The challenge of attracting Apple to India

This is a very informative and sobering article, well worth a read (HT: Rajeev Mantri). It outlines in great detail the challenges that India will have to overcome to attract the likes of Apple into manufacturing in India. Even with China's problems, attracting the global value chains into India will be a very big struggle. We should be realistic.

Since 2017, its Taiwanese supplier Wistron has been assembling older models sold in India, starting with the iPhone SE and currently the iPhone 7... However, most of the parts that go into these phones are still made in China, these people say. India-based companies supply packaging materials, phone chargers and batteries, but nothing more, they say. Last year, Apple had seven major suppliers operating in India, compared to 135 operating in China, its public disclosures show. The biggest challenge for Apple may simply be the less developed nature of manufacturing in India compared to China...

Finding suppliers that could produce components locally proved difficult. One major issue was compliance with Apple’s supplier standards for health, safety and the environment... One reason why suppliers aren’t willing to improve factories in India to meet Apple’s requirements is because of its small order sizes. Current and former employees at Apple suppliers such as Shenzhen Yuto, Salcomp and Sunwoda, which make packaging, power adapters and batteries, say order sizes in India are in the thousands per month. By contrast, Apple’s orders in China are in the hundreds of thousands per week.

This is a very telling assessment of the Indian market,

Apple’s phones are simply too expensive for the market. While India has a population of nearly 1.37 billion—almost as big as China’s—its middle class is around one-fifth the size of China’s. ... In the 12 months leading up to March 31, Apple’s India subsidiary reported sales of around $1.5 billion, down 19% from a year earlier, according to local corporate filings. It’s a tiny fraction of Apple’s global revenue over the same period of nearly $260 billion... Last year, Apple shipped 2 million iPhones to local retailers in India, compared to 30 million in mainland China, according to Canalys. Because of the low volumes, Apple can’t take advantage of economies of scale and pays 60% more per unit for Wistron to assemble the iPhone in India.

Issues like the far smaller domestic market (than is hyped), poor labour productivity, poor quality of infrastructure etc are not going to be resolved anytime soon. The only things that can be fixed are bringing more predictability to government policy (even that can be a stretch) and giving fiscal incentives (to which there are limits). But the structural negatives appear to be so daunting as to offset anything positive on the policy side.

Much the same challenges would be at play with other manufacturing areas like batteries, defence, railways, thermal BTG, renewables, telecoms equipments etc. Even textiles. 

It's difficult to understand why for a very big country with several country-sized states, India struggles with suppliers (beyond packaging material and phone chargers in case of Apple), entrepreneurs, investors and creation of eco-systems. 

It was thought that many of the structural negatives will be offset by merely being a big country. After all China is not an easy place to do business, though its other advantages more than offsets the formidable problems of navigating the bureaucracy and the Party.

A takeaway is perhaps that instead of going after big names like Apple or Tesla, India should perhaps try to court the likes of Oppo and Xiaomi, who are more likely to find a large enough local demand and also therefore have the incentives to invest in volumes and create eco-systems. And also support their  smaller component manufacturers compared to the likes of Foxconn.

Update 1 (06.04.2020)

As India eyes the unique opportunity of attracting factories from China, the challenges loom. SCMP points to the success with mobile phones PMP,

The number of cellphone and accessories manufacturing units rocketed from a mere two in 2014 to over 260 in 2019. Today, India is the world’s second-largest maker of cellphones, with 95 per cent of them assembled locally.

But the cellphone industry still imports around 75% of components from China, and only 12% is made in India. The challenges are formidable, 

Moreover, despite efforts to boost India’s manufacturing, it consists mainly of assembly lines rather than industrial clusters, so it has some way to go to develop a fully fledged capacity comparable to that of China. Modern manufacturing has a sophisticated multi-tier structure: those directly connected to the assembly line are the first-tier suppliers, who have second-tier suppliers of their own, who may, in turn, have a third tier of suppliers. For example, as one commentator pointed out in Foreign Policy, Volkswagen has around 5,000 first-tier suppliers, each with an average of 250 second-tier suppliers, so it could end up with as many as 1.25 million suppliers. So even if the Indian government successfully attracted some manufacturers with their first-tier suppliers, it would still need to cultivate a suppliers’ network and forge industrial clusters. Furthermore, as Apple’s aborted idea of relocating its iPhone 11 production to India has demonstrated, one should not take assembly-line readiness for granted. Suppliers are able to build large-scale factories in China, employing more than 250,000 semi-skilled workers because the Chinese education system prepares tens of thousands of these battle-ready technicians and engineers. By contrast, India lacks the capacity to provide such skilled and semi-skilled workers. And these problems will linger as long as the Indian government balks at critical reforms for labour and land.

Update 2 (09.04.2020)

Strategic purchases by government can be of help with localisation and shifting the value chains, like with LED bulbs,

Meanwhile, the localization content in LED bulbs is getting brighter. The broad script is similar to mobile phones. While manufacturers assemble in India and have reduced dependence on China for low-value work, component manufacturing is yet to take root. “We are far more detached from China than what we were a few years back," said Sumit Joshi, the chief executive of Signify Innovations India Ltd, the new avatar of Philips Lighting India Ltd. “About 98% of what we sell in India is made in India. I don’t depend on China for finished goods," he added. Joshi said the dependence on Chinese components is dwindling. “We are now able to make 60% of the components in India," he said. Lighting components imported from China include LED chips, transistors, and resistors.

There are two reasons why lighting companies de-risked from China. Indian consumers behave differently from global consumers and two, the Indian government became a big buyer. The Union government’s Unnat Jyoti by Affordable LEDs for All (UJALA), a scheme to provide LED bulbs to consumers and replace 770 million incandescent bulbs, began in 2015. The overall LEDs distributed through the programme totals over 360 million until 16 March 2020. Since the mandate was to make bulbs affordable, LED makers started developing local supply-chains. How are Indian consumers different? Joshi cites the example of the still-preferred tube light in Indian homes. “We came up with the Philips T bulb. It is more of a horizontal kind of a bulb because Indian consumers like brightness and a throw of light," he said. To switch over from conventional lighting to LEDs, Indian consumers also demanded longer warranties. Signify offers a two-year warranty on a bulb, a small value product. It can’t afford returns. The company, therefore, ensures that the bulbs tolerate power surges, common in many parts of the country. This customization ensures that lighting companies are less dependant on imports.

The vanities of Silicon Valley billionaires

Tyler Cowen has an interview of Reid Hoffmann, founder of LinkedIn, which has some interesting Freudian slips. It reveals the insecurities and vanities of the Silicon Valley billionaires and the hypocrisies among the liberal elites. Some thoughts.

1. It is no good merely being the founder of PayPal or Tesla or LinkedIn. There are now a good number of these founder billionaires, just in Silicon Valley itself. The megalothymos of these individuals is now driving the urge to differentiate themselves within even this rarefied group.

Being perceived as "weird" has become an aspirational imprimatur, a signature worn with great pride.

2. Then there is another urge. Founding a company, howsoever big, only makes you a successful businessman. There is the lurking realisation that they were blessed with extraordinary good fortune in their business success. After all several entrepreneurs were pursuing the same idea at exactly the same time as them in the Valley itself, and any one among them could have emerged successful. Then there is the inherent winner-takes-all feature of an industry characterised by network effects.

This subliminal realisation that they won a lottery can be intolerable for big egos. So the need to construct a new image, to weigh in on and be heard on public issues at a global stage, become opinion makers who are taken seriously, be seen as thought leaders, and some even as philosophers. Re-invent themselves as public intellectuals.

So they have enlisted the services of the so-called serious people (the likes of researchers and public figures) to re-construct their public images. In the name of endowments, some such have even been purchased. One list even names Peter Thiel, Elon Musk, and Mark Zuckerberg as "public intellectuals"! How does founding a digital payments company make a person a public intellectual on religion and epistemology?

True, there are exceptions. But we are talking about the norm.

3. The gobbledygook that drains out from publicity providing conversations like the above is symptomatic of the world in which we live. One where substance takes backstage.

Funnily, the more interesting readings are in the comments section, where wisdom, conspicuously missing in the conversation itself, often surfaces. This comment sums it all,

Wittgenstein's reputation has endured many slights but to blame him for the design of LinkedIn is surely the deepest cut of all.

4. The interview questions are so fawning as to make the whole thing very awkward for a reader/listener. Sample this,

COWEN: If we think of Peter Thiel and Elon Musk, they could arguably, by the standards of many people, be called weird. I’ve reviewed all the books you’ve written and a lot of your public talks. I can’t recall you saying a single thing that’s outrageous in any way whatsoever. Why aren’t you weirder?

And this,

COWEN: What is the early influence on your thought that you feel you have and other major figures in Silicon Valley maybe don’t? Because a lot of people have played board games growing up, right? A lot have read early science fiction. But you’re different. Is there more New England in you? Or what’s the missing variable? 

HOFFMAN: Well, it might be that the missing variable is that I deeply value intellectual work and public intellectual work, that I found —

Recasting a MUDRA 2.0

Very good article in EPW by Deepti George on the Pradhan Mantri MUDRA Yojana (PMMY). She suggests that the MUDRA Bank could shift its focus away from refinance and play a direct role in offering credit guarantees on the loan book of the lending institutions; and credit enhancements for the securitization of their MUDRA loans. 

Unlike direct lending or investment, through this the MUDRA Bank’s capital would go longer by being able to leverage a far greater volume of financing for the MUDRA loan lending institutions.

In the context of reforms to PMMY, a few more suggestions:

1. All MUDRA beneficiaries, except perhaps a few categories of Shishu loans, should be encouraged to maintain some basic level of book-keeping of its accounts. A freely available App or web-enabled solution (for different categories of businesses) for this would be the first step in creating a supporting eco-system for these businesses. Similarly, they should also be encouraged to transact digitally. APIs to this solution can be made public to allow fintech and other internet businesses offer different types of services to these MUDRA beneficiaries. This, over time, can help catalyse an enabling eco-system to support the growth of such SMEs and start-ups.

2. Rather than mandatory requirements on these, a better idea may be to make it costlier (both in terms of financial cost as well as effort required) to continue with the business as usual approach as well as more attractive and beneficial to adopt change. The latter can even involve some form of reimbursement on meeting the requirements which can be monitored digitally (say, a waiver of the last repayment tranche or a small interest subvention for compliance with verifiable book-keeping).

3. Availing at least a few types of Kishor and, especially, Tarun loans could be made contingent on some simple and easily captured/observed metric of outcomes (eg. a couple of relevant metrics on growth of output and employment from GST or some other statutory compliance reporting).

4. Tarun loans (Rs 5-10 lakhs) to start-ups could be bundled with privately provided management capacity building support (or management capacity extension service). The management capacity enhancement support could be optional, with partial subsidy. The access to such support should be made easy. Besides, some of the SMEs, as required, could be connected with GEMS, the government e-marketplace. After all, State Finance Corporations (SFCs) are supposed to provide both credit and hand-holding support to its SME borrowers. 

5. Given the massive volumes of such transactions, this is a good window for capturing credit histories and creating the conditions for lowering future credit access costs for the vast majority of compliant MUDRA beneficiaries. This becomes even more valuable and actionable given that Aadhaar is now a requirement for MUDRA transactions. Accordingly, MUDRA Bank could take the lead in creating the systems to capturing and consolidating and making available the credit trails from MUDRA transactions without compromising on privacy requirements. 

6. The objective behind this approach should be not to capture digital trails and thereby expand the tax base or detect tax evasion, but to help the SME develop capacity and grow sustainably as well as to create an eco-system to support that growth. The latter should, if at all, be a by-product. This needs to be signalled pro-actively so as to shape positive expectations and alleviate concerns. It is important that all government stakeholders involved in this approach their roles from this perspective.

7. The MUDRA Bank should have a research unit that analyses the data emerging from MUDRA loans and its refinance and generates insights that are useful to both itself and the lending institutions to continuously iterate and refine their process to enhance the quality of their service delivery and reduce losses. The Research Unit could enlist partnerships with reputed research institutions, think-tanks, and credible researchers in this effort.

These reforms cannot be decreed into action through legislations, much less guidelines or directives. Their implementation demands focused and long-drawn effort on a mission-mode. This also means appointing a co-ordinating official with the convening power to engage meaningfully and quickly on multiple dimensions. Like the example of eNAM, about which I blogged here, this too is a wicked problem, where implementation is of essence. 

Weekend reading links

1. The Times has an assessment of the possible responses to coronavirus, which it describes as a rare combination of both supply and demand-side shocks.

As the report writes, the substantive value of a monetary policy cut now is questionable, beyond being a measure to reassure the financial markets. Experts favour fiscal policy response,

Governments have tools that could limit the costs, but have been reluctant to use them, economists said. They could give cash to employees whose workplaces are shut, provide credit to small businesses and offer rescue packages to industries most affected, like airlines and other tourism-related concerns.

See also Larry Summers here - central banks to keep credit taps flowing, fiscal expansion etc.

One observation that I have heard is that the medical costs of the epidemic may be lower than the economic costs that will be inflicted due to an over-reaction. Only time will tell.

2. Good example of urban regeneration in the 30-block Hudson Square area of New York city sparked off by the decision by Disney to relocate its headquarters by building an SOM designed 19-story, 1.2 million square foot office and studio complex. The area, once a hub for printing-press businesses has struggled to reinvent itself since the 1990s.

Needless to say, this time too has come with its gentrification costs.

3. After the initial mishandling, the Chinese response to the coronavirus appears to have been very impressive. Mobilising resources in a big way and acting swiftly and decisively.

4. Tristan Harris, a former employee at Google, calls for EU to take the lead and regulate Facebook and Google as "attention extraction utilities".

First, the EU should create a new corporate classification for large, dominant social platform businesses that have created vital public digital infrastructure. These “attention utilities” should be required to operate in the public interest, according to rules and licences that guide their business models. Traditional companies have been subject to licensing for many years. Attention utilities should be required to obey limits on data extraction and message amplification practices that drive polarisation, and should be required to protect children. We should ban or limit microtargeting of advertising, recommendations and other behavioural nudges. Second, instead of relying on revenue based on advertising, attention utilities should be required to convert to a monthly licence fee model a bit like the BBC or a subscription like Netflix. They must adhere to the terms of an operating licence framed by a duty of care. EU antitrust commissioner Margrethe Vestager has suggested Facebook should use a subscription model. Third, attention utilities should be subject to a “social impact assessment”, in which their new products are evaluated for their potential impact on mental health, social isolation, fake news, polarisation and democracy. This pre-clearance would be akin to an environmental impact assessment or safety protocols used for medical devices.

5. Nice article about living in central London.

Central London had 5m residents before the second world war. That had halved by the 1980s and is now creeping up again above 3m, as city-centre living becomes cool again.

6. Morgan Hausel points to three more inevitabilities of life other than death and taxes (HT: Ananth).

1. Talent will cluster around tiny groups of people because people like associating with winners, so success snowballs.

2. The boom-bust cycle will never go away, because most people capable of driving growth don’t have the kind of personality that can quit while they’re ahead.

3. There are social problems that will never be solved, because they evolve and adapt to solutions.

This a very good article.

7. As the Covid 19 outbreak continues to spread, the biggest concern is about its likely impact on the economy. One of the strongest candidates for a channel of transmission is the debt markets. Consider this,

According to the Institute of International Finance, a trade group, the ratio of global debt to gross domestic product hit an all-time high of over 322 per cent in the third quarter of 2019, with total debt reaching close to $253tn... This is particularly important because much of the debt build-up since the global financial crisis of 2007-08 has been in the non-bank corporate sector where the current disruption to supply chains and reduced global growth imply lower earnings and greater difficulty in servicing debt. In effect, the coronavirus raises the extraordinary prospect of a credit crunch in a world of ultra-low and negative interest rates....

A recent OECD report says that at the end of December 2019 the global outstanding stock of non-financial corporate bonds reached an all-time high of $13.5tn, double the level in real terms against December 2008. The rise is most striking in the US, where the Fed estimates that corporate debt has risen from $3.3tn before the financial crisis to $6.5tn last year... The IMF’s latest global financial stability report amplifies this point with a simulation showing that a recession half as severe as 2009 would result in companies with $19tn of outstanding debt having insufficient profits to service that debt.

Consequently, as John Plender writes, the risks are also much higher now,

... banks... are not as heavily exposed to corporate debt as investors, such as insurance companies, pension funds, mutual funds and exchange traded funds. That said, banks cannot escape the consequences of a wider collapse in markets in the event of a continued loss of investor confidence and or a rise in interest rates from today’s extraordinary low levels. Such an outcome would lead to increased defaults on banks’ loans together with shrinkage in the value of collateral in the banking system... The OECD report notes that compared with previous credit cycles today’s stock of corporate bonds has lower overall credit quality, longer maturities, inferior covenant protection — bondholder rights such as restrictions on future borrowing or dividend payments — and higher payback requirements. Longer maturities are associated with higher price sensitivity to changes in interest rates, so together with declining credit quality that makes bond markets more sensitive to changes in monetary policy. Current market volatility is further exacerbated by banks’ withdrawal from market-making activities in response to tougher capital adequacy requirements since the crisis...

In a downturn, some of the disproportionately large recent issuance of BBB bonds — the lowest investment grade category — could end up being downgraded. That would lead to big increases in borrowing costs because many investors are constrained by regulation or self-imposed restrictions from investing in non-investment grade bonds. The deterioration in bond quality is particularly striking in the $1.3tn global market for leveraged loans, which are loans arranged by syndicates of banks to companies that are heavily indebted or have weak credit ratings. Such loans are called leveraged because the ratio of the borrower’s debt to assets or earnings is well above industry norms. New issuance in this sector hit a record $788bn in 2017, higher than the peak of $762bn before the crisis. The US accounted for $564bn of that total.

8. Nice set of graphics in WSJ explaining the transmission channels of coronavirus from China.

9. Finally, a good representation of China moving up the textiles value chain,

This about the difficulty of unshackling of the global textiles value chain from China is important,

China’s share of global clothing exports ebbed from 37% in 2010 to 31% in 2018, according to the World Trade Organization. Yet over the same period, China’s share of global exports of textiles—which is made into apparel—rose to 38% from 30%. Producing textiles is highly automated, which means China has remained a favored destination despite rising labor costs. Chinese fabrics are shipped to Vietnam, Pakistan and Bangladesh for labor-intensive cutting and sewing, accounting for roughly half to three-quarters of their textile imports. These newer manufacturing nations also depend on China for more complex products, such as zippers and fasteners, that rely on higher-skilled workers... In Bangladesh, factories can’t easily stockpile fabrics and other materials because they don’t know what styles of clothing buyers will request. Both Vietnam and Bangladesh are increasing textile production, but industry experts say they will rely on imports for years to come.

10. Finally, a fascinating peek at the economics of buffets (HT: MR).

Given low margins, it is economies of scale that makes buffets profitable. 

PPPs and infrastructure financing - A summary

The newly launched India Infrahub has a series of my articles on PPPs and infrastructure financing.

1. The first article outlines the challenges ahead as India pursues its INR 102 trillion infrastructure pipeline.

2. The second article categorises infrastructure from the perspective of commercial viability and private financing.

2. The third article talks about the challenge with attracting long-term finance for infrastructure.

3. The fourth article explains how China finances its infrastructure.

4. The fifth article explains the problem of delays and cost overruns with infrastructure projects.

5. The sixth article makes the qualified case for PPPs.

6. The seventh article talks about the role of development finance institutions in infrastructure.

All these articles draw from this very long working paper. 

India's capital accumulation challenge

In Can India Grow?, we argued that India did not have the capital foundations to sustain high growth rates for long periods. It needed to accumulate capital on multiple dimensions - financial, physical, human resource, and institutional.

Reinforcing the point, The Economist writes,

A lack of technology transfer is only part of the problem, however. Half of the slowdown in labour-productivity growth in recent years reflects not a failure to imitate but a failure to accumulate: weak investment has left labour with too little capital to work with. This shortfall in investment explains all the productivity slowdown in South Asia, the Middle East and north Africa, and two-thirds of that in Europe and Central Asia.

Also, I am inclined to believe in Dietz Vollrath's assessment of the institutions literature,

The point is that even if we acknowledge that "institutions matter", that does not imply that we can or should propose institutional reforms to generate economic development. It's a mistake to think of ceteris paribus changes to institutions. They are not a thing that we can easily or independently alter. If they were, then they wouldn't be *institutions* in the way that Douglass North uses the term. If you want to generate economic development, the implication of the institutions literature is that you have to reform the underlying distribution of economic power first. Once you do that institutions will endogenously evolve towards the "good" equilibrium, whatever that may be.

It helps with accumulation if you have some baseline equitable distribution of the means of production (land, in particular) coupled with investments in human resources capacity. This is one of big takeaways from the contrasting fortunes of North and South East Asian economies as chronicled by Joe Studwell. 

Does India suffer from an acute version of baseline inequitable resource distribution and sorely deficient poor human resources capacity?

Wicked problems and e-NAM

Shoumitro Chatterjee and Mekhala Krishnamurthy have a very good article in Seminar on the challenges of the implementation of the electronic National Agriculture Market (eNAM). 

eNAM is a great paradox. The vast majority of opinion makers think it is some simple plug-and-play low-hanging fruit in agriculture. Deregulate and let farmers trade their products freely. Create the enabling environment and we will have a national marketplace, an Amazon for Indian agriculture produce! Even the more perceptive ones don't realise the enormity of the challenge. In reality it is a very wicked problem, a strong reminder case study of why technology and process re-engineering interventions alone cannot make much dent on persistent development challenges.  

Consider an illustrative list challenges.

1. The barter nature of farm gate trade - farmer being yoked to the current buyer for various difficult to overcome reasons (buyer/broker offers other services which farmer needs but are not available from elsewhere)
2. Small farmers do not access the physical mandis in any case and sells their harvest at farm gate
3. The need to simplify the process at the mandi itself - so that farmer does not need to wait more than now, does not experience more uncertainty than now, buyers are not similarly hassled, and so on.
4. Limited post-harvest management at the farm gate means that there will have to be full-fledged assay at the mandi to be able to do a credible digital transaction - creation of awareness and capabilities to do a credible basic assay at farm gate, so that the mandi assay is simpler/easier
5. Credibility of the assay standards and its compliance - awareness and confidence creation among the sellers, and more importantly, the distant online buyers
6. The inherent limitations of a (physically centralised, in so far as farmers have to come to the mandi and get their produce assayed) digital mandi-based process in handling large transaction volumes when sales are usually concentrated into a few days in the immediate post-harvest aftermath
7. The widespread practice of instalments-based part payments or recovery for the downstream inputs by buyers based on mutual trust (inherently localising transactions)
8. The sanctity of an eNAM contract, in terms of the farmer delivering the tested grade and trader honouring on the payment, and mechanism to ensure compliance

It is exactly the sort of stuff that demand extremely high state capacity in both design and implementation. The authors have done a nice diagnosis. The challenge is how to go about solving the problem - or a practical design and implementation plan for an eNAM? Is there a framework to think about a comprehensive policy proposal that can then get parcelled out to different stakeholders/agencies to attend and then be monitored closely over a period of say, 5-8 years?

What is it that can be done on each and by whom and in which sequence? Given that each will take time, what is the best strategy for each challenge so that we can have a gradual process of full adoption of e-NAM? How can traditional agriculture extension system be reformed to address these new challenges? How can private sector eco-system for the associated set of services be catalysed?

No easy answers here. It is clear that this requires responses at three levels of governments - Government of India and State governments for broad policy changes; district administration for engagement with the mandis and making them work; and agricultural extension services to create awareness, change behaviours, nurture capacity etc.

In simple words, these are the domain of implementation, at the district and below, and all of them will require painstaking and long-draw action, filled with uncertainties. These are the sort of activities which defy a neat end-to-end plan. There will be partial and general equilibrium factors on which state and central governments will have to act pre-emptively as well as respond to emergent challenges. They demand initiation of a minimum viable product and then active engagement to spot both emergent problems and opportunities, and address them swiftly. In some of these cases, private providers (say, App providers, if APIs are made available) will be able to offer some services and they will require enabling conditions.

Update 1 (18.05.2020)

Ashok Gulati is excited by the Agriculture marketing side reforms announced by the Finance Minister,

They relate to amending the Essential Commodities Act (ECA) of 1955, bringing a Central legislation to allow farmers to sell their produce to anyone, outside the APMC mandi yard, and having barrier-free inter-state trade, and creating a legal framework for contract farming — the buyer can assure a price to the farmer at the time of sowing.