Tuesday, 24 March 2020. At 11:01 at night the Prime Minister’s Office posts a release.
“PM calls for complete lockdown of entire nation for 21 days,” it reads, “beginning at midnight tonight.” The address was given that evening on television. Social distancing, Narendra Modi said, was the only option. He warned that if India did not handle the 21 days well, the country would go backwards by 21 years. Crowds mobbed the shops after the broadcast.
There will be a total ban on venturing out of your homes.
Within days millions of migrant workers begin to walk home, and the Guardian, writing on 30 March, calls it an exodus unlike anything seen in India since partition. One six-person family in Gurgaon faces almost five days on foot to cover the 125 miles to its village. By the paper’s count more than twenty migrant labourers had died in a few days trying to walk back.
This chapter is about that road and another one. On the first, in March, the state had to reach people who were leaving. On the second, in November, it met people who would not move. What links them is older than either, and the chapter will go back to find it.
The first walk
The state did not stand still. Two days into the lockdown the Finance Minister announced a relief package of Rs 1.70 lakh crore. Eighty crore poor people were to get 5 kg of wheat or rice and 1 kg of pulses free every month for three months. Women holding Jan Dhan accounts were to get Rs 500 a month on the rails of chapter 22. Eight crore families were to get free gas cylinders. A provident-fund top-up was aimed at wage-earners below Rs 15,000 a month in firms with fewer than 100 workers. The channels the package used were the ration system, the Jan Dhan account and a farmer-payment scheme reused as a cash channel. The book’s inference, not a finding, is that these channels follow a person’s registered home, and the people in the Guardian’s report were away from it. The documents show the gap only indirectly. In late May the Supreme Court, acting on its own motion, directed states not to charge migrant workers fares for the journey home and to provide them with food and water. Indian Railways ran 3,736 special trains from 1 May to late that month and carried more than 48 lakh migrants.
What the lockdown bought is also on the record, and it is the government’s own account. In April the Centre said, from a simple model, that at least 14 to 29 lakh cases and 37,000 to 71,000 deaths had been averted. The cost was counted in jobs. The Centre for Monitoring Indian Economy put April unemployment at 23.5 percent, against 8.7 percent in March, and 122 million jobs lost. The Azim Premji University’s State of Working India 2021 counted about 100 million jobs lost in April and May, most recovered by June and about 15 million still gone, and 230 million people pushed below the national minimum-wage threshold of Rs 375 a day over the year. The two surveys count different things, and their direction is not in doubt. In October 2020 the IMF’s chief economist said lockdowns should be targeted and local, and that India could have given more direct support to the urban poor and migrants.
2020New DelhiOn 1 February the Budget had said that farm markets needed to be liberalised and that distortions in them should be removed. It proposed then only to encourage states that adopted the Centre’s model laws on land leasing, produce marketing and contract farming. In June, the government moved from model laws to ordinances: three of them, covering trade outside the old markets, contract farming and stock limits. The bills that replaced them were introduced in the Lok Sabha on 14 September. The trade and price-assurance bills passed the Lok Sabha on 17 September and the Rajya Sabha on 20 September. The Essential Commodities bill passed on 15 and 22 September.
Two of the bills passed the Rajya Sabha on the 20th by voice vote. The opposition, angry that the Deputy Chairman would not allow votes on the resolutions it had moved, broke microphones and stood on tables. The Parliamentary Affairs Minister replied that they should have asked for a division. The opposition called it a sad day for parliamentary democracy. The bills had gone through without committee scrutiny, a process question the chain-break below returns to.
What the laws did is simple to state. The first allowed farmers’ produce to be traded outside the premises of the state-regulated markets, the mandis, and forbade states to charge market fees on such trade. The second set up a framework for farming agreements with price assurance. The third allowed stock limits on farm produce only in times of steep price rises. The government’s stated aim was more buyers and better prices. In October the IMF’s chief economist called the farm and labour bills steps in the right direction.
2020HaryanaAt the end of November more than 300,000 farmers marched on Delhi from Punjab and Haryana, on foot and in convoys of tractors, for what they called a decisive battle. Police used tear gas and water cannon. They ripped up highways to stop the tractors. Most of the marchers were halted by barricades and barbed wire, and they camped on the roads.
What did they fear? The Indian Express put it in one sentence: since the states could no longer collect market fees on trade outside the markets, farmers believed the laws would gradually end the mandi system and leave them at the mercy of corporates. They wanted the laws withdrawn, or a new law guaranteeing the minimum support price. By December, Punjab was paying tribute to 41 farmers who had died since protests intensified on 15 September.
Lay the two roads on top of each other and you can see the whole problem. The book’s reading: in March the road ran outward from the capital, and the relief channels followed registered homes, not the road. In November it ran inward, and the people on it would not turn. In both cases the question was what the state knew about the people at the far end of the road, and what they had been promised. The second question has a date on it. It goes back to a drought.
Chapter 10 left India in the drought year. This chapter follows the seed.
According to Norman Borlaug’s Nobel lecture, in 1966 India imported 18,000 tonnes of Mexican wheat seed. The strategy that came with it assured farmers a fair price and access to seed, fertiliser, credit and machinery. The price assurance was not an afterthought. It came in the same package as the seed.
The state then became a buyer. Net imports of foodgrains fell from 10.3 million tonnes in 1966 to 5.7 million in 1968, and state procurement rose over the same two years from 4.0 million tonnes to 6.8 million. By 1969–70 about 35 percent of the wheat area grew the Mexican varieties, and farmers had sunk about 70,000 private tube-wells in a single season. By the same account, wheat yields were the thrust of the revolution in India, and rice played only a minor role there.
Output recovered from the drought year to 83.2 million tonnes by 1968, and the seed and the monsoon cannot be separated in these figures.
Chain I161966 → 2021The seed that vetoed a law19472026A rescue policy creates a constituency, and the constituency is made of people with a stake. This is not a verdict on whether the Green Revolution was right. It plainly saved a country from dependence on foreign grain. The point is narrower: the system that guaranteed a price in one set of regions became, in time, something those regions would defend against a reform written for the whole country.
Replay of an earlier scene: Anhui, 1961: the fields, and the man who changed his mindThere is a mirror to this, and it is a rhyme, not a lesson. Chapter 4 told the Anhui story: a farm reform that was spreading across the province (85.4 percent of its production teams by the end of autumn) until, in December of that year, Mao said he had changed his view. The reform was ended from the top.
In India in 2020 and 2021 a farm reform was withdrawn by the government after a year of protest by farmers, who had the right to march and camp on a highway. The Chinese reform ended by a decision from above; the Indian one ended by a decision taken under pressure from below.
The familiar line says democracy is slow and autocracy is fast. The pairing does not support it. Anhui’s fields spread across most of a province in a single season. India’s laws went from ordinance to passage in Parliament in under four months. Both were fast, and both were stopped. In neither story is speed the dividing line.
April 2021, Ahmedabad
2021AhmedabadWhile the camps stayed on the highways, the second wave arrived. On 26 April India recorded 352,991 new cases and 2,812 deaths in one day, its highest to that point. By 2 May the health ministry was reporting 3,689 deaths in 24 hours and had confirmed more than 215,000 in all, though the BBC wrote that the real toll was thought to be far higher.
On 11 April reporters from the Gujarati daily Sandesh watched the mortuary of Ahmedabad’s 1,200-bed state-run Covid hospital for 17 hours and counted 69 body bags. The next day Gujarat officially counted 55 deaths, 20 of them from Ahmedabad. On the night of 16 April the reporters visited 21 cremation grounds and found more than 200 Covid funerals. The Hindu reported 689 bodies cremated or buried in seven cities that day, when the state’s official toll for the whole of Gujarat was 94. The state government denied under-counting and said it followed federal protocols.
Here is the difference between a count and an estimate, laid out with names. The official figure reported through the end of 2021 was 481,080. A survey of about 140,000 adults by Prabhat Jha and colleagues put COVID deaths between June 2020 and July 2021 at about 3.2 million. Anand, Sandefur and Subramanian gave three further estimates of excess deaths through June 2021: about 3.4 million, 4 million and 4.9 million. The WHO’s model put excess deaths in 2020–21 at about 4.7 million. India’s health ministry strongly objected, calling the method statistically unsound, and noted that WHO had used data for 17 states taken from websites and media reports. The book prints the range with its authors and does not average it.
The lockdown’s benefit and the lockdown’s walk belong in the same ledger. The government said the first lockdown averted a very large number of cases and deaths. The Guardian counted the migrants who died on the road. A newsroom in Ahmedabad was counting deaths because the number it was given did not match what it saw.
202109:55New DelhiAt 9:55 on the morning of Friday 19 November 2021, the Prime Minister’s address is posted. It is the day of Guru Nanak’s Prakash Purab.
we could not explain to some farmers despite our efforts
He said the government had decided to withdraw all three agricultural laws, and that the constitutional process to repeal them would be completed in the Parliament session starting later that month. He said the laws were meant to strengthen small farmers, 80 of every 100 of whom hold less than two hectares. He announced a committee on zero-budget farming, crop patterns and making the support price more effective and transparent. The Repeal Bill was introduced in the Lok Sabha on 29 November. The book imputes no motive to the timing.
In December the Centre sent a formal letter to the protesters. It promised a committee on the support price that would include the Samyukta Kisan Morcha, said the status quo would be kept on procurement at the support price, and said that five states had agreed to withdraw cases against farmers. The cost of the year on the highways was counted by others. At day 100 the Morcha put the number of farmers who had died at the borders of Delhi at 248, some of them from illness and some by suicide. In August 2022 Punjab’s Chief Minister said his government had given Rs 5 lakh each to the families of 789 farmers who died during the agitation, because of various reasons including the harsh winter, road accidents and cardiac arrest.
The Supreme Court panel’s report had been submitted on 19 March 2021. It was made public in March 2022 by one of its members, Anil Ghanwat. It recommended that the three laws not be repealed, and suggested changes, among them freedom for states to make the support price legal. Nothing in this chapter says which side was right. Both are on the record: a majority of the organisations that wrote to the panel supported the laws, and a very large number of farmers camped on the highways through the winter and beyond, still there at day 100.
2026The 2025–26 Economic Survey, the government’s own, is the later word on what the system built in the 1960s has become. It judges rice and wheat procurement to have outgrown food-security needs. It finds that the ratio of nitrogen, phosphorus and potash applied to soils has drifted from 4:3.2:1 in 2009–10 to about 10.9:4.1:1 in 2023–24, against an agronomic norm near 4:2:1, driven mainly by excess urea. It does not trace either finding to 1966, and the book does not either. What the 1966 system left is a procurement apparatus, and the regions that depend on it.
Five months before the farmers reached Delhi’s borders, at least twenty Indian soldiers had died in the Galwan Valley, on a line the two countries have never agreed.