Data Plans

WERE IT NOT FOR THE SUITS, sombre faces and plush chairs, the launch of Digital India would have had the aura of a pop concert. It was the first day of July in 2015, and thousands of people had filled the arena of the Indira Gandhi Indoor Stadium in Delhi. When Prime Minister Narendra Modi entered, the crowd erupted with cheers. Before the presentations began in earnest, jumbotrons played a digital animation accompanied by heavy electronic music. Light—sometimes blue, sometimes yellow—bathed the stage. Modi himself sat at the centre of the onstage dais, flanked by the finance minister, Arun Jaitley, the electronics and information technology minister, Ravi Shankar Prasad, and a supporting cast of other officials.

Behind and just to the right of Modi, in the middle of a mostly out-of-view back row of businessmen, sat Mukesh Ambani. When it came time for these men to speak, Ambani—the richest man in India and the chairman of Reliance Industries Limited, India’s largest private company—went first. Referring to the event as “a momentous occasion in the history of modern India,” Ambani praised India’s leader. “India is fortunate to have a prime minister who not only paints a compelling and comprehensive vision but has the personal leadership and drive to convert his vision to reality,” he said. Modi, wearing a white button-down jacket, stared forward blankly. “Normally, industry moves faster than government, but with Digital India it is different,” Ambani said. “I have no hesitation in saying that government has moved faster.”

Ambani then spoke of his own company’s ambition, and how it aligned with Digital India. Reliance Jio, RIL’s telecom subsidiary, had not yet launched. Nevertheless, Ambani promised to partner with the government and invest over Rs 250,000 crore, or $39.29 billion, in its projects. He pledged to lay out a next-generation wireless network across all Indian states, create a nationwide cellular distribution network involving 150,000 retailers, encourage phone manufacturers to set up shop in the country, and help small start-ups. “I estimate that Reliance’s Digital India investments will create employment for over five lakh people,” he said.

Digital India is an amalgamation of various projects, aimed at increasing connectivity in the country. Promoted by government officials as an attempt to bridge the divide between “digital haves and have-nots,” the initiative has been pitched as an ambitious modernising measure. According to the Digital India website, its various projects, including the controversial Aadhaar, are all unified by a three-part vision: making online infrastructure available to every Indian, providing government services on demand and the “digital empowerment of citizens.”

When Jio formally launched over a year later, Indians woke up to front-page advertisements for the company in the Times of India and the Hindustan Times that featured a large image of Narendra Modi below the Jio logo. “Jio is dedicated to realising our Prime Minister’s Digital India vision for 1.2 billion Indians,” it read. Social-media users speculated whether using the prime minister’s name or photograph in promotional material without prior approval was illegal. Arvind Kejriwal, the chief minister of Delhi, tweeted: “Modiji Ambani ki jeb mein”—Modi is in Ambani’s pocket. In December 2016, the information and broadcasting minister, Rajyavardhan Singh Rathore, told the parliament in a written reply that the government was aware that RIL had used Modi’s image, but that it had not granted the company permission to do so. In March 2017, RIL apologised for putting the prime minister’s face in its ads. In an editorial, the Economic and Political Weekly suggested that beyond legality, “the question to be asked is why it suits both—Reliance and Modi—to be seen as sharing a vision and what implications that has for RJIO’s future.”

In almost every public statement about the company, Ambani has continued to reference Digital India, claim credit for the country’s growing internet usage and promise even bigger gains. At RIL’s 2016 annual general meeting, Ambani began his speech “with a dedication” to Modi and Digital India. At the following year’s meeting, he said that while “analysts believed that India can never be the largest mobile data market in the world,” Jio had “proved them wrong.” In a 2018 speech at an investment summit in Uttar Pradesh, Ambani said that Jio would “assist the administration from rajdhani Lucknow to the smallest gram panchayat.” Modi was sitting in the front row.

ACCORDING TO THE TELECOM REGULATORY AUTHORITY OF INDIA, or TRAI, there are currently 1.17 billion mobile-phone subscriptions in India, an increase of roughly 140 million subscriptions since August 2016—the month before Jio launched. The growth is especially pronounced in rural areas, where there are now over 500 million wireless subscriptions, roughly 80 million more than there were before the company formally began its operations. As more Indians gain phone subscriptions, more are coming online. A 2017 report by the Internet and Mobile Association of India estimated that nearly 50 million Indians gained internet access between December 2016 and December 2017, allowing many of them to surf the web, send WhatsApp messages and stream videos for the first time.

For Mukesh Ambani—who makes most of his money in the petrochemicals business started by his father, and who had limited experience in telecom before Jio burst onto the scene—the growth in subscriptions and data consumption are all proof of Jio’s transformative social power. “Today, India is the fastest digitising economy in the whole world,” he said in November 2018. “Jio is leading India’s digital transformation.” By offering service at almost no cost, and at first for free, Ambani argues that Jio will “democratise the digital culture in India.”

But the branding is not just about transforming India. Ambani has been described as “an extreme innovator and believer in game-changing businesses of the future.” In its annual report for the financial year 2016–17, RIL—by far the biggest of the three private-sector Indian companies to make the global Fortune 500—argued that its telecom arm is “the world’s largest startup.” The company says it now has over 250 million subscribers.

The media has largely eaten it up. “Two Years Ago, India Lacked Fast, Cheap Internet—One Billionaire Changed All That,” declared a September 2018 headline in the Wall Street Journal. In Fortune magazine’s most recent annual list of companies that “help the planet and tackle social problems,” Reliance Jio was ranked number one.

And yet, the story of Jio’s success is not as rosy as it first seems. A series of suspiciously timed changes in regulations by the TRAI have allowed the company to sail towards the top. It has undermined the health of its competitors by offering prices that many analysts argue are predatory, despite laws meant to monitor and restrict anti-competitive activities.

Mukesh Ambani established his position within the telecom industry by acquiring a small, obscure firm that had won an auction for national wireless spectrum one day earlier. To win, the firm bid roughly a hundred times what it was worth. According to the spectrum licence, the firm was only permitted to provide internet service. But three years later, the government allowed that firm, by then renamed Jio, to upgrade to a “full mobility” licence and offer every kind of cellular service for a fraction of what its full-mobility licences should have cost.

The TRAI came under sharp criticism from the rest of the telecom industry for allowing RIL to “test” the connectivity of its network for 253 days—many degrees higher than what, till then, was the norm. The regulator also raised eyebrows for eliminating data share from its definition of what constitutes “significant market power” not long after Jio became India’s largest data provider. This allowed Jio to sidestep competition laws and stricter regulatory scrutiny. Only companies that have significant market power, the TRAI wrote, could be viewed as engaging in monopolistic behaviour.

“The telecom sector in India has a multi-decadal history of regulatory capture,” Saurabh Mukherjea, a telecommunications expert and the former chief executive of Ambit Capital, told me. “Everybody in the sector knows that capturing the regulator can be a major source of competitive advantage.” Mukherjea said that while this was true of almost every regulated industry within India, telecom was especially notorious, full of periodic “strange interventions” by the government that benefit only a select few companies.

The government has denied allegations that it is favouring RIL. Jio, it explains, is just doing a good job at expanding the market. “They provide new services at a low price,” said Anil Kaushal, who was one of the TRAI’s governing members from April 2015 to April 2018. “Now, with Jio in line, everybody is offering the same low prices, and the consumer is benefitting.”

Jio may indeed be expanding India’s data market, but the company’s debut has created mounting troubles for the industry. Reliance’s rock-bottom data pricing have forced competitors to slash their tariffs. The result has been rising levels of corporate debt that will be difficult to pay off. By March 2018, India’s telecommunications companies had a cumulative debt of over $75 billion, or roughly Rs 500,000 crores, but only $27.6 billion, or about Rs 180,000 crore in revenue, a level that many telecom insiders worry has troubling implications for the broader economy. “Domestic banks and equipment vendors have been put under financial stress,” said Rajan Mathews, the director general of the Cellular Operators Association of India, an industry group. This, he and others worry, could slow the country’s overall growth.

Jio’s entry has already prompted a wave of mergers and acquisitions, resulting in fewer industry jobs and a collapse in competition. At the end of 2015, when Jio was formally launched, India had nine private-sector wireless providers. Today, there are effectively three—Bharti Airtel, Vodafone–Idea and Jio.

In a recent interview with the Economic Times, the telecom minister, Manoj Sinha, tried to allay fears that any one company might come to dominate the industry, thus giving it control over Indians’ access to the internet. “There is no question of any monopolistic situation arising in the telecom sector,” he said. “We don’t think the number of operators will reduce further.”

Not everyone agrees. Not even, it seems, the Ambanis. According to the Economic Times, at his company’s 2018 annual general meeting, Anil Ambani—Mukesh’s younger brother who runs his own business empire—said that high costs and aggressive tariffs had created an “oligopolistic structure that was now rapidly moving toward a duopoly, and could eventually become a monopoly.” Anil had nearly closed a deal to sell his own, debt-laden telecom venture, named Reliance Communications, to Mukesh. Jio, however, has insisted that it not be held accountable for Reliance Communication’s liabilities, putting much of the deal on hold. In the interim, the Swedish telecom manufacturer Ericsson has asked the Supreme Court to jail Anil until he pays the Rs 550 crore—or $77 million—he owes it for helping manage and operate Reliance Communication’s network.

Reliance Communications, previously known as Reliance Infocomm, had been a part of RIL until a mid 2000s fight between the two brothers led it to be spun off. Typically, family feuds do not have macroeconomic implications, but Reliance had disproportionate stakes in too many of India’s major industries to avoid disrupting markets.

DHIRUBHAI AMBANI, MUKESH AND ANIL’S FATHER, was born into a less-than-affluent household in Gujarat. He did not attend college and moved to Yemen as a teenager to work as a petrol-pump attendant, and later, as an operations manager for Shell—one of the world’s biggest petroleum companies. When Dhirubhai returned to India in 1958, he set up a small trading shop for yarn and spices, and began the slow but steady process of transforming it into India’s largest private-sector company: Reliance.

Mukesh, born in 1957, and Anil, born in 1959, grew up watching their father’s spectacular rise. At first, the family lived in a chawl in south Mumbai, but by the time Mukesh turned 12 years old, the family had moved to a much more comfortable residence in Altamount Road, one of the city’s most high-end neighbourhoods. Bolstered by Dhirubhai’s resourcefulness and, much more controversially, his tendency to “take advantage of a particular system” (in the words of one contemporary colleague), the company expanded rapidly. By 1976, Reliance was India’s 67th-largest private-sector business.

The next year, Dhirubhai named a 20-year-old Mukesh to the company’s board of directors. At the time, Mukesh was working towards his bachelor’s degree at the Institute of Chemical Technology in Mumbai, where he studied before going to Stanford University to pursue a master’s degree in business administration. In 1983, a 24-year old Anil was named Reliance’s co-chief executive officer. He had just graduated from Wharton with an MBA.

Reliance was not without controversy under Dhirubhai’s leadership. During the 1980s, the Indian Express discovered that the son of Mohan Katre, then the director of the Central Bureau of Investigation, was secretly employed by Reliance. During Katre’s tenure, the CBI arrested the journalist S Gurumurthy, who had made a name for himself investigating Reliance, under a colonial-era statute designed to protect the British Raj from nationalist critics. Katre was also famous for doggedly pursuing cases against Bombay Dyeing, Reliance’s main competitor in the petrochemicals industry. Once, Katre even sat in court during a passport-fraud case involving Bombay Dyeing’s chairman—a case to which the CBI was not a party. These controversies did little to dent Reliance’s rise. Neither did others. By the beginning of the twenty-first century, Reliance was the dominant player in India’s petrochemicals market. In 2012, the British Plastics Federation estimated that Reliance produced roughly 75 to 80 percent of India’s polypropylene—a versatile polymer used in everything from bags and straws to car bumpers and chairs.

Despite this, Dhirubhai was perceived “as the business underdog trying to break through the government’s red tape and the prejudices of a tired Bombay business establishment,” as Hamish McDonald, journalist and Dhirubhai’s unofficial biographer, wrote. But for Dhirubhai, good appearances and relationships were ultimately transactional. The more lasting legacy left for his sons was a lesson in the importance of having friends across political lines and being ruthless with competitors. “The orbit goes on changing,” he told McDonald. “Nobody is a permanent friend, nobody a permanent enemy. Everybody has his own self interest.”

To hear Mukesh discuss telecom today, it sounds as if providing cellular service has long been a core goal for both himself and for RIL, and that this is driven by altruism. But in the mid 1990s, when an undivided Reliance was just getting started in communications, Mukesh was preoccupied with overseeing petrochemical plants, and his interest in the telecom sector was very limited. “The guy who was the head of affairs for telecom was Anil,” Brijendra Syngal, the first chairperson of Reliance’s telecom arm, told me. “He was looking into telecom. As a matter of fact, Mukesh paid hardly any attention to telecom.”

Before joining Reliance, Syngal was the managing director of Videsh Sanchar Nigam Limited, the government monopoly charged with providing internet service within the country. The Ambanis, Syngal said, hired him and a small team of associates in 1998 to try and make something of the hodgepodge collection of telecom licences that Reliance had obtained in the free-for-all that followed telecom liberalisation. They found themselves stymied by a relative lack of interest from either brother. “Mukesh, up until then, was very busy with the refinery. So he didn’t pay much attention to what was happening in telecom,” Syngal said. “Anil was also interested some days, not interested some days.”

The Ambani family’s petrochemicals workload certainly distracted from telephony, but they were also unsure whether business in telecom could make a profit at all. Mukesh has said that his telecom ventures have been inspired by “Dhirubhai Ambani’s dream of enabling subscribers to communicate at a cost that is less than the price of a postcard.” But Dhirubhai had a corollary maxim. “He also said that unless we can make telecom as cheap as the price of a postcard, it’s not worth getting into,” Syngal told me. The logic extended to his elder son. “Unless he’s convinced of a proposal, he’ll take no decision,” Syngal said of Mukesh. “He will not invest.”

Tired of the apathy, Syngal and his partners threatened to quit. “Our group was getting very frustrated,’” Syngal told me. “We gave our letters, and Mukesh then called. He said, ‘Mr Syngal, please make me understand what is telecom. Because each day, I listen to a different tune on telecom.’” The conversation kick-started a series of meetings between Syngal, the Reliance board and senior leadership, designed to persuade Mukesh that the company could make a sizeable profit in the industry. By January 2000, Mukesh was convinced. “He decided to invest in telecom,” Syngal said. “And from there he took complete charge of it.”

Working for Mukesh was different than working for Anil, who, Syngal said, paid relatively little attention to his team and took a passive approach to decisions. Mukesh, by contrast, was far more studious and required convincing. But what both had in common was an individual coterie of friends-turned-employees, whose own behaviour belied Anil and Mukesh’s kindness. “I was never treated badly by the brothers. Never,” Syngal told me. “But the people around them, they were vicious.”

Syngal wondered if the contrast between brother and coterie was intentional. “It wasn’t clear what were the instructions to the kitchen cabinet, and what were the instructions to us,” he said. “Whether they were the same or they were different, or whether they were meant to say, ‘Oh, we’ll be nice to these guys, but you create conflicts.’” The good-cop-bad-cop routine injected a degree of chaos into the daily life of Syngal’s employees, and it turned Syngal into a fall man. “There were not very pleasant happenings in board meetings, the staff getting insulted, me taking it on me,” Syngal told me.

After a particularly bad falling out with Manoj Modi, Mukesh’s right-hand man, Syngal resigned—a decision that he blamed, in part, on himself. He had imagined that the breadth of his previous telecom experience had made him indispensable to the Ambanis. “But I didn’t realise the working of the kitchen cabinets,” Syngal said. “Maybe I should have sucked up to them and listened to them and moulded myself to their way of doing things.”

Reliance, of course, moved ahead anyway. On 27 December 2002, Mukesh formally inaugurated Reliance Infocomm. Like most major Reliance events, the launch party was a spectacular affair. Among the attendees was Pramod Mahajan, then the minister of communications and information technology. Atal Bihari Vajpayee, the sitting prime minister, read an original piece of poetry by videoconference.

Notably absent, however, was Anil, who claimed that he could not attend because of an illness. The excuse was flimsy, and it gave off more than a whiff that something was rotten in the state of Ambani. “For many months now, the corporate grapevine has been rife with reports that all is not well between the two brothers,” the business journalist Sucheta Dalal wrote. By not showing up, Dalal said, Anil had “stolen the show.”

The tension, however, preceded their father’s death in 2002. During his three-year stint at Reliance, which ended not long before Dhirubhai’s demise, Syngal told me that there was little love lost between Anil and Mukesh. “Frosty,” he said, when I asked what fraternal relations were like. “We used to discuss amongst ourselves that everything isn’t as rosy as it appears.” His team did not usually see the two attend the same conferences. “You’d go to Anil, [tell him] the meeting is starting, he’d say ‘Ah, go on, go on, go on. Mukesh is there. He can handle.’”

In public, the siblings displayed unity in the wake of their father’s death. But it didn’t take long after Reliance Infocomm’s launch for the relationship to openly deteriorate. Control over the communications subsidiary became a key component of what turned into a years-long dispute over who would govern which parts of the company. With the help of a Rs 12,000-crore investment from its parent, Reliance Infocomm was valued at Rs 60,000 crore by 2004. Fresh off being crowned India’s “Telecom Man of the Year” by Voice & Data magazine, Mukesh did not want to surrender the subsidiary to his brother, who was purportedly alleging that Mukesh had sold himself many of its shares at a heavily discounted rate.

But when the time came to divide the company between the two in a somewhat equal fashion, it was impossible for Mukesh to keep both Reliance’s core petrochemicals business and its telecommunications arm. In 2005, Reliance Infocomm went to Anil. “I have today amicably resolved the issues between my sons, Mukesh and Anil, keeping in mind the proud legacy of my husband,” Kokilaben Ambani, who oversaw the partition plan, declared upon its announcement.

In theory, Kokilaben’s plan ought to have also prevented her sons from having intercompany fights. Mukesh and Anil signed a non-compete agreement, each promising not to enter industries where the other had an established interest. But there was no real truce. Anil’s group cut off service to Mukesh’s company. Under Mukesh, RIL tried to void the terms of a gas supply agreement it had signed with Anil’s holdings. The two billionaires—whose father had died as one of India’s richest men—competed to outdo each other’s net worth. In 2008, it was close. With $42 billion, Anil Ambani was India’s third-richest man. But with $43 billion, Mukesh was number two.

Anil would likely have been wealthier if he had acquired the South African telecommunications giant MTN, as he almost did that June. At the time, Reliance Communications, as it was renamed under Anil’s group, had grown substantially, and, with 45 million subscribers, was the country’s second-largest cellular network. Anil and MTN’s chairman had settled on an agreement that would have merged the two entities and largely transferred control to Anil’s organisation, putting him in charge of a telecom entity that would have boasted more than 100 million subscribers across two continents. But the agreement was scuttled after Mukesh wrote to both MTN and Anil, asserting that each brother had the right to stop such mergers per the terms of their divorce. If the MTN deal went through, Mukesh vowed to challenge it in court. Anil announced that, if this were true, he would sue his brother for a petrochemicals deal with Chevron. But it was too late. The MTN deal collapsed.

A proposal for an Anil-owned power plant in Dadri, Uttar Pradesh, designed to help end nearby Delhi’s power cuts, was put on hold until a gas dispute between the two brothers was settled. Oxford Analytica, a major United Kingdom-based consulting firm, warned that their fighting would take a toll on the country’s energy sector. Capital markets boomed and swooned with each new development. Pranab Mukherjee, then India’s finance minister, said it was a matter of “national interest” that the two calm down.

Relief finally came in the form of a 2010 Supreme Court decision, which resolved an especially testy gas-pricing dispute between the brothers in favour of Mukesh. Two weeks later, on 24 May, the Ambanis announced that they were effectively scrapping their non-compete agreement. Immediately, the media began to speculate that Mukesh was planning to re-enter the telecom field.

IN 1856, WILLIAM O’ SHAUGHNESSY, an employee of the British East India Company, completed a 6,400-kilometre telegraph system that connected Calcutta, Agra, Bombay, Peshawar and Madras. The system, technologically innovative for its era, had the main effect of strengthening the imperial state. During the First War of Independence in 1857, telegrams quickly alerted the British government about the growing rebellion, helping the Raj stamp it out. On his way to the gallows, one defeated Indian fighter pointed at a telegraph line and declared, “There is the accursed string that strangles us.”

Naturally, Britain continued to expand the network. By 1883, the Raj had runners stationed at telegraph offices, who would travel back and forth between the country’s post offices, allowing British officials to better keep tabs on what was happening across the colony. In the early 1940s, activists in the Quit India Movement targeted the system by cutting telegraph and telephone wires. Apparently, this was to great effect. In 1942, the British secretary of state emphasised the need to protect the government’s telecom property while justifying the arrest of Mohandas Gandhi to the press. Speaking at the height of the Second World War, the secretary warned that Congress leaders were planning a litany of disruptive and destructive activities, including “the cutting of telegraph and telephone wires.” This, he continued, “would paralyse not only the ordinary civil administration of India, but her whole war effort.”

When India finally became independent, the government promptly nationalised the country’s telecom assets. In the Industrial Policy Resolution of 1956, the parliament defined “telephones and telephone cables, telegraph and wireless apparatus (excluding radio receiving sets)” as an industry in which future development was exclusively the responsibility of the state. In justifying this decision, the legislature cited a constitutional directive to ensure “that the ownership and control of the material resources of the community are so distributed as best to subserve the common good.” It added that keeping industries such as telecommunications public was essential “to prevent private monopolies and concentration of economic power in different fields in the hands of small numbers of individuals.”

At the time, talk of the telecom industry was mostly speculative. In 1947, India had roughly eighty-four thousand telephone lines for its population of 350 million. But the country’s teledensity—the number of telephone connections per hundred individuals within an area—barely moved over the next four decades, staying below one percent. To try and fix this, in 1986, the government created two public-sector telephone companies, Mahanagar Telephone Nigam Limited and Videsh Sanchar Nigam Limited. The former oversaw basic telephone service in Delhi and Mumbai. The latter—led by BK Syngal—would bring internet into the country.

Still, the country’s teledensity remained abysmally low, the waiting list for phones grew long and the government became acutely aware of the chasm between what people wanted and what it could provide. In 1992, the government allowed private firms to bid for licences across the four metropolises in the telecom sector for the first time.

It soon became clear that there was a serious lack of regulations with regard to licensing. This led to a watershed national telecom policy in 1994, in which the state estimated that it would need well over Rs 23,000 crores, or $7.3 billion, to meet consumer demand. “Clearly this is beyond the capacity of Government funding and internal generation of resources,” officials wrote. “Private investment and association of the private sector would be needed in a big way to bridge the resource gap.” The state ended its monopoly, and private companies began to buy licences and flood the industry.

The department of telecommunications laid down a list of criteria for private players. The country was divided into circles, or zones of operation, in which companies could apply for licences separately. In order to independently regulate the new entrants, the TRAI was set up in 1997. It was to fix tariffs, previously overseen by the central government, and maintain a healthy competitive environment.

The reforms were a success. As the number of telecom companies exploded, so did teledensity. In 1994, right before the government liberalised the telecom sector, only eight out of every thousand people possessed a phoneline. By March 2007, over a hundred and eighty out of thousand people did. India then had 12 companies that provided cellular services, a number that soon increased.

Since liberalisation, however, the industry has also been marred by controversies. One of the earliest of these involved Reliance Infocomm, when it belonged in the undivided Ambani empire. When the company was founded, in the late 1990s, it was not licensed to provide cellular coverage. In 2001, however, the telecom department ruled that basic licence holders could offer cellular services, what it called “limited mobility,” to subscribers who were making calls from within the boundaries of their own city or town.

Reliance was one of the first companies to avail itself of this opportunity. Another was Himachal Futuristic Communications Limited, a mid-size telecommunications technology company promoted by one Mahendra Nahata. In 1995, HFCL gained notoriety by bidding many times its worth to win some of India’s first private telecom licences. When it became clear that HFCL was unable to pay what it now owed, the telecommunications minister, Sukh Ram, changed the auction rules so that Nahata would not be financially penalised. In 1996, the CBI raided Ram’s home, where it found Rs 3.6 crore—over one million dollars—in cash, and a diary that appeared to contain Nahata’s name. Six days later, the CBI also raided Nahata’s residence, where it found nothing. Ram was eventually sentenced to jail on a variety of corruption-related charges. Nahata was never charged with wrongdoing, but his business soon found itself ensnared in the controversy surrounding the trader Ketan Parekh, who was convicted of artificially inflating the stocks of companies from 1999 to 2001. HFCL was one of Parekh’s clients—and its stocks were among those he allegedly drove up.

Reliance took a very generous interpretation of what limited mobility entailed. The company gave subscribers multiple telephone numbers, allowing them to use their phones as they travelled around the country—not just within the locality where they subscribed. This appeared to contradict the terms of the limited-mobility licence, which stipulated that subscribers’ access to wireless services had to be “limited within the local area i.e. Short Distance Charging Area (SDCA) in which the subscriber is registered.”

Companies authorised to provide full cellular services, which paid roughly four times more for their licences than Reliance had for its, sued the government, arguing that limited mobility was just regular cell service in disguise. But it was not just the roaming that angered the industry. Typically, when one company’s customers call subscribers of a different company, the first company is required to share a part of the revenue it generated from the call with the second. These fees, known as interconnect usage charges, are designed to make sure that providers are compensated whenever their infrastructure is used. But because Reliance held only a fixed-line licence, according to prevailing regulations, its IUCs were substantially lower than those of most of its opponents.

Syngal said that fighting over limited mobility was what finally led him to leave Reliance. During his tenure as Reliance Infocomm chairperson, Syngal told me he was also serving as the president of the Association of Unified Telecom Service Providers of India—one of the telecom industry’s trade associations. He said that Reliance asked him to use his AUSPI position to try and get the government to allow the company to stretch out what it could do with the limited-mobility spectrum. “I was getting calls after calls to say, if [officials] are not agreeing, you as the chairman or president of the association, just write a letter to say this is what should be done,” Syngal said. But he felt that most AUSPI members would reject limited mobility because it would undermine the value of their own licences. “I said, ‘Sorry, I’m not going to write a letter because tomorrow there will be five letters contradicting my letter, so what happens to my reputation?’”

His response, apparently, generated little sympathy from Mukesh’s team. “The retort from Manoj Modi was, ‘You are a Reliance employee, so you should be doing what Reliance wants to do,’” Syngal told me. “I said, ‘Sorry, tough luck, I’m not doing it.’” The company divested Syngal of serious responsibility, and in 2001, he resigned. He said the separation was amicable. “I said, ‘Listen, if I have to just sit and come here to my office and collect my salary and do nothing, I’m not prepared to do this, because age is with me,’” Syngal said. “Maybe I was wrong, maybe I was right. I have no regrets because for me, my reputation is supreme.”

In 2003, the government—aware that Reliance Infocomm was now a de facto cellular provider—introduced a new, unified-access licence. Reliance Infocomm received one less than a day after it became available. The company paid Rs 1,542.5 crore, or $340 million, to upgrade to the new licence, of which Rs 526 crore, or $116 million, was a penalty for its overuse of limited-access mobility.

Pradip Baijal, who was the chairman of the TRAI when the agency introduced the unified-access licence, was later accused of accepting bribes or favours in kind. In 2005, Outlook released snippets of an alleged email from Manoj Modi, in which Modi said that he had “personally spoken to the regulator PB (Pradip Baijal) today and have convinced him of our intention. We are also ensuring that he’s taken good care of.” Baijal called the email a fraud. Prashant Bhushan, a lawyer and activist who has sued the government multiple times over its telecom controversies, argued that the conglomerate wielded immense influence in the corridors of political power.

“If the bureaucracy doesn’t tow their line, they make sure that that person gets transferred or victimised by the government,” Bhushan told me. “They have the political government in their pocket, and, using their leverage over the political government, they ensure that the bureaucracy is also bending to them.”

In 2002, a reporter traced roughly ten million Reliance Infocomm shares to an accountant who had helped the politician Pramod Mahajan escape a financial predicament. Mahajan, who became telecom minister right after the department of telecommunications introduced limited mobility and left right before the government granted Reliance a unified licence, denied any wrongdoing. Mahajan’s successor, Arun Shourie, never faced allegations of impropriety, but was open about his affection for Reliance. At a 2003 memorial meeting for Dhirubhai Ambani, Shourie gave a speech in which he effusively praised the corporation. By “exceeding the limits in which restrictions sought to impound them,” Shourie said, companies like Reliance “helped create the case” for scrapping regulations.

But exceeding limits ultimately did little to help what became Anil’s Reliance Communications, which went from being the market leader to an also-ran within a decade. Part of the company’s failure owed to the onslaught of competition—including, eventually, from Jio. A large part of the blame rests with Anil Ambani, who, as Syngal put it, could not “make a meal” out of the expansive network he acquired. But some of it also rests with Mukesh, who decided in the early 2000s to use a kind of broadband technology that, by the late 2000s, had fallen out of favour in the industry. Reliance Communications was then forced to spend large sums of money switching to a different system.

INDIA’S BROADBAND WIRELESS-ACCESS SPECTRUM auctions began on 24 May 2010, the same day Anil and Mukesh scrapped their non-compete agreement. The department of telecommunications had received applications from 11 bidders. That month, most of India’s telecom energy had been centred on the country’s 3G auction, which had offered spectrum for voice, text and data, and which had concluded five days before. But the BWA auction—the first of its kind—also attracted the telecom industry’s heavyweights, such as Airtel and Vodafone.

It also attracted a curious upstart, a little-known company called Infotel Broadband Services Private Limited. It was earning only Rs 14.78 lakh, or $32,743, and had reported Rs 18 lakh, or $35,329, in cash reserves in its most recent annual financial statement. At the beginning of March 2010, it was worth just Rs 2.4 crore, or $5.215 million.

That changed on 19 March, when IBSPL received a bank guarantee of Rs 252.50 crore, or $55.5 million, from Axis Bank. Years later, this guarantee would attract the attention of the comptroller and auditor general, who concluded in a draft report that IBSPL managed to get it without providing a security deposit. In 2014, images of the bank guarantee itself were leaked. The document was entirely typed, except for the lines to the right of a field titled “Name and Address of the Applicant.” Whatever text was there appeared to have been erased, and the name and address for Infotel Broadband Services was handwritten in its place.

The guarantee, along with a variety of other documents, was sent to JS Deepak, then the joint secretary for telecommunications, who was preparing the spectrum auctions. While the bank guarantee did not reveal anything about IBSPL’s background, one of the documents did. According to the company’s ownership details, IBSPL’s director was the son of Mahendra Nahata, of HFCL.

Deepak’s department assigned Infotel’s bid the amount of eligibility points consummate with a Rs 252.5-crore loan—350, the maximum. With the exception of Qualcomm—a US-based, Fortune 100 technology giant—every other company that received 350 eligibility points was a major existing Indian telecom company with at least 35 million wireless subscribers. Infotel, by contrast, had just one internet subscriber.

IBSPL placed the largest first-round bid in each of India’s 22 telecom circles for the internet-only licence. As Infotel’s competitors raised their bids, IBSPL kept raising its bids too, pushing more and more of its chips onto the table. By the twelfth round, a tenth of the way into the auction process, IBSPL had offered over Rs 3,000 crore for pan-Indian spectrum rights. By round 60, that figure had increased to roughly Rs 8,000 crore. In round after round and on day after day, India’s 150th-largest internet provider continued to push broadband prices around the country higher and higher, towards the upper limits of what even the biggest telecommunications companies could tolerate.

On 31 May, in the middle of the auction, the Economic Times ran an article speculating that Mukesh was searching for a way to re-enter the telecom industry. The article cited Idea Cellular, which had 65.3 million subscribers, as a possible acquisition target. The Birla family, Idea Cellular’s owners, had recently sold some of it to a Malaysian company. Perhaps they were interested in giving away more. One analyst was quoted saying that the Videocon-promoted Datacom, another existing provider, might be ripe for a takeover. The story did not mention IBSPL, nor did it discuss the auction itself, where prices were still going up. At round 100, Infotel’s total bid had soared past Rs 11,000 crore, or over two billion dollars. At round 110, they had topped Rs 12,000 crore.

Then, on 10 June 2010, the Economic Times dropped a bombshell: Reliance Industries was planning to acquire the broadband-spectrum auction’s eventual winner. The next day, the auction came to a close. IBSPL was, in a sense, the victor—the only company to win broadband spectrum in each of India’s 22 telecom circles. But a few hours later, Infotel agreed to sell 95 percent of its shares to RIL. On 22 January 2013, Reliance renamed the company Reliance Jio Infocomm Limited.

“The speed with which that company changed hands, it became clear that they were fronting,” said a veteran executive at a different telecom company, who spoke on the condition of anonymity out of fear of being sued by Reliance Industries, which is notorious for taking legal action against its critics. The comptroller and auditor general’s draft report came to a similar conclusion, and noted that RIL’s 2010–11 annual report showed significant transactions with Axis Bank. It showed an investment of Rs 2,250 crore, or $503.9 million. All of these went against the “sanctity and transparency of the auction process,” the report noted.

Multiple telecom executives told me that Infotel’s fronting helped lower the spectrum’s final price. “We thought it was an ineffective, useless, underfunded company like HFCL,” the veteran executive told me. “That itself lulls you into what possibly these guys can do with their spectrum.” Companies bid less, which, in turn, reduced the amount of money that the government collected. “They won it at a price where they were disguised as a sheep,” the veteran executive said. Had his company known in advance what would ultimately become of the spectrum, “of course we would have bid much more.”

To try and guarantee that winning spectrum bids reflect the spectrum’s actual value, the Indian government requires that auction participants disclose information about themselves in advance. In 2010, each bidder had to provide the government with a document “certifying details of Promoters/Partners/Shareholders in the company.” This did not mean that bidders had to already be telecom players. The government’s “Notice Inviting Applications” allowed companies that did not have licences to formally bid in conjunction with a company that did, such as IBSPL. They just had to disclose it.

But Infotel did not make any such disclosure in its application form, and Reliance did not avail itself of the auction’s partnership provision, even though Nahata told reporters on 11 June that he had been “in talks with RIL even before the auction for BWA started.” The application notice states that the government could disqualify any bidder that “provides inaccurate information in its Application to participate in the Auction.”

Mahendra Nahata went on to become one of Reliance Jio’s directors. As of 2017, his son owned 65,753,000 shares of the company—15 percent of the total shares. In 2012, Reliance announced that HFCL would help construct its 4G network, leading the latter company’s stock prices to skyrocket. In a section of the government’s notice inviting applicants to the 2010 broadband auction, entitled “Anti-Competitive Activity,” the department of telecommunications stipulated that spectrum bidders “must not enter into any arrangements with suppliers of equipment or software that would restrict the supplier’s ability to supply such equipment or software to other Bidders for the purposes of planning, building or operating a network.” The terms and conditions of Reliance’s agreement with HFCL are unclear.

I sent a detailed questionnaire to Nahata and HFCL, asking about the controversies surrounding the auction, but received no reply. I spoke to Deepak on the phone, but he declined to comment.

Some of the analysts I interviewed argued that what happened in 2010 is as much the responsibility of Reliance’s competitors as it is the responsibility of Mukesh Ambani. “The incumbents got caught on the wrong foot because they failed to anticipate this,” said Mahesh Uppal, a director at the telecom consultancy ComFirst. The companies, after all, knew that IBSPL had 350 eligibility points.

I asked the veteran executive why the industry had not taken IBSPL more seriously. He pointed to the recently concluded 3G auction, which was far more expensive and important for providers. Indian telecom companies, after all, were still working to lay out their 3G networks. “We were fighting to either save our company or sustain our company, because if we didn’t win the auction in certain circles, if we didn’t win spectrum, than the growth path for our consumers is gone, and that leads to the atrophy of the business,” he said, referring to the 3G auction. “You blank out other things that are happening.”

Infotel, by contrast, participated only in the country’s broadband wireless auction, which offered spectrum designed for internet service. The company was not even licensed to provide cellular service. “It was obviously not HFCL’s balance sheet,” the veteran executive said. “But people did not bother because if you want to do something in internet services, go ahead, that’s not my business.”

But in 2011, Infotel, now part of Mukesh’s RIL, applied for a mobile country-code and a network code—initial steps towards becoming a full-fledged telecom provider. In April 2012, the TRAI recommended letting companies with internet-only licences upgrade to unified-access licences. In February 2013, the department of telecommunications began allowing companies to do so. RIL was the first provider to take advantage of the department’s decision, paying a Rs 1,658-crore migration fee and a Rs 15-crore entry fee.

The decision invited a cascade of criticism. The Cellular Operators Association of India, which represents telecom companies—including, now, Jio—noted that Reliance had effectively cut the line and become the first Indian telecommunications firm with access to 4G cellular technology. The government’s decision raised further questions about the 2010 auction. “If others had also known that this spectrum will also be used in addition to ISP services for full- blown voice and data services, maybe others could have bid. And maybe the price of the spectrum could have been higher,” Syngal told me. “This is clearly a favour. This is clearly a backdoor entry.”

It was also, for Reliance, a money saver. The entry fee for the internet-only auction was substantially lower than the fee for the unified-services auction, and the government charged Jio a comparatively modest amount to upgrade. As the comptroller and auditor general noted, the migration fee Jio paid was based on the price of a unified licence in 2003, which was, ironically, the same year that the Mukesh-led Reliance Infocomm upgraded to a unified licence (also with controversy). During the intervening years, the value of such a licence had increased substantially. The CAG initially estimated that the value of the spectrum had increased by Rs 20,653 crore, or $4.57 billion. Reliance, then, had managed to upgrade at less than one-thirteenth the “actual” cost of the spectrum it could now use for everything from data to phone calls. The national exchequer lost over Rs 20,000 crore, or approximately $4 billion, in the process.

I messaged Kapil Sibal, the minister of communications and information technology from 2011 to 2014, about the controversy surrounding the licence upgrade. He did not respond and neither did the TRAI.

The CAG’s final report, however, reduced the estimated loss to Rs 3,367 crore, or $525.7 million. In 2016, the journalists Paranjoy Guha Thakurta and Aditi Roy Ghatak reported that a source within the CAG’s office said two senior government bureaucrats had asked to “tone down” the report.

The deputy CAG, Suman Saxena, offered a different explanation when reporters asked about the large discrepancy between its two estimates.

“A draft is a draft,” she said.

ACCORDING TO THE AUCTION TERMS, the winning broadband bidder had five years to roll out service in almost all metropolitan areas and at least half of rural areas. RIL told reporters that it would launch services within two years of acquiring Infotel. In a June 2012 speech, Mukesh Ambani said that his company was “finalising our plans to offer services on a nation-wide basis.”

RIL, however, did not launch telecom services in 2012. Nor did it launch in the next two years. In 2015, amid intense speculation about when Jio was going to kick into action, Ambani told shareholders that Jio would begin commercial operations around December. It did not.

In the interim, Airtel and Aircel started using the broadband spectrum they had managed to win—alongside IBSPL—to offer fourth-generation cellular service in a few states. But their initial 4G rollout was limited to regions where they had broadband licences. Jio’s advantage was that it was the only company that had won broadband rights across India in 2010. It took the company until 5 September 2016—over six years—to launch commercial services. To try and ensure timely access, the notice inviting auction applications had stated: “If the licensee does not achieve its roll out obligations, its spectrum assignment shall be withdrawn.”

According to many officials I spoke to, a certain amount of delay in rolling out broadband spectrum was to be expected. “This is new technology,” Rahul Khullar, the chairman of the TRAI from 2012 to 2015, told me. “Some extension of time was necessary. But it could not be open-ended.”

When it did finally go live, Jio quickly accumulated subscribers. In October 2016, the company boasted that it had mopped up 16 million customers in one month. “Jio is built to empower every Indian with the power of data,” Mukesh Ambani said. “We are delighted that people have recognised this and are utilising our services to the fullest.” Analysts have been impressed. “It is an incredible job that they have done,” Syngal told me, referring to the company’s rapid growth.

But what Ambani did not mention was that the company had had an unusually large head start. Before offering services commercially, most telecommunications firms are allowed to test their network by inviting employees and their associates to subscribe in advance, free of cost. But Jio’s tests extended far beyond the friends and family of company staff. In July 2016, for example, Jio invited anyone who had purchased Lyf smartphones, the Xiaomi Redmi Note 3 or a variety of Samsung smartphones to try out their services. Over a month before formally launching, Jio had 1.5 million people “testing” its network. By its launch date, some estimates put the number as high as 3 million.

Existing operators were furious—and not just at the number of testers. Typically, telecom companies are allowed to test their networks for 90 days before launching services. Jio’s test phase lasted for over eight months. In a letter to the telecom department, Rajan Mathews of the Cellular Operators Association of India wrote that Reliance’s offers were “full-blown and full-fledged services masquerading as tests, which bypass regulations and can potentially game policy features.” Mathews pointed out that because it had yet to formally launch, Jio did not have to pay other operators IUCs. In addition, Jio was not yet subject to TRAI regulations regarding anti-competitive and predatory pricing behaviour, allowing it to enrol hundreds of thousands of new customers for free, with little interrogation of its practices. Jio responded that the tests were necessary for a network that soon planned on carrying traffic from tens of millions of subscribers.

In response to the controversy, the TRAI wrote to the department of telecommunications looking for guidance as to how long testing phases should last. The department, in response, sent the question back to the TRAI, and asked that the regulator issue recommendations. Eventually, the TRAI stipulated that “there should be a limit of 90 days” for trials. Extensions would be permitted on a case-by-case basis. But the back-and-forth delayed TRAI’s findings until May 2017—eight months after Reliance Jio formally launched its services.

Meanwhile, Jio had continued to grow. On 27 February 2017, Cisco Systems, an international technology manufacturer and one of Jio’s business partners, announced that Jio had 100 million customers using its data network. “Jio network has exceeded consumption of more than 1 exabyte of data per month, establishing its clear leadership as the dominant network,” they wrote. Two-and-a-half weeks later, the TRAI invited stakeholders in the telecommunications industry to write in with advice on how to restructure the agency’s regulatory policies. The “shift, from voice to data, driven by technological and other factors,” the TRAI explained, meant that it was time to reconsider regulations.

The process lasted nearly a year, and included a lengthy commenting period, a seminar conducted by industry experts and meetings with the chief executives of telecom companies. Finally, on 16 February 2018, the TRAI announced that it had changed the definition of what constituted “significant market power,” a designation that subjects companies that fit it to increased scrutiny and heightened regulation. Previously, a telecom provider had significant market power if it held “a share of 30% of total activity in a licensed telecommunication service area.” That included the volume of data traffic.

But now, the TRAI wrote, “SMP would be based on Subscriber base and Gross Revenue only as these two parameters are relevant and comparable across all the technologies.” Reliance Jio, by then, carried more data on its network than its competitors, but according to February’s TRAI data and media reports, it did not possess more than 30 percent of the customer market share or gross revenue share in any of the country’s telecom regions. Airtel and the then-merging Vodafone-Idea, by contrast, met the SMP threshold in at least ten circles apiece. Jio was now much less likely to be regulated than its competitors.

Jio’s competitors sued. In December 2018, the Telecom Disputes Settlement and Appellate Tribunal ruled against the regulator, and asked it to draw up new rules within six months. The regulator has indicated that it plans on appealing the tribunal’s decision before the Supreme Court.

Jio supporters argue that it might be a stretch to assert that regulations are being designed to benefit the company. Why can the company not, either through chance or its own volition, just be the beneficiary of TRAI rules that are, as Anil Kaushal put it, designed to make sure that consumers “get the appropriate, right price and good quality?”

I put this to former TRAI chairperson Rahul Khullar. He rejected the idea that the seemingly pro-Jio bias of the government’s decisions was just a series of lucky coincidences. “No one can be lucky continuously, not even Jio,” Khullar argued. “And to get regulatory orders that are subsequently struck down by courts, that tells you that it is more than blind luck.”

One official I spoke with pointed to Ravi Shankar Prasad’s history as evidence of government bias. Before assuming his current charge as minister for electronics and information technology, Prasad served as the minister of communications. In December 2014, the Financial Times and Outlook reported that Prasad was being retained as a lawyer and paid by a Reliance subsidiary from April 2013 to March 2014. In a statement, the then-communications minister said that his legal work as a minister “has been purely in a professional capacity.” His sister also has close ties with Reliance Jio: Anurradha Prasad (who is also married to prominent Indian National Congress politician Rajeev Shukla) is the owner of News24 and E24. In March 2014, both channels received a loan of Rs 12.5 crore, or roughly $2 million, from a company called Eminent Networks—loans that could have been converted into equity. The Jio board member Mahendra Nahata owns Eminent Networks.

Several of the people I spoke with also cited the fate of JS Deepak—the same official who oversaw the 2010 spectrum auctions—as evidence that the government is thinking about more than just consumers when it makes telecom decisions. Five-and-a-half years after the auction, Deepak was appointed the secretary of the telecom department, putting him in charge of India’s telecom commission. The Deepak-led commission expressed discomfort with the TRAI’s treatment of RIL. It criticised the regulator for allowing Jio to test its network so extensively and rebuked the regulator for trying to fine Jio’s competitors Rs 3,050 crore, or $455.9 million. In February 2017, Deepak wrote that the regulator’s tolerance of Jio’s low tariffs had cost the government Rs 685 crore, or $102.3 million, in revenues. This, Deepak said, was an “alarming decline.”

One week later, Deepak received an order transferring him out of his position as telecom secretary. It took effect immediately, three months prior to when he was scheduled to depart. Deepak was representing India at the Mobile World Conference in Barcelona when the command came.

I asked Syngal whether he thought Deepak’s transfer was the result of his criticisms. “It is very, very true,” he told me. Syngal said it was not the first time a bureaucrat had been removed in such a fashion. “I can tell you another one who was shunted out overnight,” Syngal told me. “Dr JS Sarma from the department of telecom was shunted out because he didn’t listen. He was told to screw Tatas by Maran.”

JS Sarma was the telecom secretary from June 2005 to July 2006, when Dayanidhi Maran was the telecom department’s minister. According to Syngal, Maran disliked one of Tata Communication’s main advisors, and he asked Sarma to help penalise the company. Syngal was meeting Sarma when Maran called. “I’m literally, literally witness to his treatment by the minister, because the minister wanted to screw Tatas, and [Sarma] said ‘Minister, you can do anything to me, but what you are suggesting is absolutely unfair that I screw Tatas. I will go by rulebook.’ And he was shunted out.”

I sent messages to Maran to ask about Sarma’s transfer. Maran did not respond.

“Something similar happened because Deepak didn’t want to tow the line to help Reliance,” Syngal said. When I spoke to Deepak by phone in January, I asked him why he had been transferred. He declined to comment.

THE CURRENT CHAIRMAN OF THE TRAI, and the target of the telecom commission’s criticisms, is Ram Sewak Sharma. A member of the Indian Administrative Service class of 1978, Sharma is perhaps most famous to the public for publishing his Aadhaar identification number on Twitter to prove his confidence in the system’s security. Within hours, social media users claimed to have accessed his phone number, address, and a variety of other personal information. Sharma responded that the information people obtained was already in the public domain.

Recently, critics of Sharma have focussed on the TRAI’s decision to gradually eliminate interconnect usage charges. Until October 2017, a telecom company had to pay a competitor 14 paise for every minute that their customers called someone who subscribed to the competitor’s network. But Jio protested that these payments unfairly advantaged its opponents, all of whom had a larger customer base and a more developed network and, as a result, carried more incoming telephone traffic than Jio did. Incumbent operators, including Airtel and Vodafone, countered that the fees were necessary and fair, given the costs incurred to maintain their infrastructure.

The TRAI sided with Jio. In a report issued on 19 September, the agency wrote that keeping IUCs would “be detrimental to the growth of telecommunication services sector.” It immediately moved to lower IUCs from 14 paise per minute to 6 paise per minute. The agency decided that it would eliminate the fees entirely by 2020.

In the abstract, slashing IUCs can be good for innovation and competition. It can incentivise companies to build new lines in new places, rather than doubling down on existing ones. It makes it easier for smaller firms to use the infrastructure of titans without being smothered out of existence by untenably high fees. But there are effectively no small telecom providers left in India, and analysts say that the company with the most to gain from reduced IUCs is Jio. A report by AllianceBernstein, a New York-based global asset-management firm, estimated that Jio received a net benefit of Rs 640 crore from the TRAI’s decision to slash the fees.

The decision earned plaudits from the former TRAI chief Pradip Baijal, who praised RS Sharma’s “courage” in reducing IUCs during an interview with the Economic Times. Sharma’s record has also pleased his political superiors. One day before he was scheduled to step down, the government extended Sharma’s term by two more years—making him the first ever TRAI chairman to be reappointed to the post. His new retirement date is 30 September 2020. IUCs are scheduled to go away just nine months before then.

“Everyone knows what this chairman is doing, and why,” Khullar, Sharma’s immediate predecessor, told me.

One former senior government telecommunications official minced no words when I asked for the rationale behind Sharma’s decisions. “I’ll give you four options. One, Reliance is promising a postdated payback. Two, the prime minister’s office or the prime minister is giving a hint, help Mr Ambani. Three, any combination of the two. Four, favour Mr Ambani and we will reward you in the future. We will reward you now with an extension. We will reward you with something even more important later.”

I reached out to RS Sharma and TRAI with a detailed set of questions, including allegations about direct collusion with the prime minister’s office and the Ambanis. He did not respond.

Sharma’s defenders praise his “tech-savvy” credentials—the chairman holds a masters degree in computer science from the University of California, Riverside—and his contribution towards lowering data prices. “He focussed on the needs of the consumer,” Kaushal said. His detractors are not convinced. “He is aligned to one company, and sometimes he makes a blatant show of it,” Syngal told me. “He has harmed the industry; he has harmed the competition; he has also harmed the revenues of the government.”

The controversies, however, spill beyond just the TRAI and the telecom department. They extended beyond even the central government. Rajasthan’s recent free phone scheme is the foremost example. According to a recent report in the Caravan, the principal secretary of Rajasthan’s information technology and planning department instructed district collectors to direct eligible residents to Reliance Jio for purchasing both plans and phones. The government even allegedly provided Jio with lists of NFSA beneficiaries.

The plan was controversial independent of Reliance. Rajasthan’s government introduced the scheme close to the recent state elections, and the Congress accused the ruling BJP of trying to bribe voters. Vasundhara Raje, then the chief minister of Rajasthan, framed her plan using the language of Digital India. State residents, she said, “will be able to see and access government schemes at the click of a button”— echoing one of Digital India’s key aims.

I sent queries about the programme to Rajasthan’s department of information technology and communication. I did not receive a reply.

“It’s not only the regulator,” Syngal told me. “It’s the entire setup. Everybody is hand in glove.”

WHEN RAJAN MATHEWS asked the department of telecommunications to end Reliance Jio’s testing phase, the company immediately mounted a counterattack. In a letter to the department secretary, Jio accused the Cellular Operators Association of India of levelling “malicious, unfounded, ill-informed” allegations and “promoting vested interests of the incumbent operators.” The COAI, it wrote, “has deliberately indulged in an unwarranted vilification campaign, not only against Reliance Jio Infocomm Ltd, but also against TRAI.”

Shortly after sending its reply, Jio filed a formal complaint with the TRAI, alleging that Airtel, Vodafone and Idea are a cartel that—in fighting to stop Jio from testing how its network connects with theirs—had engaged in anti-competitive behaviour and should be punished. The TRAI agreed, and asked the telecom department to fine the three players a combined Rs 3,050 crore, or $575 million. The department has yet to make a final decision on the matter.

Jio’s TRAI complaint was only the first in a series of legal manoeuvres that RIL has used against critics. In March 2017, Jio filed a complaint with the Advertising Standards Council of India, or ASCI, alleging that Bharti Airtel misled the public by claiming that it was “officially India’s fastest network.” At issue was whether the application that Airtel used to justify the claim, named Speedtest, was legitimate enough for Airtel to call its results “official.” Because Speedtest was not recognised by the TRAI or the telecom department, Jio contended that it was not.

But RIL went even further, targeting the application’s developer itself. “We are aware of the material flaw in Ookla,” a Jio spokesperson said, referring to the company behind Speedtest. “In spite of acknowledging this flaw, we are surprised that Ookla has gone ahead and released misleading results. We have initiated suitable actions at appropriate forums.”

The ASCI asked Airtel to withdraw the ads. But Airtel began airing Ookla’s data again in June, prompting Jio to sue both its competitor and the application. In a Mumbai magisterial court, Reliance alleged that Airtel and Speedtest had caused the company monetary and reputational harm to such an extent that it amounted to conspiracy, defamation and breach of trust.

The presiding judge, KG Paldewar, disagreed. “It is nothing but a marketing policy,” he wrote, and dismissed the case.

RIL has also singled out the COAI. In February 2018, the company sent a defamation notice to the trade body and its director-general, Mathews. It demanded that the group publicly apologise for claiming that the TRAI favoured Jio over other operators. The COAI refused, stating that it was “well within its rights” to object to TRAI policies and that it harboured no “malicious intentions against RJIL.”

When I spoke to Mathews in April 2018, I asked him if he was concerned about being sued. “Not particularly,” he said. In an earlier conversation with me, Mathews had reiterated the COAI’s belief that “the positions made by the regulator over the last couple of critical division points seem to have favoured one particular operator.”

In late May 2018, Reliance sued the COAI—an organisation to which it belongs. The Delhi High Court accepted the case and told both the COAI and Mathews to stop using “disparaging and defamatory” words about Jio. Mathews said that it was “disappointing that one of our members has chosen to take legal action against the association,” but that the allegations were “without merit.” The organisation is considering its legal options.

Reliance’s COAI lawsuit does not mean that the trade body is without bias. By nature, the COAI’s membership is mostly comprised of Reliance’s competitors. And there is little reason to doubt that, in their statements and actions, Airtel and Vodafone–Idea are acting out of financial self-interest. They are large, publicly traded, for-profit companies.

But RIL dwarfs both of them. Its revenue is three times that of Airtel and four times that of Vodafone–Idea. In 2017, the company made more than twice as much as the combined profits of its two competitors. “Jio is not a small new entrant in the market,” Mahesh Uppal told me. “Jio is a humongously large company with a huge balance sheet.” That balance sheet, analysts say, is helping the company sustain its devastating price war.

Reliance, however, says that its telecom subsidiary is in fact turning a sizeable profit. The company, for example, reported Rs 681 crore, or approximately $100 million, in profit for the second quarter of 2018. In a press release, Mukesh Ambani praised Jio’s financial performance and said that it reflected “the benefits of scale, technology and operational efficiencies.”

But the asset-management firm AllianceBernstein says it also reflects something else. Describing Jio’s reported profits as “a bit too good to believe,” the firm suggested that the company was taking “a unique approach” in measuring the costs of its assets, one that “results in a significantly lower D&A”—depreciation and amortisation—“expenses than seen elsewhere in the Industry.”

Depreciation and amortisation are a common accounting technique in which companies distribute the cost of an asset over the asset’s usable lifespan, rather than all at once. For example, imagine a cellular company purchases a cellular tower for Rs 20 crore in 2018. It may opt to incorporate the entire cost of the tower in its 2018 financial report. But, assuming that the tower has a 20-year lifespan, the company might instead spread, or “depreciate,” the cost of the tower over twenty years—Rs 1 crore per year. The company’s reported 2018 profit will, therefore, appear much higher than if the company had included the entire Rs 20 crore in one year. Amortisation refers to the same practice, except for “intangible” assets.

AllianceBernstein reported that Reliance’s own depreciation procedures differed greatly from that of its peers. “Given their networks are broadly similar,” it wrote, “we would generally expect to see D&A as a percentage of fixed assets and Intangibles (spectrum) also be similar.” But they are not. Jio’s depreciation is 3.3 percent of the cost of its fixed assets. The average for major international providers, including Verizon, China Telecom, and AT&T, is 8.6 percent. This would mean that Jio’s physical infrastructure—in theory, at least somewhat similar to that of other big players—has a lifespan that is more than twice as long.

There are other irregularities. Jio, for example, chose not to release financial data for the second quarter of 2017. Instead, it included its April-through-June earnings in its July-through-September report, effectively packing six months of sales into the year’s third quarter. In an article, Bloomberg estimated that the decision might have helped Jio avoid reporting $1 billion—Rs 6,621 crore—in losses.

Clever accounting is nothing new for RIL. In 1987, the company was in a tight financial position, with diminished cash flows and delayed production plants. The government rushed in to help Dhirubhai, announcing a string of import-regime changes that made it easier for Reliance to acquire raw materials, but it was not enough to save the company from a loss that year. RIL announced that instead of repeating its finances over the calendar year, as it had been doing, it would shift to reporting from each April to the following March, as the government does. This extended its 1987 financial year from 12 to 15 months. When that still was not enough to prevent reporting a loss, Reliance extended its year by another three months. Throughout the process, Reliance depreciated the costs of two new laboratories in a novel way—returning roughly Rs 24.5 crore, approximately $18 million at the time, to the company’s checkbook. At the end of the 18 months, it reported a profit.

JIO HAS PROMPTED GROWTH in India’s data consumption. Internet access in India has become more affordable since the company launched, and, consequently, it has become more widespread. “There have been immense benefits,” Uppal told me. “Thanks to Jio, who just reduced the price of data by a huge amount, certainly we have seen a huge spurt in the amount of data usage.” This kind of analysis was common among Indian telecommunications experts, who seemed—if anything—fatigued by the parade of controversies in the sector. “While some of this seedy stuff has been happening in the background, the positive side is India now has some of the lowest call rates,” Saurabh Mukherjea, the former Ambit chief executive, told me. “Against all odds and with all sorts of fun and games, the free market delivers.”

But, for the government, those games have proved very costly. The cumulative loss incurred as a result of every telecom scandal between 1994 and today, according to finalised CAG reports, is over Rs 200,000 crore, or $28 billion.

Scandal, therefore, is not unique to Reliance. A 2015 draft CAG report claimed that the telecom department looked the other way while Bharti Airtel made illicit accounting deductions during the late 2000s. Airtel was also accused of receiving unduly large quantities of spectrum. The CAG, for instance, alleged that from December 1996 to August 2000, Airtel received 1.8 megahertz more from the telecom department than it was permitted to have. The CAG estimated that these, and other, infractions cost the public exchequer at least tens of thousands of crores of rupees. In an article summarising the draft report’s findings, Aditi Roy Ghatak and Paranjoy Guha Thakurta wrote that while Airtel “is one of corporate India’s outstanding successes,” there is reason “to read between the lines and scrutinise the fine print and question if this magnificent growth is built on the foundations of accounting artifice and government indulgence.”

In response to my queries, an Airtel spokesperson said the company “has always been in complete compliance with all applicable regulations.” The dispute over its revenue, he continued, “encompasses the entire industry, as the same has been challenged by all telecom operators in various courts ... where the matters are pending and under a stay.” The spokesperson said that all spectrum allocated to Airtel had been in “strict compliance with the stated government policies.”

The telecom sector’s darkest moment came in 2008. On 10 January, the department of telecommunications abruptly announced that it would issue licences for second-generation spectrum on a first-come, first-serve basis, with little regard to what companies were willing to pay. It then instructed interested businesses to go to the department and submit applications between 3.30 pm and 4.30 pm that day. Some companies arrived with fully completed paperwork. Others scrambled to get their applications in order and rushed to the department, only to be turned away by bouncers at the door. Corruption allegations dogged the department for months, and the CBI eventually arrested the then telecom minister, Andimuthu Raja, and charged him with accepting bribes. Raja, like everyone else arrested during the 2G scam, was eventually acquitted. But by doling out licenses at below-market rate, the CAG estimated that the government lost roughly Rs 176,000 crore, or $24 billion.

During a press conference held in the wake of the 2G scam, the then prime minister, Manmohan Singh, responded to a question about lost revenue by asking the country to think of the cheaper licence costs as a kind of subsidy for a developing industry. “With regard to the loss of revenues, it is very much dependent on what is your starting point” he said. “We give subsidy to fertilisers which cost about Rs 60,000 crores every year. People can say that these fertilisers should be priced at the market rate. Would you then say that there is a loss of revenue of Rs 60,000 crores in fertilisers sale?”

The CAG report identified at least one company that was willing to pay significantly more for a licence. It also noted that most of the winners of 2G spectrum failed miserably at rolling out their services on time. These companies, the auditor wrote, were “hoarding” India’s cellular spectrum, keeping it from operators capable of rolling out service more promptly, leading to an “inordinate delay” in service, depriving Indians of access to technology and further hurting the government’s bottom line.

It is difficult to know what India’s telecom and data market would have looked like if Jio had been charged a higher amount for its unified licence, or if it had been more tightly regulated. But there is evidence that, while it has spurred an upsurge in data use, Jio is overstating its overall social impact. Reliance’s 2016–17 annual report, for example, claimed that “through Jio, employment opportunities were created directly and indirectly for more than 50 lakh people”—separately, the report clarified that this figure was a “current year outcome.” There is little reason to doubt that increased data access helps spur job creation, but there are reasons to doubt Jio’s figures. The human-resource company Team Lease and the Telecom Sector Skill Council estimated that the entirety of India’s telecom sector would create 2 million jobs in 2017, directly and indirectly. For both Jio and Team Lease’s estimates to be true, Jio alone would have had to have created at least 3 million jobs in 2016—a year when its services were not commercially available until late September.

RIL’s statements about the size of its own payroll also raise questions. The company’s chief human-resource officer told reporters that, as of April 2018, Reliance Jio had roughly 157,000 employees on its staff. This is more than the 140,483 people listed as employees in RIL’s 2016–17 annual report, and slightly fewer than the 187,729 staffers listed on the 2017–18 annual report. The latter two statistics are employment figures for the entirety of RIL, including its retail, petrochemicals production, refining, entertainment, and manufacturing segments. By contrast, Bharti Airtel employs 17,263 individuals in South Asia and 20,793 worldwide. Airtel has roughly 100 million more subscribers than Jio.

In the questionnaire I sent Reliance, I asked how the company arrived at these estimates. I received no answer.

In the same April press conference, Jio’s human-resources director told reporters that the company was planning to hire tens of thousands more employees in the months to come. In his speech at the Digital India launch, Ambani estimated that his company’s initiatives under the scheme would “create employment for over five lakh people.” But whatever Jio’s individual impact, the telecom sector as a whole is struggling to hire. Even as millions of Indians have acquired cell phones, India’s telecommunications industry is shedding tens of thousands of jobs in order to save money amid the price war.

At the end of November 2018, unions at the government-owned Bharat Sanchar Nigam Limited threatened to go on strike in part to protest the government’s seemingly preferential treatment for Jio. The union alleged that the government had not given any 4G spectrum to BSNL in order to prevent it from competing against Mukesh Ambani. They expressed fear that their jobs would disappear. “The whole game plan of Reliance Jio is to wipe out its competitors, which includes the state owned BSNL,” the unions wrote in a joint statement. Then, they argued, Reliance “will loot the people by steeply raising the call and data charges.” The union has yet to indicate if it will actually follow through with the strike.

But it is not just about employment. While the growth in data usage after Jio opened up shop accelerated, the number of cellular subscriptions did not. From March to September 2017, for example, rural tele-density and urban tele-density remained effectively stagnant—in other words, the average number of telephone connections per person did not change. Indeed, rural tele-density ticked down.

To be fair, teledensity is far from the best measurement of how many people have internet or cellular access in the modern era. Digitally connected Indians often possess multiple SIM cards, which artificially inflates measurements of how many people have service. Yet for all the articles about the “Jio effect” on India’s digital connectivity, many packed with heart-warming stories of Indians browsing the internet for the first time, there is surprisingly little data on Jio’s broader impact on India’s connectivity.

The data that does exist sends a mixed message. A 2017 report by Mary Meeker, a prominent internet expert and venture capitalist, found that the number of Indian broadband subscribers did go up in the months after Jio debuted. But the most recent TRAI data on active wireless subscriptions—which eliminates SIM cards that are not actively in use—does not indicate any unusual shift. From September 2016, when Jio began offering service, to September 2017, the rate of active wireless subscriptions increased by roughly 69 million. That’s more than the 56-million gain over the same period the prior year, but less than the 81.62-million gain from September 2014 to September 2015. From September 2017 to September 2018, the number of active wireless subscribers fell slightly.

For now, data prices remain low. As of 13 January 2019, Jio offers prepaid plans that cost roughly five rupees per day in exchange for 1.5 gigabytes of data. But the low prices are a product of a kind of fierce competition that virtually no one thinks can go on indefinitely. Jio remains well-funded, thanks to RIL’s other businesses, but it still has over $30 billion in debt. Eventually, the conglomerate will want to see sizeable profits on Reliance’s telecom investments. That means more tariff increases are not a question of if, but when.

The veteran executive argued that the company would continue offering rock-bottom deals for at least a little while—but not for the benefit of the consumer. “They know that Bharti and Idea and Vodafone have got pressure from their shareholders. They’re running losses, the losses are widening every month, every week,” he told me. “This is not the time to give them breathing time and to release pressure and increase price. This is the time to finish off the job. If you were to give companies breathing time, they will recoup, they will improve their profitability, they will recover. And then they will get back at you.”

Even if the industry does stabilise as it is, some experts say that rates will still rise substantially: internationally, areas with fewer telecom providers do typically have higher tariffs. Given the high levels of debt and cutthroat pricing, it is unclear how this could be easily fixed. “We are too far gone down the road now,” Kapil Sibal, the former telecom minister, told me. “You can’t bring back the players.” If India loses one more company, then the situation may get dire. “Let’s say that situation arises for a second,” Syngal said, sketching out a world in which Vodafone–Idea collapses. “What will happen? You are at the mercy of Airtel and Reliance Jio. BSNL is not in competition, and they will be taking the consumer for a ride.”

Syngal shook his head. “You can’t have a situation where you are at the mercy of two or even three private players,” he said. “The time to act or intervene is now.”

ON 13 FEBRUARY 2014, Comcast Corporation, the largest internet provider in the United States, announced that it intended to acquire Time Warner Cable, the country’s second-largest internet company. The chairperson and chief executive officer of Comcast, Brian Roberts, referred to the deal as “pro-consumer, pro-competitive, and strongly in the public interest.” Roberts became the CEO of Comcast in 2001, succeeding his father, Ralph Roberts, who had founded the company in the 1960s. In an interview soon after the deal was announced, the younger Roberts cited the elder Roberts as an inspiration. “Whenever my dad is asked if he ever thought Comcast would grow to this size, he likes to say that he always imagined that was possible,” Brian said. “He jokes, of course, but it was. Only in America.”

Immediately, activists began to push back on the proposed merger. Public Knowledge, an American non-profit that advocates for an open internet, declared that the deal meant Comcast would “wield unprecedented gatekeeper power” over the internet and become “the bully in the schoolyard.” Writing the day the merger was announced, the New Yorker columnist John Cassidy declared that the new Comcast was a “self-serving scheme” designed to enrich an already rich monopolist. Susan Crawford, a Harvard law professor and telecommunications expert, declared that the soundtrack for the takeover was “ominous base music.”

“We deregulated high-speed internet ten years ago and since then we’ve seen enormous consolidation and monopolies,” Crawford, who served as an adviser to the former US president Barack Obama, told the BBC in 2013. “Left to their own devices, companies that supply internet access will charge high prices, because they face neither competition nor oversight.”

Activists were initially pessimistic about their chances of stopping the takeover. The Obama administration, after all, had allowed Comcast to acquire NBC Universal three years earlier—an acquisition that made Comcast the largest media company in the world. The then chairman of the federal communications commission—America’s equivalent of the TRAI—had former Comcast workers on his staff. Comcast’s leadership was also known to be close with the president. Roberts, for example, was a major donor to Obama, played golf with him and served on his jobs council. Roberts’ right-hand man, David Cohen, was a former Democratic Party operative who raised $1.14 million for Obama’s re-election campaign.

But the proposed acquisition became a galvanising cause for American progressives, many of whom came to view the deal’s fate as an acid test for whether the government supported preserving open access to the internet. Major politicians spoke out against the deal. The FCC was inundated with angry comments. The deal also invited criticism from major online content companies, such as Netflix, which worried that the resulting company would be so large that it could force them to pay high premiums for streaming their content or risk slowdowns. Tim Wu, the law professor who coined the term “net neutrality,” argued that Comcast’s incentives to do this went beyond just bigger payments. Comcast owned NBC, which meant it owned its own television shows. Netflix and Comcast were in direct competition.

In mid April 2015, regulators at the FCC indicated that they were sceptical the takeover was in the public interest. The United States justice department, which handles antitrust lawsuits, expressed similar worries. On Monday, 20 April, Roberts rushed to Washington and made a final pitch to the FCC chairman, but was ultimately turned down. By Wednesday, the deal was dead. Wu later wrote that once the justice department and the FCC’s “lawyers and economists studied the problem carefully, it became clear that the harm to competition was serious, and that no manner of political deal making could overcome it.”

It is possible that regulators could control Jio, or that politics could damage the company. Narendra Modi’s government may lose the general election this year, and Rahul Gandhi has accused the BJP of favouring Reliance. But the Ambanis have a way of surviving and thriving during even the toughest of times. Rahul’s father, Rajiv Gandhi, began his tenure as a Reliance critic and ended it as a Reliance friend. “They are quite strategic players in business, Mukesh at least,” Prashant Bhushan told me. “He realises that the future is in IT and telecom, and in the time to come it will really be these industries which will grow.”

Mukesh, for his part, does not seem concerned. Jio rolls on. The company now offers its own phone to consumers for a fully refundable Rs 1,500 deposit. It has its own television, movie, and music applications—some of which feature original content. Recently, Reliance launched a digital payments service app called JioMoney and a messaging app called JioChat. In an interview with the Wall Street Journal, one former Jio executive said that the company’s “whole management team was clear that Jio was not a competitor to Airtel but to Google and Netflix.”

Speaking at the inaugural India Mobile Congress, Mukesh declared that “data is the new oil and India does not need to import it.” As the leader of a business that turns petroleum into products, Ambani knows that ultimately, oil’s value lies in what it helps people access. His vision is not just an India filled with data. It is an India filled with Reliance. In it, Indians will make calls and surf the web on their JioPhones, using Jio’s services. They will pay each other over JioMoney. They will shop and order goods from Reliance Retail—Reliance’s expanding brick-and-mortar subsidiary.

They will watch shows produced by Network18, India’s third-largest broadcaster and another Reliance subsidiary, on JioTV. They will view films produced by Reliance-backed production companies, on JioCinema. And they will continue to buy products made from Reliance-manufactured plastic.

“Ambani wants to be seen on the global platform as a businessman, as somebody who can rival a Bill Gates or any of the big names,” one industry insider told me. “He wants to become a statesman.”

In the final analysis, it is the scope and scale of Ambani’s vision that makes the government’s favouritism so alarming. “Democracy gets captured,” Bhushan told me. “It creates these very large, very powerful corporations which become so powerful that they are beyond all accountability—they are beyond any law. They are a law unto themselves.”

LAST DECEMBER, the Ambanis travelled to Rajasthan to hold a pre-wedding party for Isha Ambani, Mukesh’s daughter. The family spent the weekend in the desert city of Udaipur, receiving thousands of guests. Hotels across Udaipur were completely booked. The local airport was clogged with charter flights that shuttled attendees in and out.

According to Mukesh, it was Isha who inspired him to create Jio. “The idea of Jio was first seeded by my daughter Isha in 2011,” he said while accepting a business award given by the London-based Financial Times. “She was a student at Yale and was home for the holidays. She wanted to submit some coursework—and she said, ‘Dad, the internet in our house sucks.’” It was the year after Reliance Industries acquired IBSPL, and the same year that the Ambanis moved into Antilia, a private 170-metre tall home estimated to be worth at least $1 billion. Isha and her twin brother, Akash, were both appointed to Reliance Jio’s board of directors in 2014, when they were 23 years old.

The wedding, which Bloomberg News estimated to have cost $100 million, or over Rs 700 crore, became a worldwide sensation. It featured a private concert by Beyoncé, held at the high-end Oberoi Udaivilas hotel. Another event was held at Udaipur’s City Palace. The guests had access to a phone application mapping out the festivities.

As with most high-profile Indian weddings, Isha’s reception featured a variety of Indian celebrities. But RIL has a sizeable network of international subsidiaries, and the wedding reception had an unusually large number of very powerful foreign guests. Hillary Clinton and John Kerry—the two top foreign-policy officials from Obama’s presidency—were both present at the festivities. Clinton and her husband, the former US president Bill Clinton, have known the Ambani family for 18 years. As president, Bill Clinton even met with Mukesh’s father.

I reached out to organisations affiliated with Hillary Clinton and Kerry, requesting interviews to discuss the wedding. Later, I emailed these groups asking for information about the amount of money the Ambani family spent on their attendance. I never received a reply.

It did not take long for a video of the two politicians to leak. In it, Kerry and Clinton are standing in the middle of a dancing crowd on a stage, awkwardly shuffling in place. Lights are flashing and Bollywood music is playing. The movie superstar Shah Rukh Khan, standing next to Clinton, tries to get her to dance, but is rebuffed. Eventually, Nita Ambani—Mukesh’s wife—walks over to Clinton, grabs her hands, and begins yanking them back and forth. Mukesh, smiling, dances his way around the stage.

Correction: An earlier version of this article mistakenly stated that Rs 12.5 crore, using the March 2014 exchange rate, amounted to $2 billion. The correct figure is roughly $2 million. The Caravan regrets the error.