#241 Escaping the Many Traps

On the 22nd January Event, Javier Milei's Davos Talk, #ChipFails, and Sugar Pricing in India

India Policy Watch #1: Hey Ram

Insights on current policy issues in India

— RSJ

मंगल भवन अमंगल हारी, 

द्रवहु सु दसरथ अजिर बिहारी 

(He who brings prosperity and takes away the woes of people, let Lord Ram shower his compassion on me) 

That Tulsidas chaupai was what I woke up to on many Sunday mornings in the late 80s. At my home, Ram wasn’t the kind of divinity who we prayed to often. Krishna, Shiva and Ganesha were ahead in that pantheon. Yet he was ever-present. His legend was passed on to us through his stories and his conduct. So, when Doordarshan commissioned Ramayan as an hour-long TV series that would run on Sunday mornings for a year back in 1987, its success was foretold. No one stepped out on Sunday mornings till 10 AM, when the weekly episode would draw to a close. We didn’t care about the production values or the often hammy acting that characterised the show. Yet when the story of Ram, so familiar to all of us through countless stories in books and at Ramlilas, came alive on screen, we sat riveted. Religion is often understood to be in the domain of the deeply personal. It might be so. But its most effective experience is through community. As Durkheim wrote, when the pagans prayed to a totem, they were worshipping their own society that had found the divine in it. Fighting against the social appeal of the religion or trying to box it within the personal is a losing battle. But I am getting ahead of myself here.

Back to the late 80s. It was also during this time, one winter morning as I was on my way to school, that I noticed for the first time “Garv se kaho hum Hindu hai” (say with pride that we are Hindus) and “Hindu jaagega toh Desh jaagega” (this country will awaken when Hindus wake up) scrawled across walls in brick-red colour. The slogans intrigued me. Nothing in my short life had given me the impression that the community I belonged to was lying dormant or had issues with self-respect. But clearly, someone felt strongly enough to go about painting it across a small town in what was beyond the literal back of the beyond. It felt incongruous. But this feeling didn’t last long. In about a year from then, the Ram Janmanhoomi movement began dominating the Indian political landscape. The door-to-door collection of bricks for the shilanyas, the daily reports of the L.K. Advani’s Rath Yatra and the trail it left behind and lumpenisation of Jai Shree Ram and Jai Bajrang Bali all followed soon after. I don’t remember if there was any awakening of the Hindu soul that happened during those years on the back of the movement. But the political movement revived the BJP in the north, and a little over three decades later, it remains a gift that keeps giving. 

To those who grew up in the 80s, the events on the 22nd of January, when the Ram Lalla pran pratishta ceremony (an inauguration of sorts) is concluded, and the Ram temple in Ayodhya is consecrated, will bookend a political issue that dominated all our adult lives. So I watch with fascination the wall-to-wall TV coverage, the back stories about the movement that have all been sanitised now, the old faces of Uma Bharti and Govindacharya talking about the genesis of the movement, the door-to-door campaign inviting people for the inauguration, the declaration of holiday keeping the sentiments of people in mind and the long list of VVIPs who will be there on Ayodhya on Monday. The timing, about three months before the Lok Sabha elections, is perfect for the BJP. The opposition, which knew this was coming for a long time, has no plan on how to counter this political windfall. Anyway, even the most consummate political artist would have found it difficult to blunt this edge. The elections are a foregone conclusion now. It is only a matter of the margin of victory. A third consecutive Lok Sabha win. It will be hard to find a political strategy across democracies that has been so successful over such a long time. Granted, there were other issues too that have helped in electoral wins, but the temple movement has been the beating heart of political Hindutva. Which raises the question, what next? Will the temple, once built, continue to be a political plank for the future? Or, will it mean something more, politically or otherwise, to the people who hold it dear?

I think the direct political advantage of the temple ends with this Lok Sabha election. It might still resonate in the UP elections next year, but that should be that. This is the reason there is a clear shift in narrative from celebrating the construction of the temple itself to imbuing it with meaning and symbolism for the new India. That the temple represents a revival of India, its civilisation and a wresting back of its history and its image from the stereotypical portrayal of the western elites. This awakening will further build the self-belief of our people in what we believe will be India’s decade. Or, so we’d like to believe. Will it work out this way? Is there a reason to believe a grand temple of a mythological figure will galvanise a society? What kind of a role can religion play in building a modern democracy in India? Does it even have a role?

I tend to be in the Durkheim camp that religion is a sui generis social institution. It exists out of nowhere, and it is the most fundamental of the social institutions. To deny its social features in how it is observed or in its agency to drive social change is futile. History has shown the more you drive it into the background, the stronger, and often in a more virulent form, it comes back. Religion gives us much more than an understanding of the world, its conception, of the God (or gods) who created it and the rituals that hold it together. These are important in separating the sacred from the profane and defining the boundaries of one religion from another. But beyond these, religion gives us a sense of belonging. In its most popular form, religion is a communal experience, and it is this sense of belonging that can give it a purpose beyond individual redemption. Like building a society or a nation based on eternal values that are held dear in that religion, like compassion, natural justice or toil. A repudiation of religion and debunking of rituals and customs in favour of scientific explanation works to a certain extent. Beyond it, a vacuum of faith is often filled up with another dogma of some kind. Often, this looks good on paper but is untested. This is what communism was, or the certitudes of woke left today is. So, religion is important in not just what it stands for but how bad its absence could look like. The question is, what form of religion should we stand for? That’s not an easy question to answer. In the many articles and speeches that suggest the Ram temple will stand for a revival of our national consciousness, I haven’t found the answer to this. Harking to a mythical past or believing that a structure by itself will free us from prejudices and fears that have held us down is not going to move us forward. 

Maybe the answer is in reading and re-reading Ramayan with an open mind and heart. It is a gift that keeps giving. 

Like Tulisdas wrote:

पुनि पुनि कितनेहू सुने सुनाये

हिय की प्यास भुजत न भुजाये

(You can listen to Ram’s story for the umpteenth time, your heart still remains unquenched) 

---

Global Policy Watch: Hard Talk

Global policy issues relevant for India

— RSJ

Javier Milei, the Argentine President, gave an address at the World Economic Forum, Davos, this week that has gone viral for good reasons. Here’s an excerpt that I endorse:

“Unfortunately, in recent decades, the main leaders of the Western world have abandoned the model of freedom for different versions of what we call collectivism. Some have been motivated by well-meaning individuals who are willing to help others, and others have been motivated by the wish to belong to a privileged caste.

We’re here to tell you that collectivist experiments are never the solution to the problems that afflict the citizens of the world. Rather, they are the root cause. Do believe me: no one is in better place than us, Argentines, to testify to these two points.

Therefore since there is no doubt that free enterprise capitalism is superior in productive terms, the left-wing doxa has attacked capitalism, alleging matters of morality, saying – that’s what the detractors claim – that it’s unjust. They say that capitalism is evil because it’s individualistic and that collectivism is good because it’s altruistic. Of course, with the money of others. 

So they therefore advocate for social justice. But this concept, which in the developed world became fashionable in recent times, in my country has been a constant in political discourse for over 80 years. The problem is that social justice is not just, and it doesn’t contribute to general well-being. 

Quite on the contrary, it’s an intrinsically unfair idea because it’s violent. It’s unjust because the state is financed through tax and taxes are collected coercively. Or can any one of us say that we voluntarily pay taxes? This means that the state is financed through coercion and that the higher the tax burden, the higher the coercion and the lower the freedom.

Those who promote social justice start with the idea that the whole economy is a pie that can be shared differently. But that pie is not a given. It’s wealth that is generated in what Israel Kirzner, for instance, calls a market discovery process. 

If the goods or services offered by a business are not wanted, the business will fail unless it adapts to what the market is demanding. They will do well and produce more if they make a good quality product at an attractive price. So the market is a discovery process in which the capitalists will find the right path as they move forward.

But if the state punishes capitalists when they’re successful and gets in the way of the discovery process, they will destroy their incentives, and the consequence is that they will produce less. 

The pie will be smaller, and this will harm society as a whole. Collectivism, by inhibiting these discovery processes and hindering the appropriation of discoveries, ends up binding the hands of entrepreneurs and prevents them from offering better goods and services at a better price.”

---

India Policy Watch #2: Understanding Chip Manufacturing Failures of the Past

Insights on current policy issues in India

— Pranay Kotasthane

I’ve often been asked why India’s many past attempts to manufacture semiconductors failed. In answering such questions related to failures, we often fail into the trap of highlighting one catastrophic event or overarching policy failure.

In this particular case, people often zoom in on the industrial fire that debilitated India’s only commercial semiconductor manufacturing facility in 1989. Another reason often cited is the unavailability of reliable electricity and water.

However, such uni-causal explanations are simplistic and often plain wrong. Such explanations evoke in the reader a rueful reaction that if only the one failure point could have been avoided, we would be in a much better place.

The reality is more sobering. That’s because the reasons for policy failures are many. The complex system lens shows us that policy successes are often the result of many parts of the puzzle falling into place rather than a result of a masterstroke move. In the semiconductor manufacturing domain, the reasons for failure were also many. In a sense, the essential ingredients needed to create a good policy recipe were missing.

Here’s an account of past failures, excerpted from our book When the Chips Are Down.

Version 1.0: By the Government, for the Government

The story in semiconductors follows an all-too-familiar trend: there are initial breakthroughs within the government ecosystem; these successes don’t translate into exports or even domestic or commercial products; and eventually, companies resign to serving a vital yet small niche market within the government. The knowledge spillovers for the economy remain marginal.

India’s semiconductor manufacturing journey started around the same time as Taiwan. Semiconductor Complex Ltd (SCL), a Government of India enterprise in Chandigarh, began producing chips using the 5,000-nm CMOS (complementary metal-oxide semiconductor) manufacturing node in 1984. Back then, SCL had monetary backing from the government and a good pool of graduates and PhDs from top Indian engineering colleges. SCL also had access to talented scientists from defence and aerospace electronics national champions such as Bharat Electronics.

Armed with sufficient funding, a good talent pool and government support, the only remaining hurdle for SCL was technology acquisition. SCL established a joint venture with the American company Microsystems Inc. in the early 1980s. It also acquired technology from two other companies (the US firm Rockwell Automation and the Japanese firm Hitachi) as part of a deal to manufacture components from them. SCL advanced to 800-nm technology over the years, which was just a couple of years behind the global cutting edge.

However, SCL couldn’t graduate to producing large-volume commercial chips, which were the trendsetters in the advancement of semiconductor manufacturing. As a government-owned enterprise with limited resources, SCL couldn’t keep pace with Rock’s law that the capital cost of a semiconductor chip fabrication plant doubles every four years. To make matters worse, it witnessed a massive fire in its complex, which stalled the company’s progress. SCL was never able to recover from the incident.

In 2000, the Ministry of Information Technology tried selling some of SCL’s equity but couldn’t find any buyers. They cited the lack of Indian OEMs that could use the chips SCL created. The Department of Space eventually revived it in 2005, after which SCL’s focus shifted to making chips for satellites and launchers.

Even though SCL continues to make chips today, it is a shadow of its past self. While companies are making 5-nm and 3-nm technology chips, SCL is still working on upgrading the facility to produce a small number of chips using the 180-nm technology node.

Another public sector giant, Bharat Electronics Limited (BEL), was also investing in semiconductors. Its proposal to manufacture germanium chips in 1959 was approved after a long delay of two and half years. Philips helped with the knowledge transfer, and by 1962, BEL had manufactured its first transistor. Manufacturing was slowly ramped up, moving up to silicon transistors. In its heyday, BEL was manufacturing up to 20 million transistors annually. By the mid-1980s, it partnered with Metkem Silicon Ltd to produce polysilicon wafers for solar cells and electronics.

But by the late 1980s, BEL, too, hit the constraints posed by Rock’s law. The tight foreign exchange situation meant it couldn’t import manufacturing equipment easily. Though it was able to attract the interest of National Semiconductor for manufacturing analogue ICs, it couldn’t get the government’s approval for the plan. Soon, turf wars between SCL and BEL regarding who should manufacture chips picked up pace. SCL wanted to be the sole manufacturer, and so in 1997, the government asked BEL to limit its activities to packaging and marketing consumer and commercial-grade ICs. Eventually, BEL confined itself to manufacturing a few chips and continued strengthening its work on chip design for strategic applications.

As is apparent, despite showing some promise, these government-led attempts could never match up to the international competition. The Japanese company Matsushita is a relatable reference as it started its semiconductor production line around the same time as BEL with knowledge transfer from Philips. However, Matsushita was export-led and had grown to manufacture 1 billion transistors, while BEL was stuck at 20 million. As a government-run company, BEL’s primary aim was to service the small-volume demands of the military and other government stakeholders.

The parallel with Taiwan is also striking. When ICs started being widely used, BEL wanted to upgrade from manufacturing transistors and discrete components. While looking outside for a technology partner, Philips declined an offer from BEL as it was manufacturing independently and wanted to avoid competition. The American firm RCA came to the rescue. RCA was also involved in setting up plants in Taiwan. While Taiwan went on to become the central node of semiconductor manufacturing, India and BEL faded away from the semiconductor landscape.

A comparative perspective can help understand why India’s ‘by the government and for the government’ approach failed. Like the USSR and early-communist China, India, too, hoped to have government-run national champions in semiconductors. All attempts failed due to three broadly similar reasons.

First, government-run companies had no incentive to compete in a hyper-competitive space that demanded constant capital infusion and technology upgradation. They started well but couldn’t keep themselves in the race for long. Eventually, when political vectors aligned in the right direction, their products became old and costly. Even customers within the government could find better technology at cheaper rates through imports.

Second, these companies were shielded from internal competition. Competition forces companies to seek differentiation. Without it, they could sustain themselves at low-level equilibria. BEL and SCL were both a part of the race initially. But from the government’s perspective, this competition was undesirable—having two companies perform the same task wastes resources. While SCL was chosen to manufacture chips, BEL was confined to assembling them. In Taiwan, on the other hand, the government-led ERSO could spin off multiple companies and foster competition successfully while BEL and SCL were busy fighting turf wars to attain monopoly power. While this approach may have saved the government precious money, it perpetuated a structure that was fundamentally at odds with innovation.

Third, inward-looking trade and business policies proved costly, as in the case of the USSR. The dominant economic narrative was to save foreign exchange and dollars from leaving the country. This meant strict import controls and exorbitant tariffs. Even after paying these duties, equipment remained stuck at ports awaiting government approvals. The cumulative effect was that products from BEL and SCL couldn’t compete internationally. Seeking self-reliance, the government was neither interested nor confident in exporting chips.

Shielded from external and internal competitions, these companies resigned themselves to meeting niche demands of strategic sectors such as defence and space. They were left further behind in the race as chip-making costs rose yearly.

Version 2.0: The Unease of Doing Business in India

India tried getting back in the game multiple times. By the 1980s, it was well understood that government-run efforts at technology upgradation had severe defects.

By the mid-1980s, the government was inching towards economic liberalisation. Immediately, many MNCs began setting up chip design centres in the country. Texas Instruments began operations in Bengaluru in 1985. By the end of the 1990s, as many as fourteen MNCs had design operations in India. By the 2000s, these design houses were also being serviced by outsourcing centres in the country. Enabled by a relatively better Intellectual Property (IP) regime and technological upgradation, the technical skills of Indian chip designers were far ahead of those in China by the late 2000s.

Nevertheless, there were no breakthroughs in chip fabrication as there were no domestic buyers for these chips in sight. The much-hyped Fab City near the Hyderabad airport was announced with much fanfare in 2006. The then Andhra Pradesh government signed an MoU with SemIndia to set up an OSAT plant. One of the backers was the industry giant AMD, which had promised $500 million as investment and technology transfer. Singapore-based Flextronics had also announced a minority stake in the venture.

But then came the global financial crisis. Semiconductor company revenues fell, and financial backing for the Fab City evaporated. Its major investors, SemIndia and Nano Tech Silicon India, dumped their planned investments of $3 billion and $2 billion, respectively, in 2008. Moreover, AMD changed its business model to fabless chip design and lost interest in the deal. The project lagged for years, and the AMD–SemIndia technology-sharing contract eventually expired.

The story repeated itself in 2014. First, proposals from two investor groups to build fabs worth $10 billion were green-signalled. The first group was a consortium comprising Jaypee Group, TowerJazz (Israel) and IBM, while the second group was a joint venture between India’s HSMC, France’s STMicroelectronics and Malaysia’s SilTerra. The first group’s plans dissolved in 2016 after the Jaypee Group pulled out due to spiralling debt, saying the plant was not commercially viable. The second project stagnated in the intent phase after HSMC failed to submit documents for ‘demonstration of commitment’. Apparently, HSMC was unable to find buyers.

In 2015, the US firm Cricket Semiconductors announced a plan to invest $1 billion in setting up a specialised analogue fabrication unit in Madhya Pradesh. However, this project also failed to take off due to procedural delays and a lack of government support in resolving infrastructure problems.

What unites these failures is a poor environment for businesses. A combination of three factors—an uncertain tax and policy environment, poor infrastructure and high trade barriers—made chip fabrication unattractive in India. A chip fabrication unit can take up to four years to set up and requires billions of dollars as upfront capital. Unless investors have complete confidence in the country’s business, tax and trade regimes, they won’t put their necks on the line. India’s policies failed to reassure investors on these counts.

One particular example that left a bad taste in everyone’s mouth was the retrospective taxation against Vodafone. In 2007, Vodafone acquired the ‘Hutch’ Hutchison Whampoa business for $11 billion, which involved their mobile telephony wing. Hutch was then a big telecom player in the Indian market. Later that year, the Indian government accused Vodafone of withholding tax and capital gains valued at almost ₹8,000 crore ($962.4 million at the current rate). Vodafone eventually got it struck down by the Indian Supreme Court and did not have to pay any tax. However, the finance minister then did one better by amending the Finance Act to give the income tax department the power to tax such deals retrospectively. The case eventually went international to the Permanent Court of Arbitration at The Hague, which ruled in favour of Vodafone.

Capital-heavy, long-gestation projects such as chip fabrication units were spooked. How could they trust India to refrain from doing what transpired with Vodafone to their company? Overcoming infrastructural issues—another major bottleneck—such as high-quality water and electricity supply are not a complex problem for one special economic zone (SEZ), but an uncertain investment climate illustrated by the retrospective tax regime disproportionately affects investment decisions.

Another reason that has deterred investors is India’s changing trade policies. India was an early signatory to the Information Technology Agreement (ITA-1) in 1997. Signatories agreed to levy a maximum tariff of 0 per cent on a set of pre-agreed information and communication technology (ICT) goods as part of the agreement. As a developing country, India was granted concessions to lower its tariff over an extended period. Low tariffs were critical since global value chains involve several trips of intermediate products across borders. This move helped India initially, as the first generation of mobile phone makers, such as Nokia, set up their assembly manufacturing in the country. However, by the 2010s, riding on increasing R&D investments and rapidly growing capabilities in large-scale manufacturing, China and Vietnam had embedded themselves firmly in global value chains. Their electronics exports rocketed while Indian companies couldn’t go beyond phone assembly. The lesson that Indian policy-makers took away from this episode was that they need to increase tariffs to incubate the domestic industry and not expose it to the ‘pressures’ of competition. With rising tariffs and resulting costs, manufacturing or assembly of chips in India faced the risk of becoming globally uncompetitive. This trade policy decision weakened the already mild interest of major semiconductor manufacturers.

---

So, will India be third time lucky with the current set of policies? We must ask if the causes of past failures have been substantially dealt with. The book tries to compare recent policies and the current global environment with that of the past.

Looking at India’s chip manufacturing failures from a complex lens can inspire hope and reassurance as well. With better initial conditions and policy ingredients, emergent behaviour can resolve a seemingly wicked problem without an explicit government programme or policy.

---

India Policy Watch #3: Citizen Cane

Insights on current policy issues in India

— Pranay Kotasthane

Parts of India celebrated their respective harvest festivals this week. A common component across festivities is the presence of sugarcane. We, too, bought some cane home—half a stick ordered online for a princely sum of ₹10. Usually, festivals are a good reminder of the supply-demand curve in operation. Remember the surge pricing on marigolds and firecrackers around Deepavali? So I was surprised that despite the rise in demand and consequently the price, sugarcane wasn’t all that costly. That got me down the rabbit hole of the global sugar (and sugarcane) market, and here’s what I could gather.

Though there have been some fluctuations, global sugar prices today are only slightly higher in real terms than they were 25 years ago. This is strange, given the increased demand due to rising incomes during this period.

The major reason is that all major cane-producing countries provide massive subsidies to their sugarcane farmers. Combining these input subsidies with better yields has incentivised overproduction in several countries, such as Brazil and India.

Now, let’s zoom in on the situation in India. The government, as you can expect, has a complex system of price and trade controls to keep sugar affordable for consumers on one hand and to provide a fair price to cane growers on the other. The byzantine controls are listed on the Sugar Policy webpage.

The whole policy is predicated on the fact that sugar is a cheap energy source—an ‘essential’ commodity—for the poor, and hence, its price is administratively controlled, with the government procuring some of the produce for distribution through ration shops. The government also prevents mills from exporting sugar if it foresees a rise in inflation using various instruments such as export duties, bans, and restrictions. For instance, sugar is currently a restricted commodity, meaning that an exporter requires a licence or permission from the government to export sugar. This is despite India’s production far exceeding the projected domestic consumption.

To balance the interests of cane farmers and sugar mill owners, who are both politically powerful interest groups, the government does another complex juggling act. A union government committee announces a Fair and Remunerative Price (FRP) every season to take the price volatility out of the equation. The FRP is what sugar mills are supposed to pay to the farmers per unit of cane procured. Thus, an increase in FRP hurts their profit margins. Hence, the government also fixes a minimum selling price (MSP) for sale by sugar mills.

But the plot thickens further. Some state governments don’t like the FRP. They announce their State Advised Price (SAP), which is pegged far higher than the FRP. These are states where sugar is widely practised, and hence, its farmers are an even more important political interest group. SAPs are often announced ahead of elections. For instance, just this week, the UP government raised the SAP. If this isn’t a ‘freebie’, I don’t know what else is.

The overall effect is that farmers have zero incentive to stop producing sugarcane. If the intended outcome was to make India self-sufficient in sugar at a time when foreign exchange reserves were thin, the policy could be termed a reasonable success.

But the parameter of policy success today need a hard reset. The current mechanism results in two problems.

1. Sugarcane is a water-intensive crop. Cane cultivation generates negative environmental externalities like groundwater depletion, soil degradation, etc. However, these costs are not factored into the pricing process, resulting in overproduction of the subsidised crop despite sustainability issues.

2. The overproduction of cane leads to an overconsumption of sugar. More sugar is particularly terrible for Indians, given our genetic predisposition to diabetes. This factor has become far more significant in today’s context. In the past, perhaps supplying sugar as a low-cost source of carbs to a significant chunk of the poor was a far more important policy objective. But if we are to believe the government’s own projections that only 11 per cent of Indians are multidimensionally poor in FY23, surely the better option would be to subsidise sugar only for this small section and end all the complex manoeuvres in the remaining segments of the supply chain. In fact, the government needs to intervene actively to reduce the information asymmetry regarding sugar consumption rather than encourage the overproduction of sugar.

Sometimes, changing goalposts is necessary because you are playing a different game altogether. So is the case with sugarcane in India.

---

HomeWork

Reading and listening recommendations on public policy matters

1. [Article] This Indian Express article by Ashok Gulati, masterfully titled From Plate to Plough: Drowning in Sweetness, is a must-read.

2. [Paper] This week’s reading was also about critical technologies—a term that governments like to throw about rather casually. This paper analysing the dimensions of such technologies, is terrific. Perhaps the only one which systematically maps policy instruments to policy objectives desired from a technology supply chain.

3. [Blogpost] Urbanomics on high-pitched GST demands resulting from revenue bias.

---