Why Large Parts of North India Turn Dystopian Every Winter

— Picture by Neil Palmer (CIAT). Burning of rice residues in Punjab, India, prior to the wheat season.

In the end, it’s all about incentives, perverse or otherwise.

My parents moved to Delhi in 2009, after my father retired from Coal India. Since then I have spent all Diwalis in Delhi, though this Diwali due to reasons beyond my control, I will most probably be in Mumbai.

The last few Diwalis in Delhi have been very difficult for me. In fact, last year, I could smell smoke inside the house even before the Diwali day (so, it was clearly not because of crackers). Delhi and many other parts of North India go totally dystopian during winters.

On some days when one gets up and looks out of the window, there is so much smog that one gets a feeling that Armageddon is here.

One of the primary reasons for this smog/pollution is the rice stubble burning that happens in Punjab, Haryana, parts of Uttarakhand and Western Uttar Pradesh. From the looks of it, the situation doesn’t seem to be very different this year.

Newsreports suggest that stubble burning is currently on and was responsible for 40% of Delhi’s pollution on November 1, the highest it has been so far this season. Last year on the same day, the stubble burning’s contribution to Delhi’s pollution had stood at 44% on November 1.

Let’s try and understand this issue in detail and why it happens every year.

This is a great story of how noble intentions on part of politicians and bureaucrats (yes, you read that right) along with incentives that seem right when they are introduced, can really screw up things in the years to come.

And once a system is in place, right or wrong, it is difficult to change it, given that many individuals benefit from the status quo.

Why do farmers burn rice paddy stubble?

They do it primarily because the time farmers have between harvesting rice paddy and sowing the wheat crop, is very short. That’s the short answer. But there is a lot more to it than just this.

Mechanical harvesters found their way into Punjab sometime in the early 1980s. Around four-fifths of the rice crop is harvested using combine harvesters and not human beings (if you still thought humans being carry out harvesting in Punjab, you haven’t moved beyond Hindi cinema of the 1960s).

These harvesters cut and clean rice from the rice paddy, but they leave behind straws on the field. These straws are six to eight inches long and for all practical purposes are useless.

The straws remaining after harvesting of wheat can be used as animal fodder. Rice straw cannot be used as animal fodder primarily because of its high silica content. If this straw is used as animal fodder, it impacts the quality of milk, with the quantity of calcium in the milk coming down. This is not true about straw left behind after harvesting basmati rice, which has low silica content.

But basmati is grown only in a limited area. Like this year, rice was planted on a total area of 27.36 lakh hectares. Of this, basmati was planted on around 6.5 lakh hectares or 24% of the total area under rice. This was primarily because the government does not buy basmati rice under the minimum support price (MSP) structure. (We shall look at this in detail later).

Farmers have a time of around 10-15 days for removing the rice straw and get the fields ready for planting wheat. The easiest thing in this situation is to burn the rice straw. All it takes is a single well-lit matchstick. The cost is close to zero.

This burning leads to higher pollution and deterioration of air quality even in places hundreds of kilometres away. The heat from burning the rice straw leads to an increase in soil temperature which kills beneficial soil organisms. The burning is also a potential source for greenhouse gases. Also, it is worth remembering that the burning of post-harvest rice stubble forms around 50% of the all the crop residue burning in the country.

Let’s take a look at stubble burning in incidences reported in Punjab and Haryana, where most of the burning takes place.

Source: Price Policy for Kharif Crops – The Marketing Season of 2020-2021, Commission for Agriculture Costs and Prices.

As can be seen from above table, the number of stubble burning incidents have come down over the years. The total number of incidents in Punjab and Haryana have come down by 52% between 2016 and 2019. In 2016, the total number of incidents had stood at 1,18,065. By 2019, this was down to 56,742.

So there has been some improvement on the number of fires front over the years. But there are other ways of looking at the situation; the total weight of the stubble burned and the total area of stubble burned. The Chief Secretary of Punjab Vini Mahajan said on October 31, that the total straw burning area in 2020 was 7.49 lakh hectares, which was 5.23% lower in comparison to 7.90 lakh hectares last year.

A Amarender Reddy, the principal scientist at the ICAR-Central Research Institute for Dryland Agriculture recently wrote in The Wire that last year the farmers burnt 11 million tonnes of rice stubble in Punjab and Haryana. This is a little over 40% of the total stubble of 27 million tonnes of rice stubble produced in both the states. Reddy expects the figure to be the same this year.

There has been some improvement on the paddy burning front. One reason has been the distribution of Happy Seeder, a machine which cuts rice stubble and plants wheat seeds at the same time. Reddy writes that the Punjab government has distributed around 24,000 Happy Seeder machines though the state needs nearly 50,000 seeder machines to remove all the rice stubble in the short period of time available before wheat seeds are planted.

Also, farmers have complained about low germination of wheat seeds when the Happy Seeder machine is used. The central government, like most central governments, has allocated more money to solve the problem. Using this money, machines to tackle the rice stubble can be bought at a subsidy.

While, all this is fine, it doesn’t answer the most basic question: why do semi-arid states like Punjab and Haryana, grow a water-intensive crop like rice paddy in the first place?

Why Punjab and Haryana grow rice?

Punjab has the highest yield of 4,132 kgs per hectare when it comes to rice, against the all India yield of 2,659 kgs per hectare. The rice productivity in Haryana is better than the all India average and is at 3,121 kgs per hectare. But this does not take into account the total amount of water used to produce this rice.

As the document titled The Price Policy for Kharif Crops: The Marketing Season for 2016-17, brought out by the Commission for Agriculture Costs and Prices, points out:

“If water consumption is measured in terms of per kilogram of rice, West Bengal becomes the most efficient state, which consumes 2,169 litres to produce one kg of rice, followed by Assam (2,432 litres) and Karnataka (2,635 litres). The water use is high in Punjab (4,118 litres), Tamil Nadu (4,557 litres) and Uttar Pradesh (4,384 litres). … [This] shows that the most efficient state in terms of land productivity is not necessarily the most efficient if irrigation water is factored into. This is because of high rainfall in the eastern region.”

Haryana also uses a lot of water to grow rice.

What this means is that Punjab and Haryana given that they are semi-arid water deficient areas, should not be growing rice in the first place. In the early sixties, Punjab used to grow crops which did not require a lot of water. These included maize, bajra, pulses, oilseeds etc. But over the years, the share of these crops in the overall cropped area has come down dramatically. Take a look at the following table.

Source: Economic Survey of Punjab, 2019-20.

Rice paddy was grown only on 4.8% of cropped area in 1960-61. In 2018-19 it was grown on around 39.6% of cropped area. What happened here? Sometime in the mid 1960s, the central government launched the Green Revolution in Punjab, in order to build food security in India and reduce our dependence on import of American wheat under the Public Law 480 (PL -480).

The farmers were encouraged to plant a high-yielding variety of wheat. In order to incentivise them, the government bought this wheat from them at a minimum support price (MSP) which was declared every year.

A look at the above table tells us that the cropped area under wheat jumped from 27.3% in 1960-61 to 40.5% in 1970-71. The fact that the government bought the wheat at the MSP, led to an increase in wheat plantation.

The government started buying rice at an MSP as well. This led to a jump in number of farmers planting rice in Punjab and Haryana because they had a readymade customer in the government willing to buy at a fixed price. They weren’t subject to the vagaries of price and India’s underdeveloped agricultural marketing system.

The farmers were first incentivised to grow wheat (rightly) and then incentivised to grow rice as well (right from the point of food security, but wrong from all other angles).

Take a look at the following chart which plots the total amount of area on which rice has been planted in Punjab, over the years.

Source: http://punenvis.nic.in/index3.aspx?sslid=5882&subsublinkid=4993&langid=1&mid=1 and Agricultural Statistics at a Glance 2019.

As can be seen there was a major jump in the area under rice production between 1970-71 and 1990-91, from 0.39 million hectare to 2.02 million hectare. This was primarily because of rice being bought by the government at a minimum support price announced every year. The next jump came in the mid 1990s.

In the year 1997, free electricity for farmers was introduced in Punjab. This encouraged farmers to grow rice even more. They could now pump groundwater for free. This could be used to grow rice. Take a look at the following table, which plots the number of tube wells in the state over the years.

Number of tubewells (in lakhs)

Source: Economic Survey of Punjab, 2019-20.

As can be seen from the above table, the number of electrically operated tube wells has gone up dramatically over the years. In 2018-19, the number is more than 13 lakhs. With free electricity, farmers were incentivised to buy electricity operated tube wells and pump as much ground water as required to grow rice.

This has led to the exploitation of groundwater. As the latest Economic Survey of Punjab points out:

“A state-wise assessment of the groundwater resources in the country showed that 80% of 138 blocks assessed were ‘Over-exploited’, 2 blocks were ‘Critical’, 5 were ‘Semi-Critical’, and 22 were ‘Safe’.”

In fact, 95% of groundwater is extracted for the purpose of irrigation.

Along with this, the Food Corporation of India (FCI) buys up a bulk of the rice produced in the state. It is worth remembering here that Punjab is not much of a rice eating state. Take a look at the following chart.

Procurement of rice in major producing states.


Source: Price Policy for Kharif Crops – The Marketing Season of 2020-21.

In 2018-19, Punjab produced around 12.82 million tonnes of rice. Of this, 11.4 million tonnes was procured by the government through FCI and other state procurement agencies. In Haryana, around 4.5 million tonnes of rice was grown. Of this, around 85% was procured. The major reason for this lies in the fact that given that the green revolution started here, FCI has the best infrastructure to procure and store foodgrains, in this area.

The government doesn’t procure basmati under MSP because there is a huge international demand, given its fragrant smell when cooked. Hence, farmers don’t plant much of it. While there is international demand, the farmers also need to suffer the vagaries of price.

This easy procurement along with free electricity encourages farmers to grow rice in what is largely a semi-arid area. But this still does not explain how the farmers came around to burning rice paddy stubble. I mean, I have been going to Delhi for more than 35 years now, but the city was never dystopian during winters earlier. This is clearly a phenomenon of the last decade. What changed?

What led to farmers burning rice stubble?

As we have seen, the government policies over the years, have incentivised farmers to grow rice. At the same time, these policies have led to the water table in Punjab falling dramatically. Given this, the government had to something about this and it did. (I am talking more about Punjab than Haryana here, simply because the number of fires in Punjab is many times more).

As the Economic Survey of Punjab points out:

“It requires 4,500 litres of water to grow one kg of sathi rice when it is sown in April-May. But if sowing is done around mid-June, water requirement reduces to 1,500-2,000 litres. Water requirement is high in April-May because the evaporation rate is high and there is no rain. As a result, all the water used in irrigation is groundwater. In June-July, rainfall supports water needs of the crop.”

This logic essentially led to the enactment of the Punjab Preservation of Subsoil Water Act in 2009. As per this law, farmers are not allowed to sow paddy seeds in nurseries before May 10. They are not allowed to transplant the saplings before June 10. The idea being that by the time farmers start transplanting the saplings in the fields, the Monsoon would have already arrived and hence, lesser groundwater will be used to grow rice. Haryana has a similar law.

The intention behind the law was noble, but the incentive it created for the farmer was again perverse. The end to end production of rice takes 120 days. The process of growing rice used to start in April earlier. But this was pushed back by a month due to the law to prevent the overexploitation of groundwater.

This led to a situation where farmers were left with a time of around 15 days to get their fields ready for the plantation of wheat. The quickest way to turnaround is to burn the rice stubble and that is precisely what has been happening for the last decade.

History plus perverse incentives are at the heart of this problem.

What’s the way out of this?

The central government recently told the Supreme Court that it was planning to bring a new law to tackle the stubble burning problem. This is a classic way of how any government tries to tackle a long-term problem. They either bring a new law or throw money at it, in the hope of solving the problem.

But the question is will this law or any law be of help? The Punjab government did bring in a law to solve one problem and ended up creating another one, without really solving the first one.

In the short-term, innovations like the Happy Seeder have clearly helped. But the problem can only be solved if the Punjabi and other farmers in the semi-arid areas of North India are incentivised to not grow rice and to grow other crops which do not require a lot of water.

But at the risk of repeating a cliché, it is easier said than done.

Let’s take a look at this pointwise.

1) Over the years, the government has bought much more rice and wheat than it needs to maintain the operational reserve and the strategic reserve. Like in September, the rice stock in the central pool of the FCI was at 22.2 million tonnes. As of October every year, FCI needs to maintain an operational reserve of 8.25 million tonnes and a strategic reserve of 2 million tonnes. Clearly, the FCI has much more rice than is required.

This reserve can be brought down by buying lesser rice in the time to come. If this policy is followed for a few years, the farmers will automatically be disincentivised to grow rice. If they are disincentivised to grow rice, there will be lesser rice stubble to burn.

Of course, this is a politically risky move and in the process a section of farmers is bound to face losses, until they move away from growing rice.

2) Another way is to buy more rice from states like West Bengal, which is best suited to be growing rice, given it uses less water to grow rice, in comparison to other states. In fact, West Bengal produced 16.05 million tonnes of rice in 2018-19. Of this, the government purchased just 1.9 million tonnes. The point to remember here is that West Bengal is a rice eating state.

So, unlike Punjab the government cannot buy almost all the rice that is produced. Hence, buying by the central government shouldn’t lead to a shortage of rice in the state, forcing it to buy rice from other states, in the process. The solution lies in helping increase the rice yield per hectare in the state. In 2018-19, West Bengal produced 2,906 kgs of rice per hectare. This was significantly lower than Punjab’s 4,132 kgs per hectare, but more than the national average of 2,659 kgs per hectare.

3) The most important way in weaning away farmers from rice is to change incentives. Let me offer an analogy here. Why does a wealth manager/insurance agent/personal banker/mutual fund agent mis-sell? Simply because their incentives are so aligned.

Along similar lines, if the farmer has an incentive to grow rice (and unlike financial salesmen, the incentive here is an honest one), he will grow rice. We can’t judge him for this.

One way out is to encourage farmers to grow maize, like they used to in the sixties. In 1960-61, 6.9% of the total cropped area in Punjab was used to grow maize. By 2018-19, this had fallen to 1.4%. As the document titled Price Policy for Kharif Crops—The Marketing Season of 2020-21 points out: “Maize cultivation is more water efficient than rice… [It has] a great potential for crop diversification in rice-wheat cropping system areas of north-western plains, where substantial groundwater depletion has occurred.”

The trouble is that maize has low profitability in comparison to rice “due to low and fluctuating prices and yield of maize.”

As the Price Policy document points out:

“There is a need to find alternative uses of maize in the country for industrial uses like feed, starch and ethanol as well as for direct consumption, mainly value-added products… Allowing maize as raw material for ethanol production would help in crop diversification and ensure remunerative prices to farmers.”

This will help increase the demand for maize and help increase its price.

Along similar lines, there is a need to encourage and incentivise the growing of pulses and oilseeds, which we don’t grow enough of. As the Price Policy document points out:

“Instead of promoting water-intensive crops like rice… it is important to promote production of pulses and oilseeds by encouraging farmers to grow these crops by providing better quality seeds, technology and appropriate price support to address gap between domestic production and consumption and maintain stability in the domestic market.”

The fact of the matter is that the private agriculture markets in the country don’t function well. At the same time, in order to encourage farmers to grow particular crops, the government cannot buy a large amount of it, like it buys rice and wheat, simply because it doesn’t have enough money to do so or the right infrastructure to store what it has bought. Pulses are an excellent example. The reason FCI cannot buy pulses is simply because it doesn’t have the right infrastructure to store them.

In this scenario, getting farmers to grow something other than rice is going to be very difficult and will take a lot of concentrated effort on part of the politicians as well as bureaucrats. Will that happen? On that your guess is as good as mine.

The moral of the story being, just because there is a problem, doesn’t mean it has an immediately implementable solution.