Industrial policy in the age of automation

One of the most disturbing economic trends of our time is that of technological advances automating work and displacing labour. While in its early stages, there are compelling reasons to believe that unlike with earlier technologies, robots and automation could significantly shrink the labour market. 

This, by itself, should be a matter of deep concern. But this trend also adversely impacts labour’s bargaining power with capital, thereby creating the conditions for worsening the already stark inequities in the sharing of capital returns. 

The prevailing incentives and power relations of the market, left to itself, will invariably encourage firms to deepen and accelerate the adoption of labour-minimising technologies. This market failure will necessitate public policy action.

A recent FT long read examined the impact of the strike last October by some 25,000 members of the International Longshoremen’s Association (ILA) that grounded three dozen container ports on the US east and Gulf coasts, which handle one-quarter of the country’s international trade and cost the economy up to $4.5 bn a day. The strike was withdrawn after just 72 hours following negotiations and an offer of a salary increase of nearly 62% over six years. 

But, as the FT article points out, the main issue goes beyond salaries and is existential for workers.

Although it was the pay rise that caught the attention of the media, the union’s real issue is with automation — specifically proposals by the United States Maritime Alliance (USMX), which represents port operators and container carriers, to equip more US ports with semi-automated cranes. These cranes are equipped with advanced technology that makes them faster and more efficient to operate, say the owners. But the ILA claims that their introduction threatens their members’ livelihoods. Unless USMX agrees to a total ban on automated machinery, the union has threatened to strike again as early as next week. “We embrace technologies that improve safety and efficiency,” the ILA’s colourful president, Harold Daggett, said in a statement. “But only when a human being remains at the helm.”

… As more and more businesses experiment with next-generation robotics, US labour unions representing industries as varied as UPS drivers, Las Vegas casino workers and grocery store employees are fighting for provisions to be added to contracts that focus on retaining jobs and compensating displaced workers in the event of automation. What were previously run-of-the-mill negotiations over pay and conditions have mushroomed into larger, more existential disputes over the relationship between humans and machines… Whatever contract the longshoremen negotiate, say analysts, could help provide a template for agreements nationwide.

Turbocharged by the advances in artificial intelligence, robots are being experimented across sectors.

Since General Motors first put robots on assembly lines in the 1960s, carmakers have been pioneers in automation. Yet until the rise of AI, other industries — ones requiring more dexterous tasks, or where robots might need to respond to unpredictable or hazardous environments — struggled to follow suit… Manufacturing companies in particular have invested heavily, with total installations of industrial robots rising by 12 per cent to over 44,000 units in 2023 — the largest volume in at least a decade, according to the International Federation of Robotics. Again, the car industry has led the way, followed by electrical and electronics companies… 

On their annual trip to the Consumer Electronics Show in Las Vegas last year, members of the Culinary Union, which represents staff at casinos in the city, were shocked to see robots frying food and making cocktails… In 2022, real estate developer The Durst Organization’s venture arm invested in the maker of a glass-washing robot, Skyline Robotics, based in Israel. The Ozmo robot can now be seen scrubbing the windows of a skyscraper near Times Square… “It’s very understandable to me why that next generation isn’t showing up,” says Skyline Robotics president Ross Blum. “It is a really tough job . . . Who wants to go hang 1,000 feet in the air today and do manual labour outdoors?”

Worsening the problem is the challenges associated with labour market adjustments arising from automation. 

MIT economist Daron Acemoglu says that robots’ current capabilities mean that those most at risk of being displaced are in blue-collar jobs and lack college degrees, which may make it difficult for them to shift into the high-tech roles likely to be created by automation.

Consider the market forces at play that have driven the search for labour-saving technologies

Jobs that looked like they could only be done by people suddenly look risky; economists have warned of wholesale and disruptive changes to the workforce as machines are capable of more and more… Adding to the pressure in economies like the US, say business owners, is sluggish growth in the labour force, which is making it increasingly hard to recruit workers. President-elect Donald Trump’s plans for mass deportations… will probably only intensify such concerns… Salary increases experienced by many Americans in the past few years have come at a cost… it makes the US somewhat uncompetitive… As the population ages and families struggle to find childcare, the share of Americans in work or seeking work has been declining for decades — dropping from 67.3 per cent in 2000 to 62.5 per cent late last year. Economists estimate that it will sink to 60.4 per cent by 2030.

US investors have piled more than $15bn into robotics start-ups since 2019, according to PitchBook, and the remarkable growth of artificial intelligence in the past 18 months has begun to show dividends… US venture capital investment in robotics has risen from around $2bn in 2019 to more than $3.5bn last year, according to data from PitchBook. In the first nine months of 2024, there were 130 fundraising deals for robotics start-ups — more than across the entirety of 2019. Among the most high-profile was a $675mn investment last February by Amazon founder Jeff Bezos, Microsoft and Nvidia in Figure AI, a Silicon Valley start-up founded in 2022 that is working on a faceless, humanoid “general-purpose” robot. It said that month that these robots — whose cost to customers is estimated at between ​​$30,000 and $150,000 — could complete tasks including moving a box on to a conveyor belt, potentially endangering the job of anyone working in, say, a distribution centre. The first models were delivered to a “commercial client” last month.

The article points to the ILA’s concern at containerisation being a case of once bitten twice shy.

Before the advent of containerisation, longshoremen spent long days unloading individual boxes, barrels and crates, then transferring their contents on to trucks and freight trains — dangerous but reliable, well-paid work that, at its peak, employed an estimated 100,000 men in ports around the US. After the trucking entrepreneur Malcom McLean championed the 8ft-wide steel container in the mid-1950s, that world fell away. The new technology meant that cargo could be transferred with a minimum of effort and drastically reduced costs. Tens of thousands of jobs disappeared almost overnight. Despite a huge increase in world exports, the number of longshoremen employed at the Port of New York and New Jersey plummeted from 55,000 in the 1950s to about 4,000 today, says Jean-Paul Rodrigue, a professor of maritime business at Texas A&M University… 

When semi-automated cranes were first brought in to terminals on the east coast of the US in the early 2000s, ILA leaders say they agreed to the changes because it would help create jobs. But they now say that the opposite happened… A 2022 survey commissioned by the west coast dockworkers’ union found that partial automation of the ports of Los Angeles and Long Beach resulted in the loss of nearly 1,200 jobs in 2020 and 2021. USMX says that because most of the ports its members operate have no spare land available, the only choice is to “densify terminals” by adding machinery that speeds up operations. In a conventional crane, an operator sits inside a cab, lifting containers off ships and sorting them, before transferring them to trucks or trains — a highly skilled job that can earn workers as much as $200,000 annually. In a semi-automated rail-mounted gantry crane (RMG) system, the operator works remotely from an off-site office, monitoring the crane via video link but letting the system do most of the work. The job requires similar skills and training, but fewer people are required… But USMX describes calls to ban automation as “unworkable”, saying that modern crane technology has “nearly doubled” both the throughput of containers and the number of workers at the ports using it.

So how is labour adjusting to the emerging trends?

In recent years, both retail and culinary unions have negotiated clauses in contracts they hope will protect human workers. Las Vegas casinos are now required to give people six months’ notice before implementing new technologies and free training on how to use them, plus severance packages for anyone laid off because of technology. UPS has agreed to negotiate with the Teamsters, one of the most powerful unions in the US, before introducing drones or driverless pick-up vehicles. New York retail stores whose workers are represented by RWDSU, including Bloomingdale’s and Macy’s, also require management to come to an agreement before introducing new technologies.

This has relevance to developing countries like India.

The adoption of labour-saving technologies by companies in developing countries is likely to lag behind those in developed countries. This is a good thing. Given the abundance of cheap labour, the economic case for automation has limits. Further, the vast informal sector, with its deeply price-sensitive consumers, too is unlikely to be disrupted significantly. 

However, the pace of adoption will be significantly higher in those sectors competing globally and among the biggest firms, which are also the creators of good-paying jobs. Consider AI-based automation displacing lower and middle-level software jobs, which comprise the vast majority of well-paying services sector jobs. Or the automation of assembly lines in textiles, footwear, electronics, consumer durables, automotive etc., which make up the vast majority of well-paying manufacturing jobs. The disruption in the former can be immediate and more gradual in the latter. 

In other words, in developing country contexts, the biggest adverse impact of labour displacing technologies is likely to be in the large layer of good jobs that pay reasonably well at the entry and middle levels. They are the lifeboats to enter the middle class.

If this trend starts to surface, it’ll invariably trigger a backlash and create the conditions for populist narratives, with all their risks of distortions and perversions. It’s therefore important that governments realise the strong likelihood of such developments and calibrate policies in this direction. 

Accordingly, there’s an increasingly strong case to align industrial policy to both encourage job-creating economic activities and also discourage the adoption of labour-saving practices and technologies. While the former has always been a salient feature of industrial policy, it’s now time for the adoption of the latter. 

This would entail shifting away from capital investments that lead to automation towards labour subsidies. The PM Internship Scheme, if scaled up in a realistic manner, is the right kind of policy for the times. As an illustration, a Rs 100 Cr capex subsidy is equivalent to providing 50% of the salary for five years for 1250 workers with a starting cost to the company of around Rs 22000 and increasing annually at 10%. At a more realistic 33% and 3 years, the same labour subsidy can employ nearly 3500 workers. This can be a significant incentive for businesses and an entry point for a better life for blue-collar entrants. 

Such industrial policy could discourage assembly line automation in manufacturing. Similarly, it should be examined as to whether tax policies can be carefully tweaked to encourage software companies to develop business models that can accommodate the displaced manpower in other activities/areas. Taxation policies that incentivise manufacturing and R&D should be revisited to keep these challenges in mind. 

From the labour market side, there will be demands to institutionalise as policy mandates the protections being negotiated in the US by labour unions. I’m not sure about its advisability. 

Such policies can create perverse incentives, and administering them can be hard. It’s therefore essential to be flexible with them. 

But they cannot be ignored for too long given the scarcity of good jobs for a rapidly growing pool of labour market entrants, and the strong likelihood of brewing discontent spilling over into some form of undesirable populist backlash.

There are no windfall revenue mobilisation opportunities in public policy

Not all great-sounding ideas can be materialised as outcomes. Nowhere is this more common than in development and public policy. 

There are several reasons for this dissonance between ideas and their adoption at scale. For a start, implementation of the concept runs into practical challenges, including weak execution capabilities, that compromise the fidelity of execution, thereby resulting in poor implementation and failure to realise the expected outcomes. This is arguably the most commonly observed reason for the dissonance in development. 

Then there are those problems, the so-called wicked problems, that are not amenable to a reductionist approach involving elegant, logically correct, and technically appealing ideas. Further, these problems generally generally require persistent and painstaking long-drawn engagement to start to show outcomes. 

There are those ideas that sound great but generally struggle to scale because there are inherent constraints and limitations to their scaling. For example, behavioural change interventions like a financial incentive or a framing nudge, when scaled (and over time), tend to lose their effectiveness. 

Finally, there are those great-sounding ideas that are over-hyped and over-sold, whose actual benefits fall far short of the projections of experts and commentators. 

Consider two such oft-repeated ideas for windfall revenue realisations - governments must monetise vacant land holdings, and increase revenues by expanding the tax base (by collecting long-pending revenue arrears or detecting large leakages or identifying new revenue streams).

Now, both ideas have some merit, but the expectations are grossly overestimated. They are based more on misplaced hope and not on a practical understanding of the contexts and underlying problems. Reinforcing this, apart from rare one-off examples, I don’t think there are any instances (across geographies and sectors, and over time) in India where windfall revenues have been realised from an institutionalised process. 

Take the example of the monetisation of vacant lands. For sure, there are a tiny handful of one-off instances where a few state governments, after immense efforts, have managed to auction a few parcels and realise significant amounts. But they must be seen against the numerous failed attempts made across almost all states, often repeatedly, to auction lands. 

Institutionalised and focused efforts in selling government lands have generally been a failure. Some states have established dedicated entities to raise revenues by selling government lands. The Government of India established the Land Monetisation Corporation (NLMC) with the exclusive mandate to raise revenues by monetising lands. None of these efforts have anything significant to show. In fact, I don’t think there is even one instance of a state or local government having institutionalised a process whereby they have been able to continuously raise significant revenues by selling government lands in a phased manner. 

An exception is that of urban development authorities and industrial development corporations in some states and that too on the suburbs of large, perhaps metropolitan, urban agglomerations. And these sales too have happened during the early phases of those cities’ emergence (think Gurgaon, Noida, Hyderabad etc.).

Globally too, I’m not aware of instances of successes with land monetisation. The Canada Lands Company is often touted as a global best practice in land monetisation. But, for an entity in existence since 1951, its nationwide cumulative sales realisation and land holdings are too small to be considered anything but a pilot. 

Governments, and especially certain departments and entities, undoubtedly hold vast extents of vacant land. But their auction sales run into several daunting problems. For one, land prices in urban India are already too high that markets cannot absorb sales in significant volumes. Two, perhaps the biggest challenge to land auctions in India is the near inevitability of litigation. Every state and central agency has its share of large volumes of land locked up in litigation arising from land disputes (private counterclaims) or initiated by disgruntled or mischievous auction bidders. The value of lands auctioned without amounts being realised due to litigation will itself be a large amount. 

Three, government land auctions have deep political economy impacts. There will be local vested interests who lose out from the sales and the consequent redevelopment and local politicians who have an interest in maintaining the status quo. Finally, public land auctions present enormous direct and indirect rent-seeking opportunities that invariably engender scandals and controversies. Influential people with insider information indulge in speculative land transactions before such auctions, arm-twists the buyers to share rents, or manipulate markets to depress prices and knock off the auctioned land at lower prices. 

Instead of outright sales through auctions, more practical approaches to unlocking value from land would be sales to other government agencies for offices and other requirements (easiest), value capture concessions on Public Private Partnerships (involve significant transaction efforts), and allotment for affordable housing projects (this too requires structuring and transaction efforts). The only realistic land sales strategy would be to undertake phased and gradual sales of small land parcels in an institutionalised and transparent manner. 

The other example is that of increasing tax revenues. Again, I cannot think of any instance in the last two decades from India of a state or central government that has actually realised (as against merely raised demand and created a receivable) a sudden and significant increment (a windfall) in cash revenues from a new revenue source, or the detection of revenue leakages, or the collection of long pending arrears. For sure, there will be claims and even awards given to such claims, but if you dig deep the actual realisations in terms of significant amounts having been collected will be far lower. 

All actual revenue increases come from long-drawn efforts, and not from free lunches or dollars on the sidewalk. In fact, there are no such free lunches. 

Any incremental revenue realisation has counterparties who lose equivalent incomes. If the government’s revenue realisation is significant, so is the private individual’s (or entity’s) income loss. Such losses cannot not trigger immediate economic and political repercussions. The potential losers will therefore fight with all their might to avoid giving up their incomes. They will deny, delay, litigate, politicise, and corrupt the system to retain their incomes. It’s therefore no surprise that all revenue departments across governments have large amounts locked up in litigation, as arrears, and in procedural delays (which in turn end up in litigation and/or arrears). 

It’s for this reason that, as I blogged here, state and central tax and other revenue mobilisation agencies have a uniformly poor track record of actual realisations from the large amounts of demands raised by them. None of the direct and indirect taxation agencies of state and central governments can claim to have actually realised large amounts through investigations and enforcement actions even when they have raised large amounts of tax demands. Similarly, there are no instances of a new revenue stream emerging with windfall revenues realised. Nor are there instances of large realisations from long-pending arrears. 

Instead of expecting windfall realisations, government agencies should be encouraged to systematically identify revenue leakages and new revenue sources, and initiate efforts to prospectively assess, raise demand, and collect them. Broadening the tax and revenue base is the only realistic means to increase revenues. There are hardly any quick windfall revenue mobilisation opportunities in government.

Weekend reading links

1. Robin Harding writes that AI advances may not help much in the development of humanoid robots.

The obstacles to making an economically viable robot that can cook dinner and clean the toilets are a matter of hardware, not just software, and AI does not in itself address, let alone resolve them. These physical challenges are many and difficult. For example, a human arm or leg is moved by muscles, whereas a robotic limb must be actuated by motors. Each axis of motion through which the limb must move requires more motors. All of this is doable, as the robotic arms in factories demonstrate, but the high-performance motors, gears and transmissions involved create bulk, cost, power requirements and multiple components that can and will break down. After creating the desired motion, there is the challenge of sensing and feedback. If you pick up a piece of fruit, for example, then the human nerves in your hand will tell you how soft it feels and how hard you can afford to squeeze it. 

You can taste whether food is cooked and smell whether it is burning. None of those senses is easy to provide for a robot, and to the extent they are possible, they add more cost. Machine vision and AI may compensate, by observing whether the fruit is squashed or the food in the pan has gone the right colour, but they are an imperfect substitute. Then there is the issue of power. Any autonomous machine needs its own energy source. The robot arms in factories are plugged into the mains. They cannot move around. A humanoid robot is most likely to use a battery, but then there are trade-offs with bulk, power, strength, flexibility, operating time, usable life and cost. These are just some of the problems.

2. Nvidia does not seem to agree as it bets on robotocs as its next big growth driver. 

Nvidia... is set to launch its latest generation of compact computers for humanoid robots — dubbed Jetson Thor — in the first half of 2025. Nvidia is positioning itself to be the leading platform for what the tech group believes is an imminent robotics revolution. The company sells a “full stack” solution, from the layers of software for training AI-powered robots to the chips that go into them... The push into robotics comes as Nvidia is experiencing more competition for its powerful AI chips from rival chipmakers such as AMD, as well as cloud computing groups such as Amazon, Microsoft and Google that are looking to reduce their dependence on the US semiconductor group... a shift in the robotics market is being driven by two technological breakthroughs: the explosion of generative AI models and the ability to train robots on these foundational models using simulated environments. The latter has been a particularly significant development as it helps solve what roboticists call the “Sim-to-Real gap”, ensuring robots trained in virtual environments can operate effectively in the real world, he said... Nvidia offers tools at three stages of robotics development: software for training foundational models, which comes from Nvidia’s “DGX” system; simulations of real-world environments in its “Omniverse” platform; and the hardware to go inside the robots as its “brain”.

3. Pakistan drastically reduces pension benefits of retired civil and armed forces personnel to reduce its growing pension bill, the fourth largest expenditure in the budget. 

The Ministry of Finance on Wednesday issued three separate notifications to discontinue multiple pensions, reducing both the first home take pension and also lowering the base for determining future increases in pensions... According to the Ministry of Finance’s notification, on the recommendations of the Pay and Pension Commission of 2020, “it has been decided that henceforth, in an event where a person becomes entitled to more than one pension, such person shall only be authorised to opt to draw one of the pensions”... Instead of taking a pension on the basis of the last drawn salary, the new pensioner will get a pension based on the average salary of the last two years... It also ended the annual compounding of the pension and any increase would be treated separately from the base pension, a concept that is similar to the ad-hoc salary increase that is not made part of the basic salary to avoid compounding. The changes will take effect from January 1 and will be applicable to both retired civil and military personnel. Many serving federal government employees, who are taking salary and pension, would also be affected by the changes. The finance ministry’s notifications stated that the changes in the pension rules have been made on the basis of recommendations given by a commission constituted by the government of former prime minister Imran Khan in 2020.

4. A good description of how Chinese exporters are responding to US tariffs.

In industry after industry, Chinese companies have found footholds abroad that allow them to bypass trade barriers with the United States. After the United States put hefty tariffs on Chinese solar panels, for example, many Chinese companies opened solar factories in Southeast Asia... U.S. efforts to block critical minerals and electric vehicle batteries from China from receiving government subsidies have also pushed Chinese companies to set up battery-making subsidiaries in Morocco and Singapore... In some cases, global companies have also used accounting and tax tricks to make it appear that their shipments from China are lower, and thus pay fewer tariffs, without making major changes to their supply chains... For example, an electronics company might move one important stage of its supply chain out of China and into Vietnam. That could allow the company to report to U.S. customs agents that the export came from Vietnam, even if the good is still finished in China and exported from China to the United States. Another lever companies could play with, Ms. Brown said, is valuation. They can officially lower the value of the import by stripping out certain “intangible” costs, like payments for intellectual property, royalties, brand or research and development, and recording those to other global subsidiaries. By lowering the value of the import, they then pay a lower tariff.

5. Important point about the origins of uncertainty and risks in the global order today.

According to John Ikenberry of Princeton University, a leading theorist of international relations, “a revisionist state has arrived on the scene to contest the liberal international order . . . it is the United States... Trump is poised to contest almost every element of the liberal international order — trade, alliances, migration, multilateralism, solidarity between democracies, human rights”. As a result, rather than supporting the international status quo, the US is poised to become the leading disrupter. “Every talk I’ve ever given on the geopolitical risks that we face in the world started with China and Russia,” says Ivo Daalder of the Chicago Council on Global Affairs. “But the biggest risk is us. It’s America.” America’s traditional allies are among the countries that feel most threatened by a change in the way that the US exercises its power. Middle-power democracies such as the UK, Japan, Canada, South Korea, Germany and the entire EU have got used to a world in which American markets are open — and the US provides a security guarantee against threatening authoritarian powers... The question of whether and how to respond to Trump tariffs is exercising diplomatic minds across the western world. Finding an answer is all the more difficult because Trump’s true intentions remain unclear.

6. Interesting inequality trends

So we have seen no increase in aggregate inequality. The story for the lowest-paid is unambiguously good but for the bulk of people who sit somewhere in the middle, it could be argued that the two divergent trends combine for a decidedly uncomfortable situation. If the middle class looks upwards, the rich are pulling further away. A top-tier life feels further out of reach than ever. But look down, and the floor is coming up fast. This simultaneous rise of resentment and precarity is a dangerous cocktail, and could certainly have fed into recent political undercurrents.

Professions once considered aspirational are at the sharp end. In Britain, doctors, nurses and police officers have all been slipping down the income rankings in recent years. In the US, the highest-paying jobs are increasingly shared among a handful of ultra-high-status occupations. Tech workers now account for one in six of the top 5 per cent of salaries, up from one in 20 in 1990... In the 1980s, 40 per cent of the highest-paying jobs in America didn’t require a degree. The upper reaches of the income scale included plenty of engineers and doctors, but also senior schoolteachers and the most skilled factory and construction workers. People from all sorts of backgrounds with all sorts of skillsets could dream of making it. Today, the upper part of the scale is dominated by highly skilled tech and healthcare workers. Almost half of the top jobs require an advanced degree. And a huge section of the population knows at a pretty early age they’re not on that path.

7. Some numbers of household savings in India.

Net household financial savings in India rose from 7.7 per cent of gross domestic product (GDP) in 2019-20 to 11.7 per cent in 2020-21, largely because of precautionary and forced savings during the pandemic, but moderated thereafter to a multi-decade low of 5.3 per cent in 2022-23... The share of low-cost current and savings accounts in total deposits has declined the past few years from a peak of 44-45 per cent in recent years to 38-39 per cent. In contrast, mutual funds, especially equity and hybrid schemes, have seen a surge in inflows and have delivered higher returns over the past few years. The share of mutual funds constituted around 6.1 per cent of household savings in 2022-23, with the number of mutual-fund folios jumping from 146 million at the end of 2022-23 to 178 million at the end of 2023-24.

8. Finally, after years of wrangling including several lawsuits, congestion pricing in busy hours goes live in mid-town Manhattan from the morning of 6th January.

Most passenger cars will now have to pay $9 to enter Manhattan south of 60th Street at peak hours, rather than the original $15. Small trucks will have to pay $14.40; large trucks, $21.60. Discounted rates will be offered overnight when there is less traffic. M.T.A. leaders expect the new tolls to help generate $15 billion through bond financing that will pay for a long list of transit repairs and improvements, including modernizing subway signals and stations and expanding the electric bus fleet... State officials said the original plan was expected to reduce the number of vehicles in the congestion zone by roughly 17 percent. They have not specified how the scaled-back program will compare except to say they expect it to cut traffic by at least 10 percent... The tolling plan also does not directly charge drivers and owners of for-hire vehicles, which have exploded on city streets since Uber’s arrival in 2011. Instead, a small per-trip fee — $1.50 for Ubers and Lyfts; 75 cents for taxis — will be added to each fare and paid by passengers... Within the congestion zone, the average travel speed has dropped to under 7 miles an hour for the first time since records were kept in the 1970s, he said. The slowest traffic crawls along at just 4.7 miles per hour in Midtown.

See also this.  

9. As Saudi Arabia wins the bid for the 2034 Football World Cup, FT writes on the country's ongoing boom in infrastructure investments.

Saudi Arabia has launched real estate and infrastructure projects worth $1.3tn since Vision 2030 was unveiled in 2016, according to estimates by consultancy Knight Frank. These projects, such as the Neom linear smart city, include adding more than 362,000 hotel rooms and 7.4mn square metres in retail space. Football has become one of Prince Mohammed’s prime targets for sporting investment. The country’s sovereign wealth fund acquired English Premier League side Newcastle United, while superstars such as Ronaldo and Neymar have been lured to play in the Saudi Pro League. Saudi Arabia’s bid said the 48-team World Cup would be played in 15 stadiums across five cities. Eight stadiums would be in or close to Riyadh, which is already undergoing a construction boom that includes an entertainment zone to the west and a major expansion to the capital’s airport.

10. Adam Tooze points to a UNTD report that highlights the large differences in the cost of debt between Western countries and low-income ones.

11. Doing business, Huawei style.

Huawei’s first business was importing telephone switches before building its own, cheaper versions, copying foreign designs in the process. It later benefited from a government policy to rip out foreign technology in China’s communications network. Huawei developed a reputation for generosity towards government officials and telecoms executives, paying for international travel and hosting lavish banquets at its campus. Dou portrays Ren as an expert networker, including sending birthday cakes to retired telecoms experts who had helped Huawei.

12. Divisions in the Trump coalition that will only grow.

The Maga crowd and the globalists disagree not only on immigration, but on defence, employment and free speech. This is a coalition whose most significant overlap was a desire to take down the previous government. Now that they have, I think it’s unlikely they’ll come together on anything else.

13. Stunning figures about the US economy from Ruchir Sharma

Following the pandemic, government spending rose sharply as a share of GDP. More than 20 per cent of new US jobs are now created by government, up from 1 per cent in the 2010s. Public transfers including Social Security account for more than a quarter of residents’ income in more than 50 per cent of US counties — up from just 10 per cent in 2000.

14. Corporate India's acute deficit of global brands

Many Indian brands have either disappeared or ceded space to foreign competition. Where Onida and Videocon once dominated the domestic market for TVs, washing machines, and household appliances, Japanese, Korean, and, increasingly, Chinese brands now rule the showrooms. In cars, the Premier Padmini and Ambassador vanished when Japan’s Suzuki set up its joint venture to launch the Maruti, an Indian brand only in name. Here, too, it is the Japanese, Koreans, Germans, and Chinese that offer consumer choices, with Tata and Mahindra & Mahindra being the only indigenous exceptions. In fast-moving consumer goods, brands such as Anchor, Nirma, Uncle Chipps, and Binny’s, which once gave multinational players a run for their money, have all vanished or receded to the margins of the market... The abdication of Indian brands to global competition — with many of them converting themselves into contract manufacturers — reflects the lack of long-term thinking and strategic imagination, which are critical to brand-building.

15. This is what US Treasury Secretary-nominee Scott Bessent thinks.

Bessent has also suggested that countries with military protection from America should be forced to buy more dollar debt, as a quid pro quo. “Is there some kind of statecraft to do where you go to [these countries] and say we have these 40- or 50-year military bonds [to buy]?” he said, citing Japan, Nato members and Saudi Arabia.

16. Finally, Tej Parikh presents some facts about European stocks that go against conventional wisdom. Excluding Nvidia, European stocks outperform S&P 500 since the latest bull market started in October 2022!

European stocks are undervalued compared to their American peers.

Tech accounts for around just 8 per cent of the Stoxx Europe 600. AI euphoria has mostly passed the continent by... The Granolas... covers a diverse group of international companies spanning the pharmaceutical, consumer and health sectors. Together, they account for about one-fifth of the Stoxx 600. Their performance against the Magnificent Seven has only recently diverged. The S&P 500 — which has around 70 per cent revenue exposure to the US — got a jolt following the election of Donald Trump... Small listed European businesses also tend to outperform their American counterparts. About 40 per cent of US small caps have negative earnings, compared with just over 10 per cent in Europe. The winner-takes-all dynamic may be stronger in the US, where tech behemoths suck capital and talent away from smaller companies... European corporates also rely more on relationship-based, illiquid funding, unlike in the US, where listed equity dominates. That may encourage longer-term corporate governance in Europe... Regarding the Trump tariff threat, it’s not all disaster for European companies either. Stoxx 600 groups derive only 40 per cent of their revenues from the continent... A stronger dollar would also boost the earnings of European companies with sizeable US sales.

India infrastructure update - January 2025

Some graphics from a very informative report by real estate consultancy Knight Frank on Indian infrastructure.

India spends less on infrastructure than other major economies.

A disturbing feature of India’s infrastructure finance landscape over the last decade and a half has been the decline of private finance and the dominance of public finance.

In absolute terms, over the last three four-year periods, private investments have declined by three-fourths in the four years ending 2003, even as central government public investments have risen five-fold. Total public investments rose from $185 bn to $505 bn. Private investments contributed just 7.2% of all infrastructure investments in the 2019-23 period.

In the 1990-2023 period, private investments amounted to $305 bn, most of which going to energy and roads. Along with ports and airports, they have formed 95% of all investments in this period. 

Interestingly, in this period, urban transport projects attracted $7.6 bn, water supply and sanitation $1.6 bn, and waste management $1.3 bn of private investments. Railways attracted a mere $0.02 bn in private private investments.

Banks and NBFCs, the major source of infrastructure debt finance for private investors, has plateaued since 2018-19 (or has grown little since 2014-15). The total outstanding loans to infrastructure sector has risen by just Rs 2.8 trillion since 2014-15, an annual growth rate of 3.38%. 

Again, mirroring private investments in general, power, telecommunications, and roads made up 85% of all loans.

FDI has been in the range of 0.15-0.25% of GDP, compared to the central government spending of 2.5-3% of GDP and total public spending on infrastructure of 5-6% of GDP. 

Railway projects formed nearly half by value of all cost overruns. 

Underlining the point about strategic misrepresentation in traffic estimates of metro rail projects, actual ridership has fallen short of estimates in all cities except Mumbai and Delhi. 

Finally, the point about FDI is underlined by this graphic from a McKinsey report. The total Asia-focused private capital flowing into infrastructure and natural resources is less than $10 bn and declining.

Against this backdrop, here are some observations.

1. At the behest of experts and multilateral institutions, the efforts of central and state governments in the mobilisation of infrastructure finance have hitherto been focused on bond market financing. As I have blogged multiple times, this is a misguided endeavour. Capital market financing of projects in even de-risked sectors will always remain a small proportion. 

Instead, policy attention should focus on expanding the coverage of bank finance for revenue-generating projects in those sectors with limited private participation and debt exposure. There’s a need for concerted efforts in this direction, like the decades-long policy campaign to deepen municipal bond markets in India. 

2. The debt exposure of infrastructure sectors like railways, urban mass transit, water and sewerage, solid waste management, electricity distribution etc., is negligible. As I blogged here, commercial debt formed just 2% of all urban infrastructure finance. This points to a valuable opportunity (albeit very less discussed) to leverage debt finance to fund projects in these sectors. 

But there are deep ideological and daunting political economy challenges that must be overcome to capitalise on this opportunity and leverage debt finance. 

All these sectors are also those where private capital is unlikely, and most of these investments must come from public finance. Since these sectors generate some revenue, even if not enough to cover the capex, there’s a strong case for blending public finance with debt capital. This will also expand the shelf of projects that can be funded with scarce public finance. 

There are at least two arguments against blending publicly financed projects with debt. One, debt finance comes with a cost of capital, whereas public finance is a grant. Even if debt is to be mobilised, it’s cheaper for governments to borrow directly and fund these projects from the budget instead of allowing the project entity to raise project finance. However, there are limits to direct public borrowings. 

Two, borrowings by public entities have the potential to generate perverse incentives arising from the political economy. The disturbing rise in government-guaranteed debts assumed by state government public sector enterprises points to this concern. Finance Departments are therefore averse to allowing public entities to raise debt. 

Addressing these concerns requires attention on both supply and demand sides. 

On the supply side, it’s required to create confidence among banks and NBFCs to lend to such projects. Given the political economy associated with tariffs and user fees, lenders prefer to stay away from such projects. Asset-liability mismatches from bank lending should be minimised through the likes of loan syndication and takeout financing. 

On the demand side, one strategy is a proposal outlined here on allowing government entities to create Special Purpose Vehicles (SPVs) that could raise project debt against the project revenue streams. This structure would align incentives and address the political economy distortions. In the initial years, lenders could be de-risked through a guarantee fund mechanism that would provide them partial guarantees. 

To support this, the Government of India should restrict grant funding in these sectors only to the extent required to meet the viability gap that would allow the project entity to raise debt. This would also allow the government to prioritise scarce public finance to those areas where private capital or debt cannot be mobilised. The GoI’s Viability Gap Funding (VGF) scheme should prioritise public-funded projects in these sectors that utilise debt. 

3. Public policy engagement on private participation in infrastructure should no longer be about sectors like power generation and transmission, highway roads, airports and ports. These are largely de-risked sectors in India, except for certain remote areas where their project’s commercial viability is questionable. 

Instead, the policy focus should be to attract private capital towards electricity distribution, railways, fishing harbours, and the large segment of urban infrastructure services - water supply, sewerage, street lighting, solid waste management, mass transit (buses and metro rail), parking facilities etc. 

For a start, the current levels of private participation in these areas is minimal, and given the difficult political economy of tariffs and user fees, there’s a need to de-risk them for private investors. More importantly, projects in these sectors require varying levels of VGF to cover the upfront investments. Therefore, public funding for infrastructure should prioritise these areas. A part of the GoI’s 50-year interest-free capex loans to state governments could be allotted for such financing (to meet the state government's share of the VGF funding). 

Weekend reading links

1. Some facts about India's trade trajectory since the millennium. Exports and imports have surged 12 and 15 times respectively.

But closer scrutiny of the data shows that over the last 13 years, from 2010 to 2023, India's export share saw only modest growth, rising from 1.5 per cent to 1.8 per cent... despite efforts to diversify, India's share in world trade has more or less stagnated over the past decade... Exports make up a significant portion of GDP in smaller economies — Vietnam (93 per cent) and Singapore (174 per cent) — but also leave them vulnerable to global economic instability, he explained. In contrast, larger economies have a much smaller share of exports in their GDP: the US (11.6 per cent), China (19.7 per cent), Japan (20.5 per cent), and India (21.9 per cent).

2. Fascinating article about a Russian satellite Cosmos 2553 located at an orbit of 1240 miles (a high radiation area where there are no other satellites), circling earth once every two hours, and with the potential to threaten military and commercial satellites with a nuclear blast in space. The US scientists believe that it's part of a Russian attempt to develop a nuclear weapon to obliterate hundreds, if not thousands, of critical satellites. But this Russian effort is a sharp advance in anti-satellite technologies.

Once considered a largely peaceful domain, space is now viewed by many American lawmakers and military commanders as a place where the next major global conflict might unfold. If Moscow is working on a space nuke, it would be merely one of dozens of space weapons under development or already in use by Russia, China and the United States. All three nations have tested high-flying missiles capable of targeting space systems from the surface and have lasers, signal jammers and other devices that can disrupt space operations. Russia has deployed nesting doll satellites (in which one satellite births a smaller satellite that is maneuverable and armed with a projectile) and China and the United States have demonstrated grappling satellites, which can sidle up to another satellite and tug it out of its orbit with robotic arms. It may sound as if these technologies were torn from the pages of a science fiction novel, but none of them come close to doing what a nuclear weapon could in space: wipe out clusters of satellites at once... 

The detonation would disable and destroy everything in its immediate vicinity, turning satellites into unguided projectiles that could crash into one another. Objects in low orbits travel at around 17,000 miles per hour. Any debris — even as small and light as a paint chip — would pose real danger to other objects or people in space... Swirling away from the blast point, the charged particles would form a shell of radiation that would linger for weeks, if not years — long enough to gradually fry the onboard electronics of surviving satellites orbiting close to Earth. U.S. intelligence analysts have determined low-Earth orbit would be unusable for an unknown period, depending on the size of the blast.

3. Coffee prices soar on the back of bad weather.

According to the U.S. Department of Agriculture, around 57 percent of the world’s coffee production last year came from arabica beans, and Brazil is the largest exporter. But a severe drought there this summer devastated the harvest, which typically runs from May to September, and it could threaten next year’s crop as well. In Vietnam, a severe drought followed by heavy rains harmed the world’s largest reserves of robusta, which is the second-most-popular variety globally and is commonly used in instant coffee blends... The wholesale price of beans has jumped more than 30 percent just since the start of November. Futures prices for arabica beans — or what buyers pay for beans to be delivered from producer countries to ports in the United States and Europe — rose to more than $3.30 per pound in mid-December, breaking a 47-year-old record.

4. Good FT explainer on China's urban migration control system, hukou. 

Because a hukou governs access to local government services, having one from China’s wealthiest municipalities, particularly Beijing and Shanghai, entitles the holder to the best education and healthcare, easier access to stable government jobs and other privileges. Abandoning hukou would encourage more rural workers to move to more productive jobs in cities... at least a quarter of city dwellers lack an urban hukou. Many migrant workers crowd into neighbourhoods on the margins of the big cities, such as Yuxinzhuang in northern Beijing. Scrapping the system would give migrant workers better access to public health and education, leaving them with more money free for consumption... Residents of first-tier municipalities would resist losing their privileges, while city governments would balk at the cost of investing in additional infrastructure and services. Many of the 298mn migrant workers who already live in urban areas lack access to the best healthcare, education and public services...

While some smaller cities have abandoned hukou altogether as part of gradual reforms, obtaining one in China’s largest municipalities, particularly Beijing and Shanghai, has become even more difficult, according to Yao Yang, a prominent Chinese professor and author. Outsiders seeking a Beijing hukou need to satisfy a points system that takes into consideration their university degree, professional abilities and ability to pay tax — tougher criteria than what many countries impose on foreign immigrants. A Beijing hukou can also be obtained through marriage rules or by birth. Shanghai’s system is almost equally strict. In China’s first-tier cities, a hukou gives access to 20 different rights, said Yao in a public speech earlier this year. This starts with direct access to the nation’s best schools, which are often in central Beijing and Shanghai and offer students superior preparation for university...

 

One of the most important privileges a Beijing hukou grants the holder is the right to complete high school and sit the gaokao, the university entrance exam, in the capital. The country’s most elite universities, which are concentrated in Beijing and Shanghai, generally offer higher entrance quotas for students sitting the exam in those cities. The populous central province of Henan, for instance, has about 20 times more students sitting the gaokao than Shanghai and Beijing, but admission rates for Henan’s students are only about a fifth of the rates achieved by applicants from the biggest cities. Privileges such as these make a Beijing hukou so valuable that some employers offer it as part of their pay packages... Without a Beijing hukou, she could only buy a less desirable flat with higher management fees. Beijing’s traffic congestion rules, which favour hukou holders, forced her to pay more to own a car.

5. From Adam Tooze, a graphic which points to how low electricity prices are in China, which is an important factor in its economic competitiveness. 

The low prices also reflect the high level of hidden subsidies.

Another graphic points to the very tight correlation between electricity consumption and national income - there are no low-energy rich country.

In other words, unless there's some miracle, forcing down electricity consumption is only going to keep the country poor. This is a big challenge to climate mitigation. 

6. Income distributions of India and China

7. Important points about constraints to the rapid expansion of industrial housing.

Private-sector initiatives in this area are often hampered by regulatory hurdles. Inflexible zoning regulations, illiberal building laws and approval processes, and high operating costs restrict land usage near industrial clusters, resulting in suboptimal usage of land. For instance, both the floor area ratio (FAR) and ground coverage ratio (GCR) are quite low in the country compared to those in developed countries, limiting both vertical and horizontal expansion, respectively. High setbacks, which mandate minimum distances from property lines, roads, and adjacent buildings, lead to wastage of space. Industrial housing faces additional challenges, including mandatory parking requirements, which are unnecessary because most industrial workers do not own vehicles.

Recommendations of the NITI Report on this.

The report suggests that government support in the sector should come in the form of VGF to bring down the cost of construction, and give tax relief and interest subvention to subsidise market interest rates to reduce the cost of borrowing for builders. There is also a need to relax FAR and GCR norms such that building height and size can be decided on the basis of cost-efficiency considerations.

8. Debashis Basu has some figures on the Indian economy.

India’s GDP (gross domestic product) growth, after reaching 8.2 per cent in FY24, fell to 6.6 per cent in Q1 (the first quarter) of FY25 and then further to 5.4 per cent in Q2... Government capex as a percentage of expenditure reached 28 per cent in FY24, up from just 14 per cent in FY14. However, while government capex increased by 33.7 per cent last year, it has contracted by 6.6 per cent in April-October of FY25... Growth in central-government gross tax revenues has fallen to 10.8 per cent in April-October FY25, compared to 18 per cent in FY23 and 14 per cent in FY24. Similarly, GST (goods and services tax) collection has slowed significantly. After 26.2 per cent growth in FY23, as the economy rebounded from the pandemic, GST collection grew just 11.9 per cent in FY25 and 9.3 per cent in April-November FY25. Other worrying signs are emerging as well. In April-November FY25, growth in power consumption slowed to 3.9 per cent, down from 9.7 per cent in FY24. Cement production increased only 1.8 per cent during the same period, while fuel consumption grew just 3.3 per cent.

9. Some snippets on Reliance's Jamnagar refinery, which celebrates its 25th anniversary.

The refinery overnight turned India from a fuel deficit nation to a self-sufficient one and later into a surplus, exporting gasoline and gasoil to Europe and the US. Today, Jamnagar has become the world's refining hub... experts had said that it would be impossible for an Indian company to set up the world's largest grassroots refinery in three years... Reliance achieved that in a world-record time of just 33 months, notwithstanding lack of infrastructure and a severe cyclone that had hit Jamnagar then... More importantly, the 27 million tonnes a year (560,000 barrels per day) capacity unit was built at nearly 40 per cent lesser cost (per tonne) in comparison to contemporary refineries in Asia. The unit was later expanded to 33 million tonnes... The first private sector refinery of India single-handedly added 25 per cent to India's total refining capacity and made India self-sufficient in transport fuels...

A decade later, Reliance, through a subsidiary, build another refinery adjacent to the old one. The new unit capable of processing a whopping 580,000 barrels per day (29 million tonnes) turned Jamnagar into world's largest single site refining complex. The refinery's most interesting feature is that it is one of the world's most complex. This enables it to turn cheaper heavy crudes into top quality products that meet increasingly tough specifications in western fuel markets. And in doing so it is able to compete with almost every refinery in the world. The new refinery caters to only the export market while the older one meets domestic market demand...

Reliance's focused efforts created a green zone in the arid land, resulting in the lowering of temperature and improved rainfall in the region. The Jamnagar refining complex now boasts of Asia's largest mango orchard, with over 1.5 lakh mango trees. The huge mangrove belt there has become a haven for migratory birds, and the surrounding dense forest houses theVantara --the one-of-its-kind rehabilitation home for rescued wild species.

10. A reminder of the extraordinary achievement that India's economic liberalisation is. It's extraordinary also as to how such radical measures involving multi-departmental co-ordination were done in such quick time. 

11. Corporate profits in four five year periods from 2004-24.

Excluding banks and oil companies, corporate sales and profits during the 2004-09 stands out even more.

12. Primer on India's cement industry. 

13. The paradox of India’s mango cultivation - the largest cultivator in the world that also struggles to export.

The ‘king of fruits’ has been cultivated in India for 4,000 years, and the country is known to grow about 1,000 varieties. Indeed, India is the largest producer of mangoes in the world, accounting for half of the global output. And yet, less than 0.5% of its annual production of 20-22 million tonnes is exported. That’s because the country’s best mango cultivars are ill-suited for commercial export and only a handful of popular ones drive the market. The birthplace of the fruit is yet to hit upon a variety that can dominate global markets. Consequently, in 2023, India’s share in global mango shipments stood at a mere 6.3%, lower than Mexico (21%), Thailand (15%), Brazil (12%) and Peru (11.6%)… In 2023-24, India exported 93,000 tonnes of fresh mango and pulp (valued at ₹1,120 crore)—a decline of nearly 40% from 2021-22.

So, why isn’t the world’s largest grower exporting more mangoes? Experts cite two reasons; first, as mentioned earlier, the best varieties grown in India do not travel well over long distances. On the other hand, mangoes grown in Mexico and Brazil, such as the Tommy Atkins and Kent varieties, are better travellers, thanks to having thick skins. But what they enjoy by way of a longer shelf life is offset by the lack of sweetness and complexity of flavour. In addition, Latin American exporters Mexico, Brazil and Peru, among others in the region, have been able to capture the premium US and Canada markets not so much because of the quality of their mangoes as their geographical proximity, which makes for lower freight costs.

Since most Indian mangoes have short shelf lives—between two days to two weeks after harvest—the air route is the preferred mode of transport. This pushes freight costs up. For instance, the average air freight for Siddiqui is around ₹70 per kg this summer. This, too, is from smaller international airports such as Lucknow and Jaipur to destinations in the Middle East. The payout on freight is higher than the price at the farm gate—it costs more to transport a mango than it costs to grow one. For destinations in the US, the transport costs are in multiples of wholesale mango prices. This summer, the air freight charge from Bengaluru to a US destination is about ₹450 per kg—over four times what it costs an exporter to procure mangoes ( ₹110 per kg for Banaganapalle)…

The second reason India’s mango exports have remained low is the strict entry standards set by importing countries, imposing a high compliance burden on anyone who wants to ship agricultural produce. The United States, for instance, requires imported mangoes to be irradiated (exposing the fruit to gamma rays) to ensure they are pest and disease-free. The European Union asks for hot water treatment (immersing fruits in water heated to 48 degrees Celsius, for an hour). Japan and New Zealand require vapour heat treatment (heating the fruit with air saturated with water vapour).

14. More on the thin personnel capabilities of Indian state

At the end of January 2024, 331 of the total sanctioned 1,114 vacancies for judges in various high courts were vacant. Similarly, there were over 5,000 vacancies across various subordinate courts in the country... As of November 2024, the National Company Law Tribunal had 43 members in service against the authorised strength of 63. The average time taken for the resolution process of an insolvency petition today has gone up to 716 days, far exceeding the stipulated maximum of 330 days, even for cases involving litigation. Another tribunal — the Debt Recovery Tribunal — is paralysed because many do not have a presiding officer.

15. Noushad Forbes summarises corporate India's R&D problem

Indian industry invests 0.3 per cent of gross domestic product (GDP) in in-house research and development (R&D), compared to a world average of 1.5 per cent. We spend $7 billion annually on industrial R&D, compared to $625 billion in the US, $335 billion in China, $130 billion in Japan, and $90 billion in Germany. We are the world’s fifth-largest economy and manufacturer, but rank 21st in industrial R&D. Our 10 most successful non-financial firms have a very healthy profit by world standards but invest little in R&D: A mere 2 per cent of profit. By contrast, firms in the US, China, Japan, and Germany invest between 29 and 55 per cent of their profits in R&D. To put this in perspective, 25 individual firms — from Alphabet ($40 billion) to BMW ($7.6 billion) —invest more in R&D than all Indian firms combined.

16. Important tourism fact

Dubai, a single city, now attracts twice as many tourists as all of India.

And this

Foreign tourist arrivals peaked in 2019 at 10.9m. That year Dubai (World Heritage Sites: zero) attracted 16.7m visitors. In the first half of 2024 Dubai’s numbers grew by 11% compared with 2019. India’s fell by 10%.

17. Finally, Indian economy graphics. On corporate investment decline.

And the anemic demand for consumer goods.