members have received some form of political, administrative or judicial punishment, Pei calculates. These include at least 632 “tigers” — senior officials at or above vice-ministerial or deputy governor ranks. Over the 11 of Xi’s 13 years of rule for which data is available, about 628,000 cadres have been criminally prosecuted and have probably received jail terms. (Generally, far fewer than 1 per cent of prosecutions in China fail to end in conviction.) Those convicted are thought to represent a sizeable portion of China’s 1.69mn jail population as of 2024, Pei says…

This war on corruption extends to bureaucratic lethargy and apathy. 

In January this year, the Chinese president delivered a speech behind closed doors to the CCDI, the internal investigation body, signalling rising exasperation with top officials’ poor delivery of his policies… he lectured senior officials on the need for vigilance against “two-faced” cadres who, he said, “take many forms”. Xi listed six manifestations of duplicity. These include people who “quickly and loudly” expressed their commitment to carrying out the central committee’s plans only to “inwardly resist” and disobey orders. Then there are those, he said, who are “ideologically liberal”, adopting his doctrine — officially dubbed Xi Jinping Thought on Socialism with Chinese Characteristics for a New Era — “only when it suits them”. A further type is “factional” and pursues the interests of their small cliques. Others are concerned only with self-interest, or shift “with the political winds” or take a “laissez-faire” attitude towards their responsibilities, allowing problems to “fester and spread”. Xi said all six types were “extremely harmful to the party’s cause and image” as he urged the commission’s members to “resolutely eliminate them”.

The article also points to the perils of such disciplining in terms of distorting incentives within the bureaucracy. The fear of disciplining can have a paralysing effect on the bureaucracy and result in behaviours that seek to maintain the status quo and avoid decision-making. 

It is remarkable that the Chinese bureaucratic performance management goes beyond the traditional performance measures and covers areas involving behaviours and attitudes. Accordingly, there is a category called “formalism and bureaucratism” in the country’s bureaucratic conduct rules. Formalism is considered any activity that produces a record rather than a result — documents, meetings, inspections, and photographs of compliance. Bureaucratism is detachment and non-performance. 

It comes from the four work-style failings Xi Jinping named in 2013 — formalism, bureaucratism, hedonism, and extravagance. They are monitored and enforced under the Eight Point Regulation, and violations are reported monthly by the Central Commission for Discipline Inspection (CCDI). 

The CCDI’s monthly reporting groups the two under four heads — failure to act on duties, covering neglect, inaction, and improper or false action; hollow compliance, where orders are nominally accepted but not implemented, with “grand statements and little follow-through”; unresponsive service, meaning passive, cold or rigid conduct toward the public; and the return of excessive documents, meetings, inspections, assessments and traceability requirements that burden subordinate units. It is a remarkable leap of faith that photographs as proof of work, study notes, login duration, likes, votes, shares, and installation rates are each named and prohibited as evaluation criteria. Imagine any government level in India biting the bullet and issuing a direction on these lines. 

In January 2025, cases of inaction, misconduct, and falsified action alone made up 88.1% of all formalism and bureaucratism cases investigated that month.

The “three distinctions” (san ge qufen kailai), formulated in 2016, separate mistakes made in pursuit of reform from deliberate wrongdoing, acting for the public good from seeking personal gain, and exploratory error from knowing violation. These are operationalised through rongcuo jiucuo, fault-tolerance and error-correction mechanisms, with local implementation. It also outlines the exclusions — no protection where discipline or law was broken, or where serious safety, pollution, or public-order consequences followed.

In August 2024, the Party and State Council, with approval of the Politburo no less, issued China’s first intra-Party regulation on easing burdens at the grassroots, consisting of 21 articles in seven chapters, covering control over the number of documents and meetings, clearer delineation of what lower levels are and are not responsible for, and limits on data collection and redundant reporting. 

The details on streamlining documents and meetings, rationalisation of mobile Apps, and guidance on inspections and checks resonate strongly with the Indian context. In fact, here is a table that captures the Indian equivalent in terms of actions/omissions.

These behaviours are the predictable output of an accountability system that can observe activity but not results. As a framework, we can categorise any official’s work in terms of how much is visible and what its outcome is. 

This framework exposes the fundamental problem. On the left, both the performers and ritual compliers have the traces of activity, and are scored equally since the outcome is not measured. Similarly, on the right, the officer who quietly solves a problem without generating paper is scored the same as the one who does nothing. 

In the circumstances, ritual compliance becomes the rational strategy. It is cheaper than performance, scores the same, and carries none of the risk.

I can think of three drivers of such an accountability system. First, an asymmetric sanction means that an error of commission is visible, attributable, and punishable, whereas an error of omission is invisible and, to the individual, costless. This problem is exacerbated since a decision taken in good faith can be reopened a decade later by an investigative agency or an audit paragraph, with the benefit of hindsight and facts nobody held at the time. 

Second, a measurement asymmetry means that measurable dimensions distort incentives. Since what is measurable in government is almost entirely process — files disposed, meetings held, inspections conducted, targets reported — the stronger the incentive, the more effort flows to the record and away from the result. Third, officials face an overload of responsibilities (and targets) than their staffing, financing, powers, and capabilities permit. When everything cannot be done, what is counted gets done, and the rest is reported as done. 

So what can India do? 

In the circumstances, I can think of a tiered performance management ladder. 

Four things can be done immediately to operationalise the above in the Indian context. 

The simplest and essential step would be to adopt an operational classification of non-performance — inaction, delay, buck-passing, ritual compliance, target ratcheting — and report against it. The CCDI’s Eight Point Regulation outlined above is a good reference point. India currently does not have an equivalent to the Eight Point Regulation that identifies and categorises the kinds of bureaucratic conduct of relevance. This is a glaring gap and is the lowest-hanging fruit for India, and can have a significant impact on the efficiency and effectiveness at the frontline of public service delivery. 

Second, a one-time standing burden reduction instrument, like that done by China in the August 2024 regulation on easing burdens at the grassroots, is another low-hanging fruit. Like there, it could cover administrative data collection, reporting, and clarity on what are not lower-level responsibilities. It is important to note that this regulation binds the principal, not the agent, and almost every provision constrains what a superior body may demand (how many documents it may issue, how long its meetings may run, how often it may inspect, what data it may ask for again). It can be done as a time-bound mission for 3 months and then issued as a Government Order. 

Three, build the omission dashboard from e-Office data to detect decision latency, pendency ageing, referral loops, and first-time-right rates, published at the office level. This might require some tweaks and additional fields to the e-Office software, but it can be done with limited administrative effort. This information can be used for internal reviews that shine light on the counterfactuals and to create accountability. 

Fourth, issue a clear directive on safe harbour that offers protection for bona fide decisions, a vigilance-intake filter, and a limitation on reopening old policy and commercial decisions. This should illustrate and guide by listing the commonly observed scenarios and situations.

It is important to start by avoiding using these for high-stakes performance appraisals. At most, they can be indicated in the performance appraisal reports, guided with illustrative examples, and be part of the qualitative appraisal by the reporting officers. 

Yes, there will be challenges with the credible measurement of some of these behaviours. In such cases, a second-best proxy should suffice. But these inevitable challenges should not be a deterrent to putting in place some forms of such appraisals to prime the bureaucracy towards thinking individually and collectively on the costs of the counterfactuals. 

Finally, one cannot help but marvel at the level of thoughtful engagement at the highest levels of the government in China on routine matters of administration, which are, however critical for state capability. Their codification is itself a great step, and the diligent enforcement, such as by way of the monthly reporting by the CCDI, points to a very high level of state capability (and Party discipline). 

Weekend reading links

1. Private equity faces its reckoning?

Over the past five years, US PE fund returns trail the S&P 500 by more than three percentage points a year; over 10 and 15 years, the industry has barely managed to match that index. Fund hold times have lengthened to seven years from five a decade and a half ago. Some $1.3tn of unvested “dry powder” waits as the industry searches for good targets; over 40 per cent of it has been sitting idle for more than two years. Fundraising is set to decline for the third year in a row. Fees are under pressure too, as big institutional investors demand no-fee co-investment opportunities alongside participation in funds... Two decades ago, according to Preqin, the industry in Europe and the US managed $630bn. Today, it manages $6tn... 

Many buyouts in software — PE’s favourite industry during the boom times — are heading for restructuring as a debt maturity wall approaches. The shares of the big “alternative asset managers” that dominate the industry (Blackstone, Apollo and KKR) have had a poor year... Five years ago, when all financial markets were flooded with cash in the post-pandemic boom and rates were low, the private equity industry went on a giddy spree, paying high prices for companies and loading them with debt. Many of these companies have not grown as hoped. As their financing has reset at higher rates, the residual equity value represents an unacceptably low return. PE fund managers have largely avoided exiting these investments, which would crystallise the poor results, hoping for a profit renaissance or return to low rates that look equally unlikely to arrive. 

2. This graphic is the clearest indication that politics in the US is becoming similar to that in the developing countries, as 88% of the electorate are extremely concerned about corruption.

And 89% think that corruption is widespread in the government.

3. Good graphical summary of how China has come to dominate manufacturing, gradually moving up the value chain. 

And this about the AI industry is striking.

4. China's transportation infrastructure rollout facts.

China’s annual investment in transport infrastructure is 2.6% of GDP. That is a far lower share than after the global financial crisis of 2007-09, when it reached 9% as the country unleashed a massive stimulus package. But it remains strikingly high compared with, say, America, where the outlay was just 0.6% of gdp in 2024. By the end of last year, China had built more than 50,000km of high-speed rail, about 70% of the global total. It plans to build another 20,000km by 2035—roughly 1.5 times the length of such tracks in operation in Europe... Since opening its first expressway in 1990, China has built about 200,000km of such whizzy roads (see map)—about twice the length of motorways in America. There are now about 270 commercial airports, three times as many as in 1990. America has twice the number, but the gap is narrowing—China is adding half a dozen entirely new airports a year.

So how is this being used?

A lot of China’s new infrastructure is underused. Fast rail services in poorer inland regions have more empty seats than, say, the highly profitable line from Beijing to Shanghai... Commercial airports have a similar problem. Between 2019 and 2025, 43% of them saw a drop in traveller numbers (not counting airports newly opened since 2019)... In 2023 China had just under 1,800 vehicles for each kilometre of motorway, far lower than the ratio in most rich countries... In small poorer cities, rail stations and airport terminals are often colossal—built in extravagant ultra-modern style to awe visitors... Local authorities have borne much of the cost of building expressways and local sections of rail. Local-government debt held in off-balance-sheet vehicles may amount to 72trn yuan, or 51% of GDP.

5. Recursive self-improvement (RSI) AI, where AI keeps improving AI applications, is arriving rapidly. Anthropic recently published a chart showing how much of its AI research and development is being done by AI, thereby raising risk of AI development getting out of control. 

It took a whole village of Claudes to produce the data in this chart. A Claude agent dug through the company’s Slack messages and internal documents to identify 15,000 tasks of AI R&D work. A Claude detective investigated exactly how that work was done. Then a Claude judge weighed the evidence and rated the work on a six-level scale, from no AI to fully autonomous AI... The most telling bars are the ones in orange for “AI leads,” meaning it performs most of the R&D while a human supervises—the fifth of the six levels. A year ago, AI led none of this work. As recently as March, it only led 1%. By August, Claude was leading 26% of the work on future Claudes. And that number is almost certainly higher today. Even as the company’s researchers delegate to AI, it’s worth remembering that 0% of this work is being done entirely by AI. It still hasn’t reached the sixth level of automation. But at this rate of progress, AI taking us out of the loop might just be a matter of time, computing power and the decisions of human beings inside the labs.

6. Good long read in Mint about how Waree Energy's solar module manufacturing facility in South Gujarat has run into problems of effluent emissions polluting ground water and local water bodies and inviting opposition from the locals. 

7. Fascinating caricature of the Victorian elite education.

Most socially elite Victorian boys began formal education at the age of seven or eight, spending about five years in a ‘preparatory school’ before passing on to a ‘public school’ (which, to the enduring confusion of international observers, is a kind of elite private school)... Most students at both preparatory and public schools were boarders, meaning that elite Victorians were generally sent away from home by the age of eight and spent their childhoods more with peers than with family. In this respect they were distinctive even in their own time: elite families in continental Europe and the United States were far more likely to keep their children at home and educate them at small local private schools. It is plausible that this contributed to the lack of subnational loyalties in the British elite, in contrast to the distinct elites of, for example, the Southern United States, Catalonia in Spain, or Hungary in the Austrian Empire. The public schools were of course single-sex, generating a somewhat febrile atmosphere in which intense quasi-romantic (or sometimes actually romantic) relationships between boys were common... Demonstrative affection remained normal in adult life, and it was completely standard for pairs of Victorian men to walk arm-in-arm in public.

Academic life was dominated by Latin and, to a lesser extent, Greek. The exact share varied, but the classical languages took up generally at least half of teaching hours, and sometimes as much as 80 percent, especially earlier in the century. Trailing in third and fourth places were mathematics and modern languages. There was resistance to the teaching of science, not because Victorians were hostile to science in principle, but because they thought it irrelevant to the kind of moral and cultural formation that a school for gentlemen was supposed to provide... The most that can be said for Victorians’ classical education is perhaps that, like schooling in any difficult subject, it trained students in self-discipline, memorization, and applying rules... The focus on character formation led to increasing interest in how students spent their time outside the classroom... During the Victorian period, school authorities introduced programs of organized sports, chiefly cricket in summer and various forms of ‘football’ (including what became rugby) in winter. Many schools ultimately had four, five, or even six afternoons of sports per week: a student would certainly spend ten hours a week on the sports field, and might well spend twenty. The Victorians thought extremely highly of sports, believing that they instilled team spirit and endurance, fostered mutual loyalty, channeled energy and competitiveness, and taught men how to win and lose graciously. An extreme but not entirely unrepresentative example was Hely Hutchinson Almond, headmaster of Loretto School, who described his school’s priorities as ‘First – character, second – physique, third – intelligence, fourth – manners, fifth – information’...

Oxbridge tutors believed overwhelmingly that the purpose of an undergraduate degree was neither training for work nor even preparation for academic research, but rather a general moral and intellectual formation that would fit students for any professional role they might later pursue. ‘No man will be a first-rate physician or engineer who is not something more than either, who has not some taste for art, some feeling for literature, or some other interest external to his profession’, said Benjamin Jowett, the influential master of Balliol College... 

By one definition, then, elite Victorians did remarkably little work: they could easily spend just four or five hours a day actually in the office. But by another definition, they were working all the time. Lunches, balls, and dinners were certainly fun, but they were not just fun. In the words of the historian Leonore Davidoff, these social occasions provided ‘access to a vast information network: information about jobs, investment possibilities, secret political decisions’... The distinctive feature of elite Victorian lifestyles was not laziness, but an extreme weighting toward what we might call team building and professional networking, and the near-total fusion of personal and professional life... The sociologist Ferdinand Tönnies famously distinguished Gemeinschaft – community, held together by ties of sentiment and tradition – from Gesellschaft – society, held together by explicit laws and contracts. As societies grow more complex, he thought, Gesellschaft tends to replace Gemeinschaft. But the Victorian elite was a little Gemeinschaft at the apex of an enormous Gesellschaft, an intimate community bound together by ties of common culture, shared memory, intense status competition, loyalty, and love. Its members strove for one another’s envy, esteem and gratitude, and to achieve this they would go far beyond what was contractually obligated or financially advantageous.

8. MS Sriram has a good assessment of the Narasimhan Committee (II) recommendations. 

9. The pension reform challenges facing France and Europe.

French voters enjoy both one of the world’s longest life expectancies and some of its earliest statutory retirement ages. Reaching pension age in the country is widely considered a rite of passage into a second phase of life dedicated to travel, hobbies and leisure that, for the luckiest, can stretch well into three decades. As a result, voters have fiercely resisted government policies designed to make them work longer. A recent government report revealed that roughly one-third of the Education Ministry’s budget is siphoned off to cover pension checks for former civil servants. At the Interior Ministry, which oversees policing, pensions absorb 30% of total departmental spending... French Finance Minister Roland Lescure said raising the nation’s legal retirement age—now between 62 and 64—was mathematically unavoidable...

Countries including France, Italy and Spain rely on pure pay-as-you-go financing. There, state pensions replace between 74% and 82% of previous earnings. Unlike in the U.S., where more than $35 trillion in private retirement assets sits in stocks and bonds, employees’ pension insurance contributions flow directly into the pockets of retirees. These contributions exceed 27% of gross wages in France and 33% in Italy. When borrowing costs surged during the 2010-15 eurozone sovereign-debt crisis, Greece, Italy, Spain and Portugal pushed through some pension cuts. In 2011, Italy lifted its retirement age to 67 and linked it to life expectancy. Spain lengthened the wage history used to calculate payouts. Greece slashed pensions by up to 40%. France, however, largely stayed put. When Macron pushed a bill progressively raising the retirement age from 62 to 64 in 2023, he had to bypass parliament without a vote, triggering months of street clashes, burned trash heaps and nationwide transit strikes. That reform has since been put on hold.

In a second camp, Northern European countries have developed hybrid models that economists generally see as more sustainable. Denmark, the Netherlands and Sweden combine a basic state pension floor with quasi-mandatory workplace pension plans invested globally. Pension assets in these three countries range from 100% to 200% of gross domestic product. This is comparable to the U.S., but the funds are managed by nonprofit industry trusts that convert accumulated savings into lifetime annuities rather than lump sums. This model delivers guaranteed income support, high retirement benefits and large pools of risk capital that help build national wealth, Kirkegaard said. But it also requires longer working lives: In Denmark, the retirement age is linked by law to life expectancy and is projected to pass 74 for younger cohorts around 2070. 

The third group, comprising Germany and the U.K., sits in the middle. The U.K. runs a low, flat state pension paired with automatic workplace enrollment into private funds. Germany’s system is also pay-as-you-go but pensions are substantially lower than in neighboring France, and monthly payments are capped. Germany also uses an automatic formula that slows benefit growth when the contributor-to-retiree ratio worsens.

10. One industry where Indian firms had a leadership role is in motorcycles, at least in the developing country markets. Hero Motors, Bajaj, and TVS have been leaders, holding more than half the market in Africa. Now, their dominance is being threatened by electric-bikes, where they have been laggards. They are noe facing compeitition from the likes of Spiro, which assembles motorcycles in Africa using Chinese kits and components while building a battery-swapping network to support their use.

TVS said in February that Africa accounts for nearly 70% of its export volumes and more than half of its global operational revenue. Analyst estimates suggest that about a third of Bajaj's exports go to the continent. While Spiro's current sales remain smaller than the combined volumes of the Indian manufacturers in Africa, the growing electric motorcycle market means a rising share of incremental sales is coming from a segment in which Bajaj and TVS are not yet active. Both companies have established electric vehicle businesses in India, led by scooters. Bajaj has said its first electric motorcycle is expected in FY28, with export markets likely to be its initial focus.

11. AI industry fact of the week.

AI-related debt issuance is likely to exceed $600bn in 2026, up 10-fold from 2024.

12. The Irish paradox.

The resources available to the state stand in stark contrast to many of the country’s European peers. Next year, Ireland expects a budget surplus of nearly €10bn at a time when France, Germany, the UK and others are battling painful deficits. Such fiscal good fortunes are the result of a fivefold increase in corporation tax revenues over the past decade paid by a handful US multinational tech and pharma companies. Almost half of Ireland’s corporation tax haul comes from just three companies: Apple, Eli Lilly and Microsoft. A quirk of international accounting allows the government to collect tax from companies even if they produce their goods elsewhere, provided they manage intellectual property in Ireland and have a real presence in the country... The country’s economic fortunes have rarely been better, with rapid growth and billions in tax revenues in its coffers. 

Yet many of its people feel worse off than ever. More fundamentally, many worry that the sources of Ireland’s wealth are fragile and that it is not making best use of its windfall. Housing is far from the only problem. Much of the country’s infrastructure and services — including transport, health, water and energy — are decades out of date or simply insufficient for a population that has increased by 10 per cent since 2020. Despite record levels of employment and average wages growing faster than prices, a recent opinion poll found 51 per cent in Ireland were finding it “a lot” harder to manage. Nearly 18,000 people are homeless — a record figure. Ireland has the highest electricity bills in the EU, and the value of household energy arrears is going up... Ireland’s health service has more than 1mn people on hospital waiting lists for the first time. The capital city of a country famous for its rain is running out of water amid delays to a landmark project.

And this about the tenuous tax balance.

Ireland now charges 12.5 per cent but big companies have to pay a top-up to 15 per cent under OECD rules agreed in 2021. The government says that will bring in another €3.5bn a year. Such trends have made the country dependent not only on the revenues from a few companies but also on their signature products: Eli Lilly’s tirzepatide, Apple’s iPhone and computers and Microsoft’s software applications. If their contribution to corporation tax receipts is stripped out on the basis that it is a potentially transitory windfall that cannot be relied on, Ireland looks dangerously like it is living beyond its means. The government estimates that €23bn of next year’s corporation tax receipts are related to activity abroad that is taxed in Ireland. Without that, Ireland would have been running a deficit since 2022 and would face a budget hole of more than €13bn next year, rather than a surplus of nearly €10bn. By the end of the decade, the country could be €20bn in the red.

When faced with such windfalls, fiscal profligacy is never far away.

Irish Fiscal Advisory Council, the fiscal watchdog has no issue with the government’s capital expenditure plans. But it has been fiercely critical of Dublin’s “blistering” overspending. Far from sticking to the 5 per cent limit in spending growth it announced in 2021, the government has increased spending by 10 per cent on average in each of the past five years, the Fiscal Advisory Council says. With cash at its disposal, the government has resorted to additional spending in between budgets, for goals such as easing fuel price rises after large-scale protests this year, even though pumping cash into an expanding economy can feed inflation.

But it is also saving for the rainy day.

Ireland has set up two sovereign wealth funds and has hired the former head of Norway’s fund, Knut Kjaer, to help oversee them. It has committed to saving 0.8 per cent of GDP every year into the Future Ireland Fund, which it will not be able to touch until 2041, and will also pay €2bn into the Infrastructure, Climate and Nature Fund every year until 2030, although it can draw on that earlier... By the end of the year, Ireland will have accumulated €23bn across the two funds.

13.  Sarah O'Connor argues that the AI labour market problem may be one of entry rather than exits. In other words, it may be less about people in occupations losing jobs to AI, but those more exposed occuptions shrinking by way of reduced entry. 

Occupations threatened by technology shrank over time because fewer young people were hired into them, rather than because existing workers were ejected from them. This was true for Victorian boot-making, for example, where research by academic Hillary Vipond shows the arrival of boot-sewing machines led to a “collapse of entry into artisanal boot-making” rather than the mass displacement of artisans. This is also largely how technology helped to hollow out routine manual and cognitive jobs over the past half century. A study of labour market flows in the US between 1976 and 2018 found that, outside of recessions, outflow rates from routine employment were pretty unremarkable. “Instead, declining routine employment overall is largely due to declining inflow rates for those who used to find employment in these occupations,” the authors wrote.

14. Janan Ganesh get many things right in this assessment of populists. It is interesting that the only surviving populists are those who have never been in power, like in Germany and France. 

15. This graphic captures the scale of the repudiation of the Trump presidency. 

After losing the national vote for the House of Representatives by two points in 2024, Democrats are on track to lead Republicans by more than 12 points — the largest swing in their favour in over 75 years... In Seward and Finney counties, Kansas, where Trump won by more than 30 points in 2024, voters picked Democratic Senate candidate Adam Hamilton over his Republican counterpart Roger Marshall.

JD Vance and Rupert Murdoch have squarely blamed the Iran war and by extension Trump for the election debacle.  

Sean Hannity, one of Trump’s staunchest defenders, was asked whether the election would have turned out differently had the president not gone to war with Iran. “One hundred per cent. I don’t think there’s any doubt about it,” he said. Laura Ingraham, another longtime Trump loyalist, said the conflict had been “really tough and not well explained”. The shift comes as Republicans begin to reckon with the scale of their electoral defeat. It raises questions about the future of an alliance that has helped sustain Trump’s dominance of US conservatism over the past decade. The Wall Street Journal blamed Trump’s “erratic and self-indulgent” second term for squandering a “rare chance to advance the cause of conservative reform”.

This is what the WSJ headline spoke.

The Iran war, gas prices and Trump’s miscalculations led to one of the most decisive repudiations of an American president.

This is what is spooking the AI trade.

The top five states for data centre building are all now governed by Democrats with a hawkish bent when it comes to new construction of server-filled sheds. That political arrangement could affect 45 per cent of planned data centre expansion, Goldman Sachs reckons. Curbs on new building, already in place in New York state, are likely to proliferate. Out of 47 local votes on data centres counted by Redbud Consulting, at least 38 came out in favour of restrictions.

But this may actually be the rightly needed sand on the wheels of the out of control AI spending. 

16. The AI bubble in one graphic.

The US is also undergoing its longest lasting span of growth expansion.

17. After the tech rout of Thursday, Robert Armstrong makes some important points. 

The case against panic starts with the fact that, if we were in a classic FOMO market riven with euphoria, it ended a while ago. The biggest US semiconductor companies peaked in early October or, as in the case of Micron, well before that. The memory hardware companies, which have risen even more than the semis, peaked back in June. The industrial stocks most associated with the data centre boom have been falling steadily since May. 

The point is that the election has accelerated a process that was already in place. Risk has been coming off the table for a while, and until yesterday, in a reasonably orderly way. At some point, we can hope, the AI complex will find a new equilibrium, and the market can go on about its business — at a meaningfully lower level, perhaps, but without an all-consuming crash. Remember that the economic fundamentals underlying the market remain strong, as witnessed just this week by solid Jolts, jobless claims, and ISM manufacturing and services surveys. Earnings growth, even outside of technology and energy, is solid too, and share performance outside of tech has been improving in recent months. None of this is a guarantee against a market collapse, but has to make one less likely.

He has four questions for the foreseeable future.

Are there signs of investor capitulation outside of tech stocks, that is, a widening (as opposed to a deepening) of the equity sell-off? Are there any signs of emerging macroeconomic weakness that could reinforce the troubles in the equity market? Does the confidence of the leading companies in AI crack, as expressed by abandoned plans to raise and deploy capital? For example, will Anthropic indefinitely suspend their plans to list? Despite persistent inflation and sound macroeconomic data, do bond yields take another major leg down from here, foreshadowing a slowdown? If the answers to those four questions are no, the equity market should be OK.

18. India-China trade facts of the week.

In FY20, the year India walked out, it bought $65 billion of goods from China and sold ~$17 billion, a deficit of $48 billion. In FY26, it bought $132 billion and sold $19.5 billion, a deficit of $112 billion, according to government data... The trade body Global Trade Research Initiative (GTRI) found that in 2023–24 about half of India’s imports from China were capital goods and machinery, and 37% were intermediate goods such as chemicals, drug ingredients, and plastics. The electronics numbers are sharper. A report in September found that China supplied at least 80% of India’s imports in 71 electronics product lines in FY26, up from 44 in 2018–19. Electrical machinery alone accounted for $43 billion of the deficit, about 38% of the total, while India’s exports in the same category to China were $3.2 billion. China supplied 83% of India’s lithium-ion battery imports in FY26, worth $4 billion, more than double the level of 2021–22.

See also this.

19. PE in India's healthcare sector.

Private-equity funds, such as Temasek, Blackstone, TPG, and KKR, which have collectively invested over $15.5 billion in healthcare in India over the last five years. Of this, 60% went to hospitals.

Analysing the challenges facing China’s economy

China’s economy is unique in multiple respects. And its economic growth presents a conundrum. 

Arguably, the biggest structural anomaly with the Chinese economy is its ultra-low consumption share compared to any other country. The other side of this is the very high investment share. 

Ken Rogoff has this anecdotal measure of the investment-driven growth reliance.

China has built more square metres of residential property per capita than either the United Kingdom or France, despite still being a much poorer country. Its road and railway networks are on par with those of the world’s most advanced economies and, in many ways, far superior to America’s.

But this investment-based growth is a diminishing returns story and finds its limits. This is clearly reflected in its declining contribution to the GDP growth. Not only is the investment rate itself declining, but its incremental impact on output growth is falling too. 

Fixed asset investment declined 7.2 per cent year on year in the first nine months of 2026. Assuming fixed investment at 40% of GDP, this means that the rest of the economy must grow at 3 percentage points more just to prevent contraction. Even the traditional workhorses, like infrastructure investment, are declining.

The new driver of growth is exports, where the trade surplus hit $1.2 trillion in 2025. Net exports supplied a staggering 32.7% of 2025 growth (it was 30% in 2024), and the value-added embodied in exports is ~17–18% of GDP and underpins tens of millions of jobs. More on this latter. 

The declining investment contribution to GDP growth also points to less and less efficiency in capital utilisation. This is captured in the incremental capital-output ratio (ICOR), what a unit of investment buys in terms of GDP growth. The ICOR has more than doubled since the 2000s. China is now roughly twice (or thrice if independent estimates are taken) as capital-hungry as Korea or Taiwan at the same stage, or as India today.

This investment is increasingly being funded by debt. The total social financing (TSF) has reached 315% of GDP. And this debt sits overwhelmingly in corporates, state firms, and local-government vehicles, hitherto the drivers of investment and growth. 

The public debt is much higher if the off-balance sheet vehicles of the local governments are added. A new study from the Conference Board says that debt servicing payments alone are expected to account for 19.2% of the central government’s general public budget this year, up from 12% in 2014. Local government debt interest payments have risen 16% annually over the same period. Morgan Stanley estimates the public debt-to-GDP ratio to hit 122% by the end of 2026.

To put this debt accumulation in a global perspective, China now accounts for more than half the rise in the world’s debt-to-GDP ratio since 2008. 

The danger is that once it reaches these levels, each incremental unit of credit buys less and less. Unsurprisingly, in 2025, one yuan of GDP required ¥6 of credit, as credit growth was double the nominal GDP growth rate. New lending increasingly refinances old liabilities rather than funding new output. Local-government vehicles alone owe ¥60tn (48% of GDP), much of it against land whose value has fallen. The credit boom's real legacy is a banking system propping up unproductive firms.

All this cascades into output growth trends. 

The government has targeted the 2026-27 growth target at 4.5-5%, and the official estimates indicate that the economy grew 4.5% in the third quarter of 2026 and 4.6% for the first nine months. But several independent reconstructions put China’s true 2025 growth at roughly half the official figure.

One estimate is this.

Capital Economics estimates that growth in China in the third quarter was 2%, dragged down by declining industrial activity, particularly in construction and petrochemicals. Its model draws on hundreds of lower-profile official data series, like intercity passenger traffic and industrial electricity consumption, which the firm says makes them less susceptible to manipulation.

Another puts it lower still.

Rhodium Group estimates that full-year GDP will be around 0% in 2026, arguing that it is impossible for China to hit 4.5% growth with sluggish retail sales growth and steep drops in fixed-asset investment.

On NBS accounting, net exports added about 1.5–1.6 points to growth in 2024 and 2025. But that is inflated by collapsing imports. If we strip that out, the genuine external contribution is between a fifth and a quarter. Either way, remove net exports and domestic-demand growth falls to 3.4–4.0% on official data, and to 1–1.8% on independently-estimated growth.

But this reliance on exports is a big vulnerability, given the rising backlash against cheap Chinese imports destroying manufacturing bases in both developed and developing countries, and the protectionist barriers that are being erected by them. It is safe to say that the export markets will be squeezed in the immediate future. 

Besides, the race to capture export markets has also spawned an unhealthy race to the bottom among companies, also spurred by equally intense competition among local governments. 

The electric vehicle (EV) market is emblematic of this trend, with local governments pouring billions into creating local champions, resulting in a staggering 143 EV brands selling at least one model at the end of 2025. Since factory closures mean job losses and admission of failure, besides also impacting the promotion prospects of senior city leaders, cities are reluctant to pull the plug and continue to support them. Exports have emerged as a convenient outlet to channel the excess capacity, and production has continued to rise. Exports are expected to touch 12 million units of passenger vehicles, up from less than a million units at the turn of the decade, thereby overtaking Japan as the world’s biggest auto exporter. 

However, the intensity of the competition has meant that even as Chinese EVs are capturing the global market, their EV companies are struggling to remain profitable. As an example, NIO, one of the leaders, has been losing more than a billion dollars each year since 2021, and its share price has fallen by 95%. In the first eight months of this year, the industry’s profit margin fell to a multiyear low of 3.6%, as companies compete in a pricing race to the bottom. Chinese officials call this price competition neijuan or involution. 

Worse still, globally, the cut-throat pricing has also forced price reductions by Western car makers, thereby triggering a race to the bottom globally in the industry. For example, BYD is selling its flagship Seal hybrid sedan for the equivalent of around $44,000 in the UK, more than $10,000 less than a similarly equipped Volkswagen Passat. 

On the one hand, there is the weakening consumer demand, which has resulted in 22% decline in the domestic sales of passenger vehicles in September (compared to the same period last year), and sales for the full year are estimated to reach only 18.5 million units, the worst performance since 2014. This must be seen against the backdrop of an annual production capacity of around 63 million light vehicles, more than triple the number of cars expected to be sold in the domestic market. 

So what does all this portend for China’s future?

China is facing a prolonged real estate slump, dismal consumer expectations, falling retail sales (and that too in a country with private consumption at two-thirds of peers), declining fixed investment, and struggling corporate profitability (due to intense competition). 

A recent FT essay argued that the combination of persistent deflation, weakening economy, dismal consumer sentiments, rising public and private debt, and ageing population means that China is facing its Japanification. 

According to Goldman Sachs, an updated version of the Japanisation index — a gauge originally devised by the late economist Takatoshi Ito — shows that for the past five years, China has been consistently more “Japanised” than Japan itself… Japan’s population started to shrink in 2009; China’s decline began in 2022 and is on track to be steeper, worsening the country’s economic prospects... In another echo of Japan’s post 1990s-slump, the IMF warned last week of growing numbers of “zombie” companies across Asia, kept alive only by lenient banks. This is now evident across China’s heavily subsidised manufacturing economy. Outside favoured sectors, companies are increasingly refusing to invest while households hoard cash instead of spending... Consumers are deleveraging, with bank lending to households as of August this year declining outright for the first time in history... China’s weakening economic conditions meant that up to 10.6 per cent of the adult population had overdue loans last year, compared with about 7.8 per cent a year earlier, according to a report by Gavekal Dragonomics analyst Xiaoxi Zhang...

Phrases coined in 1990s Japan, such as “job hunting ice age” and “price destruction”, are creeping into China’s vernacular... economists can see the signals of Japanification in the data... Restaurants across China are competing to offer meals for under Rmb10 ($1.50). This too is redolent of Japan, where restaurants entered a margin-crushing price war to attract an ever-shrinking pool of consumer spending as deflation took hold. The price of a basic bowl of rice and beef at the nation’s leading gyudon chain hit a high of ¥400 ($2.50 at current exchange rates) at the end of 1990 before falling. It only reached ¥400 again in 2021... The “expectation channel”, where households and businesses saw no prospect of longer-term growth, created a vicious circle of reduced consumption and investment in Japan which in turn lowered expectations...

China’s real estate bubble was the largest in history, outpacing even Japan’s speculative frenzy of the late 1980s and at its peak accounting for a third of the country’s economy. The contraction since it burst in 2021 has been dramatic, with new housing starts down 78 per cent from their peak, according to Macquarie. That compares with a 73 per cent fall in the US between 2005 and 2009 and Japan’s 65 per cent drop between 1996 and 2025. The collapse of the property market has set off what Richard Koo, the Nomura Research Institute’s chief economist, calls a “balance sheet recession” — a term he originally coined in the 1990s to describe the tendency of Japanese companies and individuals to pay down debt rather than borrow to invest or spend. Weaker demand and deflation then worsen the debt burden, leading to a vicious cycle. Investment opportunities decline, the labour market calcifies and salaries stagnate... The ultimate sign of China’s weak credit demand is the 1.7 per cent yield on 10-year government bonds, which Nomura’s Koo describes as “absolutely ridiculous”...

One reason for Beijing’s reluctance to resort to domestic stimulus, economists believe, is China’s strong export performance. The trade surplus hit $1.2tn last year — the biggest recorded by any country, ever — and may meet or exceed that level this year. Policymakers in Beijing think the country’s export performance will buy China time while it waits for the property market to stabilise. In the meantime, they are pursuing investment in areas such as chipmaking and AI as the country competes with the US for geopolitical dominance... Gross fixed capital formation — the acquisition of assets for the production of goods and services — was 40 per cent of GDP in 2024 against Japan’s 28 per cent and the OECD average of 22 per cent, fuelled by an augmented fiscal deficit that Goldman Sachs forecasts will hit 11 per cent of GDP this year. No one knows when this debt will become unsustainable.

Michael Nicoletos has an excellent post whose conclusion is worth reproducing.

Household debt is rising, defaults are accelerating, the population is shrinking and aging, and precautionary savings behavior suggests Chinese citizens understand their economy better than policymakers do. The question is not whether China can rebalance. Structurally, it can, if it allows wealth redistribution to work, implements genuine pension reform, and expands healthcare spending. The question is whether it will. Current policy signals suggest the answer is no.

Until that changes, China will remain trapped in the same pattern: high investment, low consumption, rising debt, shrinking population, slowing growth, and increasing vulnerability to external pressure on its export lifeline… The demographic projections make this even more dire. A country that is losing 3 million people per year now will be losing 14 million per year by 2062. A workforce that shrinks by 225 million cannot sustain the growth rates Beijing demands. An aging population that nearly doubles its dependency ratio cannot suddenly transform into a consumption-driven economy… 

The Chinese government will have a tough time boosting domestic consumption. Consider what it is up against: households that are over-indebted, scared, and saving at record rates; a population set to shrink by up to 280 million by 2050 and potentially by half by 2100; a dependency ratio that will nearly double, leaving just 1.2 workers to support each retiree; an export machine that relies on rerouting through third countries to avoid tariffs, with that loophole closing; and an energy import bill exceeding $450 billion a year, payable in foreign currency… Demographics make this worse, not better. You cannot boost consumption when your population is shrinking and aging. You cannot grow your way out of debt when your labor force is declining by 225 million workers. You cannot fund pensions and healthcare when the worker-to-retiree ratio collapses from 4:1 to 1.2:1.

All this means that catching up with the US GDP will most likely remain elusive, unless the US GDP itself starts falling. Even escaping the middle-income trap may become a challenge. 

China’s investment-led growth model started with the likes of textiles and footwear, and gradually moved up into more sophisticated products like consumer durables, heavy equipment, renewables, transportation, and now technology goods. Heavy doses of financial repression, massive subsidies, and real estate revenues financed this model. The prolonged real estate market downturn has seriously strained the model and now threatens to upend it. 

In conclusion, it is hard not to argue that any economy which has seen over ten quarters of deflation, an enduring property bust (housing starts down 78% from peak), total debt past 300% of GDP, a population shrinking ~3mn/year, and ageing toward 1.2 workers per retiree, is facing a reckoning. Each raises the ICOR further and shrinks the base that services the debt. The investment led growth model has run its course. Goldman’s Japanisation index already reads China as more “Japanised” than Japan. The export surplus is merely buying time and kicking the can down the road. Even that, too, is already getting squeezed. Factory closures and job losses are not far away. When that starts to happen in a meaningful manner, the tenuous contract between the Party and the citizens will be strained. It remains to be seen how China will navigate this inflection point. In the circumstances, Carl Minzner points to a long twilight.