Weekend reading links

1. Good chronicle of Trump at War in the NYT. 

2. List of a few impacts of the Gulf War from across the world. 

3. Foreign national government holdings of US Treasuries drop to their lowest value in years as they sell the Treasuries to prop up their economies and currencies.

4. For all its cultural salience and ubiquity, Bollywood box office receipts were $1.45 bn in 2025!

Interesting that video subscription revenues rose 61% in 2025 to $1.56 bn, according to EY.

Bollywood is another example of an Indian private sector/industry struggling in the face of global competition.

One top producer who has worked with many of Bollywood’s top actors says the combination of cultural drift and waning star power is a deadly one. “If you’re bankrupt on creativity, then the only thing you’re banking on is stars,” he says. “Nobody’s coming to the movie theatre to watch a star any more.” This, he adds, is a major reason why several international studios abandoned their attempts to enter the Indian industry...

Bollywood has remained resistant to outsiders. International studios such as Fox Star, Paramount and Sony entered the country more than a decade ago, looking to streamline the world’s largest movie-producing market, which remains parochial and dominated by family-run production houses. But through either mergers or losses, almost none exist in the country any more. Its films have also struggled to find audiences overseas. Unlike K-dramas from South Korea or Turkish soap series, Bollywood is yet to find breakout success globally, beyond the south Asian diaspora.

And this is an important market failure

The decline in Bollywood stars’ pulling power has not been reflected in either their remuneration or their creative influence. Male leads in particular wield immense power over which films get made and command salaries that are far in excess of what female actors can demand. Although Bollywood budgets are much smaller overall than those of the big five studios in the US, male stars often demand half or more of a film’s budget — compared with at most a quarter in Hollywood. The eminent producer says the industry still wants to “helicopter the star, do the least amount of work [possible] and try and do some razzmatazz, and everyone’s followed the herd mentality”. Shailesh Kapoor, chief executive of Ormax, says the multimillion-dollar fees that some top actors demand are driven by the free market. “Stars do not downgrade their fee if their last movie doesn’t work well . . . because somebody is willing to pay.”

5. Gulf war scenarios.

Oxford Economics said this week it estimates a six-month interruption of flows through the Strait of Hormuz would create a 13mn barrels per day gap in global oil supplies, triggering a worldwide recession. “That represents an unprecedented shortage of around 12 per cent of consumption, leading to widespread rationing concentrated in emerging economies, with significant hits to activity and supply chain disruption,” said the advisory group.

The Strait of Hormuz chokehold provides China with enough ideas on how to implement a similar strategy in the Taiwan Straits.

Beijing would declare the right to control who and what comes and goes from the island. China could demonstrate resolve by firing missiles or bullets and declaring “exclusion zones”. Even short of outright conflict, if escalation risk seemed high, private carriers would face pressure to avoid the waters and airspace around Taiwan. The standard “Five Powers Clause” in commercial war-risk insurance terminates coverage in any conflict between the US and China. If it has proved hard to persuade shipping companies to risk sailing amid a few unguided Iranian drones, imagine asking them to take on the People’s Liberation Army.

More on the oil supply squeeze.

Fatih Birol, the head of the International Energy Agency, has called this “the greatest global energy security threat in history” — much worse than the 1970s oil crisis, the Covid pandemic or Russia’s invasion of Ukraine in 2022. This conflict has disrupted a bigger share of the global oil and gas trade, and there is no way to quickly fill the gap.

6. FT series on how Apple became obsessed with quality. The first one is about how it learned the quality focus from Japan. 

7. NYT on Pakistan's remarkable pivot to green energy.

For governments weighing how quickly to pivot to clean energy, Pakistan’s recent history may offer some lessons. The country was hit hard by the energy shock that followed Russia’s invasion of Ukraine in 2022. It was unable to afford suddenly exorbitant gas imports, so many of its scheduled shipments were rerouted to wealthier European buyers. But a deluge of inexpensive solar panels from China has transformed Pakistan’s energy system and helped insulate it from the current scarcity of liquefied natural gas. Solar now generates nearly 30 percent of its electricity, up from just 3 percent in 2020. The Center for Research on Energy and Clean Air has estimated that the solar boom will help Pakistan avoid $7 billion in fossil fuel imports this year. That’s shielding Pakistanis from real pain.

8. Sekhar Gupta makes an important point about India's fertilizer import dependency and advocates the use of coal gassification to domestically manufacture them. 

Coal, we know, is an unpopular fuel... And while coal-to-chemicals is also a polluting activity, it is much less so than burning it in power plants... while burning coal for electricity may be evil, turning it into gas is much less so... Fertiliser shortages are an even bigger threat because this affects our food security... China produces more than 90 per cent of its ammonia from coal gasification. Ammonia is essential for DAP. India imports much of it and so severe are the shortages that to prevent rioting or looting by desperate farmers, many states store the supplies in their police stations, allocating to farmers on the basis of land holdings and Aadhaar-based registrations. Now hold your breath. China uses syngas (synthetic gas) derived from coal to produce 40 per cent of the entire world’s urea. It also produces 54 per cent of all of the world’s methanol, around 70 per cent of it from coal... Among the large agricultural producers, we’re the most import-dependent for our fertilisers... Our crippling dependence on fertiliser imports is also ship-to-mouth existence by another name. This war in the Gulf has driven home these vulnerabilities.

9. The GPU race

Graphic processing units (GPU) are the “oil” on which AI runs... India aims to increase its current stock of 38,000 GPUs to 100,000 by the end of the year, with a target of 1 to 2 million by 2030. The state is providing GPU “time” as a service at a subsidised rate of only ₹65 per hour to foster application development by startups... The US has a current inventory of 5 to 7 million GPUs, and China has 1.5 to 2.5 million. The target for the US is 25 to 30 million by 2035 and China plans to achieve 10 to 15 million by the same date.

Weekend reading links

1. Toronto's pedestrian tunnel system, The Path, that emerged in response to congestion in the main streets.

In the early twentieth century, Toronto’s businesses developed a novel response to this. They began to create pedestrian tunnels from their offices to the metro stations so that their employees could flow in smoothly, avoiding the congested streets (and, in winter, the cold). Shops quickly started to be added. After a few businesses had done this, a ‘network effect’ emerged whereby other businesses started to add their own tunnels to the system, benefiting from the existing tunnels while also making them more useful. It became routine for downtown developers to tie new office blocks into the network. Over many decades, a sort of ‘pedestrian metro’ emerged.

Known as the Path, the network today stretches for more than 30 kilometers, linking nearly all central metro and railway stations with many of the major office buildings. Although the Path forms a unified network, it is not in unified ownership: it is divided into some 35 chunks, each of which is still owned and managed separately by descendants of whichever business originally contributed it. Many branches of the Path thus terminate in the lobbies of office buildings, with the curious result that these grand spaces function as metro entrances for the general public. The municipal authorities play only a limited regulatory role.

The Path is unlike the gloomy and malodorous underpasses with which most of us are familiar. It is expensively decorated and feels like a high-end shopping mall, which in a way it is. It is extremely clean and closely policed by dozens of private security teams. Until recently, it was thronged with shoppers: this use suffered in the pandemic and has not wholly recovered, but the Path is still used for its original commuting purpose by hundreds of thousands of people every weekday.

2. Nvidia has a new rival for its GPU chips, Google's tensor processing units (TPUs), which it used to train its latest Gemini 3 LLM. 

Nvidia’s customers have a big incentive to explore cheaper alternatives. Bernstein, an investment-research firm, estimates that Nvidia’s GPUs account for over two-thirds of the cost of a typical AI server rack. Google’s TPUs cost between a half and a tenth as much as an equivalent Nvidia chip. Those savings matter, given the vast sums currently being poured into computing power for AI. Bloomberg Intelligence, another research group, expects Google’s capital expenditures to hit $95bn next year, with nearly three-quarters of that being used to train and run AI models.

But Nvidia's moat is unlikely to disappear.

For Nvidia’s other customers, however, switching to Google’s chips will not be straightforward. Nvidia’s edge lies partly in CUDA, the software platform that helps programmers make use of its GPUs. AI developers have become accustomed to it. And whereas the software surrounding TPUs has been created with Google’s own products in mind, including search, CUDA is intended to cater to a wide range of applications. What is more, reckons Jay Goldberg of Seaport Research Partners, an industry analyst, there may be a limit to Google’s willingness to sell its TPU; it could prefer instead to steer prospective customers towards its lucrative cloud-computing service. To stymie its AI competitors, Google may also be tempted to keep prices for its chips high.

3. AI adoption is stalling, so says a survey by statisticians at the US Census Bureau. It finds that the employement-weighted share of Americans using AI at work has fallen by a percentage point to 11%, and has fallen sharply at the largest businesses. 

The article has this description of the challenge facing the AI market.

From today until 2030 big tech firms will spend $5trn on infrastructure to supply AI services. To make those investments worthwhile, they will need on the order of $650bn a year in AI revenues, according to JPMorgan Chase, a bank, up from about $50bn a year today. People paying for AI in their personal lives will probably buy only a fraction of what is ultimately required. Businesses must do the rest... Jon Hartley of Stanford University and colleagues found that in September 37% of Americans used generative ai at work, down from 46% in June. A tracker by Alex Bick of the Federal Reserve Bank of St Louis and colleagues revealed that, in August 2024, 12.1% of working-age adults used generative AI every day at work. A year later 12.6% did. Ramp, a fintech firm, finds that in early 2025 AI use soared at American firms to 40%, before levelling off. The growth in adoption really does seem to be slowing...

According to a poll of executives by Deloitte, a consultancy, and the Centre for AI, Management and Organisation at Hong Kong University, 45% reported returns from AI initiatives that were below their expectations. Only 10% reported their expectations being exceeded. A study by McKinsey, another consultancy, argued that for most organisations, the use of aihas not yet significantly affected enterprise-wide profits... A paper by Yvonne Chen of ShanghaiTech University and colleagues refers to “genAI's mediocrity trap”. With the assistance of the tech, people can produce something “good enough”. This helps weaker workers. But the paper finds it can harm the productivity of better ones, who decide to work less hard.

4. India has caught up with China in education imports from the US (or students travelling to the US to study there). India's exports have doubled in just three years.

And India's share of US exports has now caught up with China. 

5. Striking statistic about private credit and insurance.

Close to 37 per cent of North American life insurance investments are allocated to private credit. But private credit encompasses a world of different types of lending — from private placements and commercial real estate lending to asset-based finance and fund finance — in which they have long-standing expertise.

Banks are also big lenders to private credit firms. 

6. Comparing AI investments with those in railways in the 19th century.

The railways, for example, were considered similarly transformative in the 19th century... As a percentage of investment in the United Kingdom, the world’s economic behemoth in the mid-Victorian period, domestic railways accounted for perhaps half. This was at the peak of the mania, in the 1840s. But even over the longer term, railway investment accounted for around a fifth of total investment over the four decades after George IV died in 1830. Around the turn of the century, railway bonds and shares accounted for between a quarter and a third of household financial portfolios in Britain. In the United States, during various railway-building booms (the 1840s, the 1870s), investment in the sector accounted for 40 per cent of total investment in the economy. At some points, it accounted for almost 10 per cent of gross domestic product (GDP)... AI investment may have accounted for all of US GDP growth in the first half of this year, but that still means only a couple of percentage points of GDP, as compared to the 6-10 per cent that was routinely achieved during railway booms.

7. Indian equity market facts

The MSCI India index has returned 2.5 per cent in dollar terms this year compared with 27.7 per cent for the MSCI Emerging Markets index, India’s weakest relative performance since 1993. Foreign investors have pulled more than $16bn out of India this year, the second-largest drawdown on record

Real interest rates have risen.

8. Indian economy update

9. Rising interest rates in Japan are bringing to an end decades of the yen carry trade which allowed western markets to benefit from Japan's savings. 

10. African electricity facts.

To get a sense of how little electricity people use in sub-Saharan Africa, imagine each person there turning on a single 50-watt light bulb. That alone would instantly double electricity consumption. Nigeria, with 240m people, generates less electricity than the American state of Wyoming, which has 0.6m inhabitants. Uganda, with 50m people, produces less than Latvia, which has a population of 1.9m. Around 600m Africans have no electricity at all.

11. A good summary of the Indigo fiasco that is now playing out in India.

The story begins with pilot associations filing petitions in the Delhi High Court around 2019-20, pushing for stricter rules regarding pilot fatigue. The High Court repeatedly questioned India’s Directorate General of Civil Aviation about updating its rules on pilot fatigue and, in response, the DGCA unleashed the Flight Duty Time Limitations (FDTL) rules... FDTL is a set of safety regulations that determines how long a pilot can work, how much rest they must get, and how many night flights they can operate... These rules were notified in January 2024 and were to come into full effect on November 1, 2025. IndiGo has now admitted that it did not plan adequately for these changes and thus faced pilot and crew shortages, which have led to this mayhem.

This is an important aspect about what happens when a dominant market player defaults in an infrastructure sector. 

In advanced economies, the repercussions would be through the markets and the courts. The markets are the first recourse – customers would punish the airline by choosing alternatives, and the company would have to go out of its way to win back market share. The problem, however, is the lack of alternatives in India, which was one of the main causes of such a scenario in the first place. IndiGo commands about 65 per cent of the market share, which translates to roughly two out of every three domestic flyers in India being on IndiGo on an average day, with the rest of the industry effectively competing for the remaining third. This enormous market power means that customers, despite themselves, might have to choose IndiGo again for their travel needs due to the lack of options.

Weekend reading links

1. As grocery prices rise and popularity ratings dive, President Trump rolls back tariffs on certain agricultural products where import reliance is high. 

Trump issued an executive order on Friday afternoon saying that imports of certain goods that were generally not grown or produced in the US would no longer be subject to “reciprocal tariffs” — the high levies he set based on emergency powers starting in April. The president’s order said the tariff exemptions would apply to common and tropical fruits including oranges, tomatoes and bananas — as well as cocoa, coffee and tea. Beef imports were also included in the list, as well as spices and some fertilisers, according to a factsheet provided by the White House.

Beef prices have surged in the US, with the average price of a pound of ground beef rising 13% in a year, while uncooked steaks rose 11%, leading to steakhouses raising prices or trimming portions or both.

The exorbitant tariffs of 50% on Brazil, the world's largest beef exporter, have been a major contributor. 

Before the Trump administration levied a sweeping 50 per cent tariff on Brazil in July, the US had been steadily increasing imports from Brazil in order to keep up with domestic demand. In the first five months of 2025, the US imported some 215,000 tonnes, more than double during the same period in 2024. After July, the effective rate for out-of-quota Brazilian beef rose to more than 76 per cent. Exports to the US, Brazil’s second-largest beef market year-to-date, fell 41 per cent in September to $102.9mn.

2. Continuing on the tariff front, Switzerland has reached an agreement with the US to lower tariffs from 39% to 15%, the same rate as EU exports to the US. In return for the deal, Swiss companies have promised to invest $200 billion in the US by the end of 2028. White House has said at least $67 of the investment would occur in 2026, and Swiss businesses would set up apprenticeships and training programs in the US. 

This also follows trade deals with Argentina, Guatemala, El Salavador, and Ecuador over the week. 

3. Good news on the South African economy, as S&P upgrades sovereign ratings for the first time in two decades to BB, two notches below investment grade, on the back of reforms and fiscal revenues. 

The rolling blackouts that hamstrung the economy have largely been avoided this year and Eskom, the state power company, returned to profit after eight years of losses and reliance on government bailouts... S&P said the upgrade reflected South Africa’s recent record of budget surpluses, excluding interest payments, and less financial pressure from Eskom... After a decade in which GDP expansion remained below 1 per cent, there have been other positive developments. The country was recently removed from the Financial Action Task Force’s grey list while the survival of the government of national unity has improved investor confidence. This week, the government cut its inflation target for the first time this century to 3 per cent, bolstering a rand rally...

S&P said it expected South Africa’s GDP growth to pick up to 1.1 per cent this year, from 0.5 per cent in 2024. South African assets have stood out this year even in the midst of a rally in other emerging markets, while the rand is up about a tenth against the dollar in spot terms. The Johannesburg all-share index has risen about a third this year, or nearly 50 per cent in dollar terms. The yield on South Africa’s 10-year rand government debt has fallen from 11 per cent in April to about 8.7 per cent.

3. The latest in rent-seeking by the Trump family is a report that the Trump Organisation is in talks to bring a Trump-branded property to a $63 billion government-owned project that is set to transform the historic Saudi town of Diriyah into a luxury destination with hotels, retail shops, and office space. The Organisation is also talking to bring Trum branded property to other developments in Saud Arabia. 

The negotiations are the latest example of Mr. Trump blending governance and family business, particularly in Persian Gulf countries. Since returning to office, the president’s family and businesses have announced new ventures abroad involving billions of dollars, made hundreds of millions from cryptocurrency, and sold tickets to a private dinner hosted by Mr. Trump... In Saudi Arabia, a Trump tower is planned for Jeddah, and two projects have been announced in Riyadh. A Trump hotel and tower has moved forward in Dubai, the largest city in the United Arab Emirates. And a golf course deal in Qatar has put the Trump family in business with a government-owned real estate firm there... Each venture generates licensing fees for using the Trump name... Licensing deals can be lucrative, particularly if a development does well. Often, a company is paid for the use of its name and is not required to invest any money in the project itself. The Trump Organization’s licensing agreements are not public, making it impossible to know the terms.

4. Sustainable high growth rates in India is not possible without broad-basing aggregate demand. More here. 

5. Corporate India's R&D problem in a graphic.

And global comparison.

And it does not seem to be improving.

Most of the top six sectors saw a dip in their share in the R&D expenditure by Nifty 100 companies during FY23-25.

6.  Retail is an illustration of the deeply price-sensitive and low-margin nature of Indian market. 

Foreign brands including West Elm, Pottery Barn and Superdry have stores in Reliance’s shopping malls in upmarket Mumbai. However, those joint ventures have largely struggled to gain traction with shoppers in India, where the per capita income remains less than $3,000. The conglomerate’s foreign brands business housing these joint ventures lost Rs2.7bn ($30mn) in the financial year through March 2025, according to the latest available accounts... Reliance’s high-profile partnership with fast-fashion retailer Shein has also been underwhelming... Shein’s app has been downloaded just 11mn times so far, according to market intelligence firm Sensor Tower. Its discount prices are largely matched, if not undercut, by many Indian ecommerce and fashion retailers, say analysts... Blinkit, Swiggy and Zepto, which together control more than 90 per cent of the quick commerce delivery market and compete with Amazon and Walmart-owned Flipkart. None of the companies are profitable.

7. Arvind Datar has a good explainer of the telecommunications adjusted gross revenue issue that the Supreme Court has just allowed for reconsideration (after having created the problem in the first place).

8. Even as Delhi grapples with toxic air pollution levels, here's something from England.

In my home country of England, levels of PM2.5 — fine particulate matter which is widely seen as the most damaging pollutant to human health — have plummeted. A report by the Institute for Fiscal Studies describes “remarkable progress” over the past two decades. Between 2003 and 2023, the average level of PM2.5 roughly halved in every region of England, and almost everywhere is now already below the target the UK government set for England for 2040.

Also India's VC industry facts

The country has created more than 120 unicorns, start-ups valued at more than $1bn, according to Tracxn — the third highest number after the US and China. Indian and international venture capital firms have invested $96bn over the past five years, according to consultants Bain, in around 8,000 funding rounds. Most of this has come from foreign investors but the domestic long-term capital base is developing fast, from family offices to insurance companies and pension funds.

9. The remarkable precision of Chinese economic forecasts raises red flags.

10. The spectacular reduction in the cost of renewable energy.

Since 1976, the price of solar modules has fallen by 99.6 per cent. With each doubling of installed capacity, the price fell by 20 per cent.

11. Janan Ganesh may well be spot on with his assessment of Rachel Reeves, the British Chancellor of Exchequer,

Rachel Reeves: one of life’s triers, but never cut out for this particular office at this particular time. At next week’s Budget, she will announce a second round of tax rises, which she said would never come. She has spent 2025 fanning and then dousing speculation about certain levies, such as higher income tax, with predictable effects on confidence. (A Tory who behaved like this would be called a vandal.) Most workplaces, including newspapers, contain staff who are out of their depth but survive because the boss is too embarrassed to fire them. They just tend not to be the second-highest person in the organisation.

And on her boss, the Prime Minister, Keir Starmer.

There were warnings about Starmer’s character in opposition. He let others stand up to Jeremy Corbyn, whom he served in shadow cabinet. He let others fight woke dogma, until the tide turned against it. Even now, he makes liberal use of human shields. Notice that every crisis for Starmer quickly becomes a conversation about his underlings. His then chief of staff Sue Gray used to be the problem. Now it is her successor Morgan McSweeney. What rotten luck the prime minister has with recruitment. The British are having to relearn a lesson that Theresa May should have fixed in their minds forever. Don’t assume that uncharismatic people have hidden depths. Being boring does not make someone a “technocrat”. One can be dull and inept.

12. Britain has the biggest difference between the bottom and top tax slabs. 

According to the latest figures from the OECD, 45 per cent of top earners’ salaries goes on taxes and social contributions, compared with 29 per cent for the average worker, for a top-to-middle gap of 16 percentage points. Scandinavian gaps come in at about 12 points. Northern Europe’s social democracies tax everyone from bottom to top at a moderately high rate. In Britain, taxes at the top are comparable to Denmark and Norway but the average Briton is taxed less than the average American.

13. John Martinis, Professor of Physics at the University of California, Santa Barbara, and the winner of 2025 Nobel Prize in Physics, poses a manufacturing challenge to realise quantum computing.

Anyone who has looked inside a modern quantum system can see the truth of this. Look at the diagrams or pictures of devices and what do you see? A jungle of wires and discrete components, all designed to cool and control a single, small chip hidden at the bottom of the cryostat. We have reached a stage where the complexity of the plumbing completely overwhelms the quantum device itself. My vision is that the entire, spaghetti-like control system must be replaced by a single, integrated chip. Think of it as the transition from the room-sized mainframe computers of the 1960s to the microchips of the 1970s and beyond. That transition wasn’t an innovation in abstract mathematics; it was an industrial engineering marvel. We need cryogenic integrated circuits to operate at the very low temperatures required for superconducting qubits. Using this approach, we can put not hundreds but 20,000 high-fidelity qubits on a single, clean wafer, and then achieve the target of millions of qubits per system by interconnecting those wafers. Quantum computing must adopt state-of-the-art chip manufacturing — the same technology that builds billions of transistors into every modern smartphone. This means getting rid of outdated, inefficient methods, such as the 60-year-old lift-off fabrication process used in the development of quantum computing chips, which simply is not clean or scalable enough.

He sees this as an industrial engineering challenge for the US and wonders whether the modern culture distracts from its realisation.

When the classical semiconductor industry offshored much of its fabrication capacity, it shifted technological leadership overseas. I do not wish my scientific legacy to simply mint a few more billionaires... I wonder whether modern culture, with its focus on the latest result and aggressive marketing, makes the necessary, difficult and frankly less glamorous work of deep industrial engineering harder to justify and fund. But the path to scalable quantum computers is paved with high-tech fabrication equipment, not just high-impact papers. It is time for the superconducting qubit community to shift its focus from chasing the next algorithmic demonstration to tackling the immense manufacturing and engineering challenge that lies ahead. The moment for foundational scientific discovery needs to give way to the era of industrial manufacturing.

Weekend reading links

1. Subsea cable is fast becoming a sector of strategic importance. Globally, there are four major companies - Japan's NEC, New Jersey-based SubCom, France's state-owned Alcatel Submarine Networks (ASN), and China's HMN Tech, a former Huawei subsidiary. The last three own cable-laying shipping fleets, while the first charters its vessels and is now set to receive government support to buy ships to put them on a par with their counterparts. A subsea cable-laying vessel could cost about $300 million. 

It currently relies on a subsea cable-laying vessel leased from a Norwegian group in 2022 on a four-year charter, partnerships with other companies and on renting the specialist ships on an ad hoc basis to meet surging demand for fibre-optic cabling in the Indo-Pacific. NEC dominates installations in Asia and has laid more than 400,000km of cables globally. It also specialises in armoured cables that can better withstand sabotage. Globally, there are 63 cable-laying ships, according to the International Cable Protection Committee. ASN owns seven, while SubCom and HMN Tech are believed to own seven and two, respectively, but did not reply to requests for confirmation. Japanese telecoms groups NTT and KDDI own cable-laying vessels, which are rented out to NEC, but they are not the larger kind of vessels required to lay transocean cables.

This is a description of the challenges being faced by the industry.

A shortage of ships is one of the major bottlenecks to achieving the 26 per cent rise per year predicted for global data transmission to 2031, driven by video streaming and artificial intelligence services led by Big Tech groups such as Meta and Google, according to the TeleGeography telecoms data service... Up to 200 cables are damaged annually, primarily due to fishing or anchors, but also through sabotage, as seen in the disruption caused to two cables in the Baltic Sea last year. Chartering for an accurate period has become increasingly difficult due to the unpredictable timelines for securing the cables’ passage. There is now a two-to-three-year wait to get permissions from countries whose waters the cables pass through, up from six months to a year about a decade ago.

2. Salaries in India have not kept pace, even with inflation in the FY16-24 period.

More here. 

From Diwali 2023 onwards, Indian companies’ earnings growth has decelerated at a rapid rate. Underpinning this deceleration is a sharp conk-off in consumption growth, long the mainstay of the Indian economy. Key drivers of this consumption downturn are a sharp deceleration in white collar job creation alongside a reduction in real wages for white collar workers over the past eight years.

And here

This should be a matter of deep concern.

The weak consumption demand will constrain India's sustained high growth ambitions. It will restrict private investment, limit job creation, squeeze salary growth, and increase household indebtedness. The only solution is broad-based, equitable economic growth. But that, in turn, requires conditions that are far from 

3. Rhodium Group has a report on the challenges faced by foreign car manufacturers in China. The German car makers and Tesla have the largest China exposure, while the Japanese and Korean car makers have far smaller engagement. 

It is important to note that while Tesla is likely even more dependent on China for profits than German OEMs, the sources of those profits are fundamentally different. German automakers still earn most of their China income—though at shrinking margins—by selling vehicles to Chinese consumers, whether locally produced or imported. Tesla, by contrast, likely earns the bulk of its China profits from two sources: the sale of regulatory credits—its CFO disclosed that three-quarters of Tesla’s global credit sales in 2024 (Q1–Q3) occurred in China—and a highly profitable export business built on China’s low production costs. In short, German OEMs depend on Chinese consumers, while Tesla depends on Chinese workers and credits... The experience of Toyota, Hyundai, and Kia shows, however, that success in China is not a prerequisite for global competitiveness. Toyota has become the world’s largest carmaker despite a relatively modest China footprint. Hyundai and Kia, for their part, have grown despite running loss-making China operations after the 2017 THAAD dispute triggered consumer boycotts that cratered sales.

4. The US Congress has become ideologically polarised since the turn of the millennium.

And immigration has been at the forefront of the polarisation.

5. A summary of the destructive chainsaw Trump has taken to the US Science establishment.

The proposed 2026 Budget outlines cuts of 34 per cent for basic research and 22 per cent for all research. It demands a 55.7 per cent cut for the National Science Foundation (NSF) — from $8.8 billion to $3.9 billion — and a 39.3 per cent cut for the National Institutes of Health (NIH) — from $46 billion to $27.9 billion. These two agencies are the primary sources for TRL 1-2 basic research. The story for TRL 3-6 applied research is also rough. The Department of Energy’s (DoE’s) Office of Science faces a 14 per cent cut, and Nasa’s science-research budget is slated for a 46.6 per cent reduction — from $7.3 billion to $3.9 billion... 

The Nazi government, which took charge in 1933, was a populist one, harnessing mass anger against the cosmopolitan elites. An estimated 25 per cent of all physicists in Germany, including 11 past or future Nobel laureates like Albert Einstein, Max Born, and Leo Szilard, fled Germany. The best research organisations of the world — like the University of Gottingen — were destroyed. It only took one year. In 1934, the great mathematician David Hilbert said to the Nazi education minister “mathematics in Gottingen? There is really no such thing any more”.

6. Comparing dotcom era telecom investments with AI investments today.

There were more than 80mn miles of fibre optic cable laid from the mid-1990s until the end of the dotcom boom, and much of that investment took years to pay off. US telecom companies spent $444bn in capital expenditure between 1996 and 2001. Still, compare this with the $342bn that will be spent this year alone in the US by the top investors in AI data centres and computing infrastructure, including Microsoft, Alphabet, Amazon and Meta. At the current rate of power consumption needed to fuel AI development, estimated investment will stretch to nearly $7tn by 2030.

7. Nvidia makes a $1 bn investment in Nokia to take a 2.9% stake in the Finnish telecom manufacturer, following which the shares in Nokia surged 21% and Nvidia by 5%. Nokia is seeking to diversify away from network infrastructure into AI and cloud services. 

This is part of the vendor-financing model of the emerging AI market. Is Ericsson next for Nvidia?

8. Two graphics that question the argument that we are in a bubble. One, the PE multiples of the Big Tech firms compared to those in earlier bubbles. 

The capex boom is largely (at least till now) being funded by free cash flows, unlike debt in earlier booms.

9. It's useful to keep in mind that much of the infrastructure in the West was built long ago.

In Britain we are using rail lines, bridges and sewers built by the Victorians, and even the Romans’ roads. It has taken over 150 years for London to need an expansion of the sewerage system begun by Joseph Bazalgette in 1859 (and largely finished within a decade). The US built its railroads in the 19th century, and interstate highways from the mid-1950s.

10. The reforms currently underway in the US to reverse safeguards in the banking sector in the name of deregulation may be an instance of bad deregulation. 

Last week, the Federal Reserve announced plans to overhaul its annual banking stress tests to make them less onerous. US banking watchdogs are widely expected to follow up with other changes to capital and leverage rules that could unlock $2.6tn in additional lending capacity, according to consultants Alvarez & Marsal. To backers, there is a logic to the Trump administration’s moves. Shackling banks with high capital requirements has not eliminated risky lending. Instead it has led to regulatory arbitrage that makes the danger harder to supervise, they say. As the IMF pointed out, banks now lend to private capital, which uses the funds to leverage investor money while making loans and buying securitised debt. In theory, the investors absorb the first losses, keeping bank deposits safe. In reality, layers of borrowing by companies like First Brands make it hard to tell who is on the hook and may lead to complacency. If banks lent directly, they say, they would do more due diligence and pick their borrowers more carefully... 

Another Trump administration initiative to allow ordinary investors to put their money in alternative assets, which have long been restricted to institutions and the super wealthy. Those changes are expected to channel floods of retail and retirement money to private capital groups, giving them bigger pots with which to make loans and buy asset-backed securities. These retail funds will be under particular pressure to deploy capital quickly because of the way they are structured... Bank of England governor Andrew Bailey said last week that “alarm bells” are going off around the rapid growth of structured products, and JPMorgan chief executive Jamie Dimon has proclaimed that the recent collapses of subprime auto lender Tricolor and car-parts maker First Brands are evidence of “cockroaches” in the credit market. The failures have uncovered complex webs of borrowing and allegations of fraud, leading Apollo chief Marc Rowan to warn that eroding lending standards are leading to “late-cycle accidents”.

11. Stock market concentration in the US is at all time high.

Eight of the 10 biggest stocks in the S&P 500 are tech stocks. Those eight companies account for 36 per cent of the entire US market’s value, 60 per cent of the gains in the index since the market bottomed in April and almost 80 per cent of the S&P 500’s net income growth in the last year... MSCI All World index, which comprises over 2,000 companies from more than 40 markets, currently has almost a quarter of its capitalisation in just eight US tech groups.

 On Tuesday afternoon, when US Stocks hit their latest highs, 397 stocks in the S&P 500 lost ground. In 35 years, the index never posted a gain on a day when do many of its components sold off. 

Since the launch of ChatGPT in November 2022, the US markets have been on a tear, underpinned by AI stocks.

This is an interesting snippet.

Since 1970, the total value of all publicly traded US stocks has averaged about 85 per cent of US GDP. Warren Buffett once described this as “probably the single best measure of where valuations stand at any given moment”. On Tuesday, the metric rose to a record 225 per cent.

12. More on the First Brands bankruptcy case in the US in this story of how a small Draper, Utah-based equipment finance specialist, Onset Financial, ended up with a $1.9 bn loan exposure to the Ohio-based automotive parts maker which borrowed close to $12 bn to finance acquisitions. 

First Brands’ reliance on Onset, which claims to eschew the “rigid” and “strict” approach of banks in favour of “speed” and “flexibility”, illustrates how unconventional corners of credit markets facilitated its borrowing binge, with many of the Ohio-based company’s lenders unaware of the true scale of its debts until it was too late... Onset’s corporate identity to date has been marked by promotional videos, a high-tempo sales culture and links to both prominent local investment firms and sports players. The fallout could ripple through the community in Utah, where Onset has built up an image of growth and glamour. Founded in the depths of the 2008 financial crisis in Draper, a small city 20 miles south of Utah’s state capital Salt Lake City, Onset’s triumph over stodgy rivals in a key area of business lending is a recurring theme of its corporate lore. Equipment leasing is a $1.3tn industry in the US that allows companies to rent machinery rather than sink large amounts of upfront capital into building or improving their facilities... People familiar with Onset’s operations describe a model fuelled by a direct and ambitious sales force. “What does our product do? We sell money, simple as that,” Taylor Weeks, Onset’s vice-president of sales, told a podcast in 2023. Weeks added that the company provides “rocket fuel” for “high-growth” businesses.

Equipment leasing is a $1.3 trillion industry!

13. FT writes about the rise of China's biotech firms, licensing technology and selling drugs outside the country. From having no biotech sector to speak of ten years back, in the first eight months of 2025, there have been 93 overseas licensing deals worth a total of $85 bn on drugs developed in China. 

China’s transformation from a copycat manufacturer of drugs developed overseas to a hub of homegrown research is exemplified by Jiangsu Hengrui. Founded in 1970, it spent the first two decades as a small-scale state-owned manufacturer of low-cost antiseptics. In the 1990s, it started developing generic anticancer drugs. It was privatised in 1997 and began investing in building its own research capabilities. Today, it has one of the most diversified pipelines in the country, spanning weight-loss therapies, oncology drugs and Alzheimer’s treatments...

Hengrui has struck licensing agreements with Merck, Braveheart Bio and Glenmark in the past year alone. In July, it agreed a deal with UK pharmaceutical company GSK to develop up to 12 medicines. “In China, you can scale and test medicines in human beings much faster than in the US or Europe,” said Loncar. “If you have an idea for a drug, you can get an answer about whether it works a year or two earlier in China.” For many Chinese biotechs, the surge in international partnerships has provided much-needed capital after a difficult few years marked by drug pricing reforms that squeezed profit margins on domestic sales... Hengrui’s international deals have highlighted a concern for Chinese pharma companies seeking to get international approval for new drugs. The US Food and Drug Administration has repeatedly rejected one of Hengrui’s cancer drugs, citing questions about quality control at its manufacturing sites.

US and European pharma companies are doing what their manufacturing counterparts elsewhere did by outsourcing to China, only to realise that they have been outmuscled by them over time. 

This also raises questions about where India's established pharmaceutical firms are. 

14. Finally, NYT has a good article on OpenAI's circular financing deals. 

Many of the deals OpenAI has struck — with chipmakers, cloud computing companies and others — are strangely circular. OpenAI receives billions from tech companies before sending those billions back to the same companies to pay for computing power and other services.

Process knowledge to build manufacturing capabilities

China may be the best modern illustration of Joel Mokyr’s argument that growth and development happen when propositional knowledge is translated into prescriptive knowledge through dense networks. Without explicitly articulating using this framework, the narratives described in Patrick McGee’s Apple in China (see this and this) and Dan Wang’s Breakneck demonstrate the essence of Mokyr’s argument.

The fundamental point is that both Apple (through iPhone) and Western multinational corporations in general helped China by transferring useful knowledge (what Wang calls ‘process knowledge’) and creating rich industrial ecosystems seeped with technical expertise acquired iteratively over several years. It helped that China, like the pre-Industrial Revolution England, had a culture that gave primacy to the acquisition of useful knowledge over theoretical propositional knowledge. 

I just completed Dan Wang’s book, Breakneck. The big premise of the book, the reason why the US is lagging behind China in making things is because the former is ruled by lawyers and the latter by engineers, is too simplistic and questionable. Jonathon Sine has an excellent critique. 

However, the book has a compelling explanation for how and why China built up its manufacturing prowess. It echoes the description by McGee of how Apple’s intense and long-drawn engagement in the co-creation of Foxconn’s iPhone manufacturing capabilities played a critical role in the development of China’s world-beating manufacturing ecosystem of designers, tooling engineers, component suppliers, managers, and assembly-line workers. This is a good description of the emergence of the ecosystem.

The smartphone components were getting better every year, part of a trend that Chris Anderson, former editor of Wired, called “the peace dividend of the smartphone wars.” The hundreds of billions of dollars invested in the smartphone supply chain have caused the cost of electronic components – cameras, sensors, batteries, modems – to plummet. That’s why we are able to carry around sensors in our pockets that used to be available to only a select few military powers.

Many companies have grown around this peace dividend. Indeed, Shenzhen is the headquarters of many of China’s most dynamic companies, including BYD, the world’s largest EV maker; DJL, the world’s largest consumer drone maker; and Huawei, the beleaguered company that is the world’s largest telecommunications equipment maker. Electric vehicles are full of electronic components borrowed from smartphones; the consumer drone is roughly a reassembly of a smartphone camera and sensor with propellers for flight. The magic of Shenzhen in the combination of the world’s most creative hardware engineers sitting in a sea of components that improve every year amid a labour force of millions who know how to put together electronics. This buzzing ecosystem has produced many other products that follow in Apple’s wake, like hoverboards, electric scooters, virtual reality headsets, and who knows what’s next?

Wang highlights the importance of three aspects of technology in production, specifically the last aspect of process knowledge. 

First, technology means tools. These are the pots, pans, knives, and ovens required to prepare a dish. Second, technology means explicit instruction. These are the recipes, the blueprints, the patents that can be written down. Third and most important, technology is process knowledge. That is the proficiency gained from practical experience, which isn’t easily communicated. Ask someone who has never cooked before to do something as simple as fry an egg. Give him a beautiful kitchen and the most exquisitely detailed recipe, and he might still make a mess…

Process knowledge is hard to measure because it exists mostly in people’s heads and the pattern of their relationships to other technical workers. We tend to refer to these intangibles as know-how, institutional memory, or tacit knowledge. They are embodied by an experienced workforce like Shenzhen’s. There, someone might work at an iPhone plant one year, for a rival phone maker the next, and then start a drone company. If an engineer in Shenzhen has an idea for a new product, it’s easy to tap into an eager network of investors. Shenzhen is a community of engineering practice where factory owners, skilled engineers, entrepreneurs, investors, and researchers mix with the world’s most experienced workforce at producing high-end electronics.

Silicon Valley used to be like this too, but now it lacks a critical link in the chain – the manufacturing workforce. The value of these communities of engineering practice is greater than any single company or engineer. Rather, they have to be understood as ecosystems of technology. The American imagination has been too focused on the creation of tooling and blueprints. Andy Grove, the legendary former CEO of Intel, said it best in 2010: that the US needs to focus less on “the mythical moment of creation” and more on the “scaling up” of products. Grove saw Silicon Valley transition from doing both invention and production to specialising only in the former. And he understood quite well that technology ecosystems would rust if the research and development no longer had a learning loop from the production process…

American manufacturers spent the better part of the last three decades unwinding its stock of process knowledge when it opened so many factories in China. Every US factory closure represents a likely permanent loss of production skill and knowledge. Line workers, machinists, and product designers are thrown out of work; then their suppliers and technical advisers struggle as well. Entire American communities of engineering practice have dissolved, leaving behind a region known as the Rust Belt. Some mayors and governors tried to step this receding tide. But they were continuously scorned by economists and executives, who sought low-wage production in the name of globalisation. Still today, many American economists doubt there is anything special about manufacturing and put their faith in the inevitable march to a service economy…

It became part of the elite consensus that the US could lose manufacturing. This consensus portrayed union bosses, as well as the handful of heterodox economists, as sentimentalists for resisting offshoring. Neither the Clinton nor the George W Bush administration restrained American firms from moving manufacturing operations to China. Now, it’s more obvious that the departure of manufacturing has created economic and political ruination for the US. We are still only beginning to understand how much it set the country back technologically…

If we think about technology ecosystems as communities of engineering practice, it makes sense that factory closures accelerated as process knowledge dissolved, prompting production problems and more job losses. And it also makes sense that Chinese workers went from merely assembling iPhones to producing some of their most valuable components as well. As one country lost its process knowledge, the other gained whole industries.

He describes how Tesla contributed to the development of China’s emerging dominance of EV car manufacturing.

Beijing did something unprecedented for Tesla in 2018: it allowed the company to fully own its plant in Shanghai. Previously, any automaker that wanted to produce in China had to partner with a domestic company. So Japanese, German, and American companies dutifully partnered with state-owned enterprises in order to access the enormous market. The state had hoped that these domestic companies would learn from the likes of Toyota and Mercedes-Benz and match their quality. In reality, Chinese automakers were sluggish from their research dependence on their foreign friends.

Tesla’s presence jolted China’s EV market. China’s business community began using the term “catfishing” for what Tesla was doing in China. The idea was that introducing a powerful new creature into the domestic environment would make Chinese firms swim faster. That’s exactly what they did to raise their game. When Tesla vehicles started rolling out of the Shanghai Gigafactory in 2019, BYD saw its sales decline by 11%, while profits fell by 42%. But Tesla would eventually do the whole market a favour. As in the US, the company’s audacious branding stimulated consumers to think of EVs as more than high-powered golf carts. And Tesla made investments in China’s tooling ecosystem that other automakers exploited to produce better cars. BYD benefited as well, reporting record profits in 2023 and becoming the world’s largest EV maker. And even the Communist Party’s main newspaper praised how Tesla produced the “catfish effect” for Chinese firms.

As Grace Wang, founder of Shenzhen-based Luxshare poetically expressed, “Flying with phoenixes will nurture outstanding birds.” It is another lesson that capitalist Shenzhen has taught the Communist Party: Market competition tends to lower prices and raise quality. 

Apple and Tesla have made a huge effort to train its Chinese workers to manufacture their products – and earned fabulous sums of money by doing so. These stories are replicated in varying degrees across China’s other communities of engineering practice, production hubs for shoes and garments in the eastern city of Wenzhou, medical equipment in Wuxi and Suzhou, and, most wonderfully of all, guitars in the mountains of Guizhou’s Zheng’an County. Overall, China’s manufacturing workforce employs more than a hundred million people, around eight times that of the US (13 million in 2025)… 

American companies have spent two decades building communities of engineering practice in China, made up of people who roll up their sleeves to figure out how to overcome their daily bottlenecks.

At the heart of this narrative about economic growth is the importance of persistent and boring implementation, over the Aha! moment of inspiration presented by ideas and inventions. This is a good summary:

Americans expect innovations from scientists working at NASA, in universities, or in research labs. They celebrate the moment of invention: the first solar cell, the first personal computer, first in flight. In China, on the other hand, tech innovation emerges from the factory floor, when a new product is scaled up into mass production. At the heart of China’s ascendancy in advanced technology is its spectacular capacity for learning by doing and consistently improving things… The US likes to celebrate the light-bulb moment of genius innovators. But there is, I submit, more glory in having big firms making a product rather than a science lab claiming its invention… Every day, millions of workers in factories to build up technological process knowledge. That is the basis of China’s tech power. China has become a tech superpower by exalting process knowledge and the communities of engineering practice that keep it alive.

There are a few observations that are of relevance for India as it charts its journey of building a globally competitive manufacturing base. 

1. Mokyr points out that the Industrial Revolution happened in England and not continental Europe, and attributes this to its culture of tinkering and refining, and interest in material progress. This required useful knowledge that the fabricants could use, as against the more theoretical and esoteric knowledge that savants created. 

Similarly, China had a culture of pursuing process knowledge. When faced with competition from Apple and Tesla, instead of folding up and retreating, these companies leveraged the ecosystems of process knowledge and engineering practice created by Apple and Tesla. They fought back to build their own products and compete (and outcompete) the foreigners. Huawei, Oppo, Xiaomi, Vivo, BYD, Geely, SAIC, Nio, Xpeng, and so on are the outcomes of this. 

Do Indian manufacturers and startups have the culture to embrace process knowledge, iterate diligently, build ecosystems of engineering practice, and move up the value chain? What are examples from India of such technology or industrial ecosystems that stand at the cutting-edge of their sectors?

2. The conventional wisdom on industrial progress, especially in technology-intensive sectors, is that of academia and industry working together to co-create products by transferring research into development at an industrial scale. While this is the approach that the US and other Western economies followed, China appears to have taken a different direction. Instead of spending time on the generation of propositional knowledge in its research institutions, it borrowed (or copied) this knowledge from wherever available and focused on the prescriptive knowledge required to scale up manufacturing. This is the classic industrial tinkering model.

Chinese firms rely on academia more in terms of getting skilled manpower to work in their factories. With the government’s guidance, Chinese colleges and universities have been running courses on niche areas like battery chemistry and rare earth processing for several years. They have kept flowing a continuous supply of high quality skilled workforce. 

This is of particular relevance for India, where academia-industry collaboration is limited, but also where industrial R&D is abysmally low (0.7% of GDP compared to China’s above 2.5% of GDP). Indian firms must necessarily multiply their R&D expenses manifold if they are to create the vibrant and high-productivity industrial ecosystems like those in China. They must invest in nurturing a culture of engineering excellence that seeks to constantly innovate and aspire to the global frontier of quality and technology. 

3. Another feature of Mokyr’s arguments and China’s success is the importance of clusters in producing things. A manufacturing base does not emerge in isolation from the production of individual firms. It requires the creation of ecosystems of engineering practice, where skills and expertise are transferred through learning by doing, and knowledge and technology spillovers. Clusters enable the diffusion and spread of process knowledge among businesses and their suppliers, and among competing firms. 

Massive industrial clusters have been central to the development of China’s manufacturing prowess. China has more than 500 towns that specialise in specific products for the global market, with some being responsible for 63% of world’s shoes, 70% of its spectacles, and 90% of its energy-saving lamps. Sample this from Wang,

Every year, as new models emerge, Apple needs new components or processes that a new design requires, like a certain type of adhesive or a screw of a slightly different size. Therefore, Apple constantly had to scramble to find suppliers on short notice. “Almost always, “the engineer continued, “we found someone in Shenzhen by asking a guy who knows a guy whose cousin might be able to produce a few hundred thousand new screws.” Virtually everything one needs to produce any electronic product can be found in a short drive around Shenzhen. Proximity creates efficiency. When it’s time to do stuff, a company can collapse coordination that usually takes weeks into a business meeting lasting hours by convening all the relevant suppliers in one room the next morning. And if something goes wrong, there are a lot of friendly neighbouring factories to call.

Unfortunately, India’s efforts in this direction, primarily in the form of SEZs, have been constrained by the size of these clusters. I blogged here on how the lack of clusters has been a constraint to the emergence of scale manufacturing in India. 

4. The creation of globally competitive ecosystems of engineering practice that value process knowledge for continuous improvement invariably sets in motion something like the Red Queen Effect. Companies must adapt and evolve continuously to even maintain the relative status quo in the highly competitive and dynamic global marketplace. 

Only firms exposed to global competition have the incentive to pursue this strategy. The incentives of firms in a closed market, or those happy with serving a large segment of their domestic market, will be only to minimise costs. They are likely to discount quality and technology. In simple terms, Indian firms must necessarily Make in India for the World.

There’s the real risk that Indian firms will remain entrapped in the bad equilibrium of making less than competitive products for India’s large price-sensitive consumer base. 

5. The example of how Chinese companies responded to the competition from Tesla is a testament to the aspirations of Chinese companies. In a short period of time, BYD and Co. were able to turn the tables on Tesla and emerge as undisputed global leaders in EVs. They first copied the engineering excellence of Tesla and continuously improved on it to then gradually surpass it. Underpinning this was the communities of engineering practice created by Tesla. 

In contrast, as I have blogged on numerous occasions, Indian companies, across sectors, have struggled to innovate and become globally competitive. 

The Ken has an article about how the garment makers of Tirrupur, who contribute over half of all India’s knitwear exports, failed to innovate and are now fighting to survive the loss of the US market. They remained stuck with the commoditised and lower value-added bulk supplies of cotton garments to the large US retail clients, with their low margins and large volumes, but stable and assured demand. They missed the opportunity to shift to man-made fabrics earlier and to blended fabrics in recent years, and did not venture into the higher value-added and higher margin mid-sized European who are much more demanding on fashion trends and quality. 

It will be extremely challenging for Indian companies to match the resolve and ambition shown by the Chinese EV manufacturers, nurture communities of engineering excellence, and become globally competitive. 

In conclusion, stripped of all the jargon and in simple terms, Mokyr’s work and the success of China is a big shout-out to the strategy of progress through continuous problem-solving and iterative adaptation, one which has relevance far beyond mere building of manufacturing capabilities to many things in life in general.