The Green Revolution in Punjab

The Green Revolution is usually told as a seed story: two Mexican dwarf wheats, a response to fertiliser, a famine that did not happen. The seed mattered, but the variety that reached Punjab in the mid-1960s was one component of a package, and it was the package that got locked in.

Dwarf wheat yields more only if it gets water on schedule and nitrogen in quantity. The dwarfing gene was bred to stop lodging — a tall stem collapsing under a heavy ear — and it works only when the nitrogen is there. A farmer will plant it only if someone will take the harvest at a price he can bank on. The public system supplied water, nitrogen and the buyer.

That is the decision the following sixty years are still paying for: not "grow more wheat" but "make wheat the crop that carries a guaranteed return", then build the irrigation, the power and the procurement around it. Three bills were deferred.

Why wheat and rice, and not the crops they displaced

Agronomically, wheat and rice answered the package — assured canal and tubewell water, deep alluvial soils, a cold winter for wheat, a hot monsoon for rice — in a way pulses and millets did not. Millet yields are lower; pulses fix their own nitrogen, so they respond to fertiliser less.

Institutionally, a price floor only operates if there is a buyer who will take unlimited quantity at that price. Wheat and rice keep for a season without much loss, can be milled and moved in bulk, and a public distribution system will absorb them. A government that announces a support price for a pulse without a procuring agency behind it has announced a number, not a floor; the farmer who plants on it finds at harvest that the local trader is the only buyer.

So what the farmer read, correctly, was not "wheat is the better crop" but "wheat has a buyer, and the buyer is the state". Since the early 1970s the paddy area in Punjab has grown roughly tenfold, while the wheat area was already near saturation. That is a supply response to an institutional fact, not a preference.

Bill one: the aquifer

Paddy is a water crop: transplanted into standing water and kept flooded for weeks. Estimates of the water it takes per kilogram in the north-west run into the thousands of litres, and the spread between those estimates is part of the argument: nobody agrees on the denominator.

What turned the aquifer into a bill was not paddy on its own but the price of pumping. Agricultural power in Punjab has been free or flat-rated since the late 1970s. A flat or zero tariff breaks the link between water lifted and money paid: the marginal cost of an extra hour of pumping is zero, so the only limit on the draw is what the tubewell can yield. An electric pump on a free connection is an open tap.

The consequence shows in the water table, not in a price. Central Ground Water Board assessments class most of Punjab's blocks as over-exploited, with extraction well above recharge; the state's stage of extraction has been reported at around one and a half times its recharge, and in the central districts the table has fallen by roughly half a metre to a metre a year for two decades.

Because the state could not price water, it tried to ration it by calendar. The Punjab Preservation of Subsoil Water Act, 2009, bars transplanting paddy before a notified date, pushing the crop into the monsoon. A real law with a measurable effect — and evidence of the shape of the problem: when price is unavailable as a lever, the remaining lever is a date.

Bill two: the soil

What got squeezed was the rotation. A traditional rotation put a maize, a pulse or a fodder crop into the summer slot; rice and wheat take both slots, so the same two crops draw the same nutrients from the same depth in the same order, every year.

Three things follow, none dramatic in a single season. Organic carbon declines, because less biomass is returned and more removed as grain and straw. Micronutrient deficiency — zinc, iron, manganese — becomes common enough that those nutrients are added rather than inherited. And the response to nitrogen falls, so the same yield needs more urea: the input bill rises to hold the output constant.

The calendar adds a second problem. Paddy is harvested in October and wheat sown in November, and the window between them is two to three weeks. There is no time to clear and incorporate the residue, so the cheapest option is to burn it. The stubble-burning season is not a behaviour problem; it is what a two-week window looks like when the next crop has a fixed sowing date and a guaranteed buyer.

Bill three: the budget

The other two arrive in the physical world. This one arrives every year, before the season starts.

A minimum support price is not a payment. It is an offer to buy at a floor, and it costs the exchequer in proportion to how much is offered and at what price relative to the market. In wheat and rice the offer is effectively open-ended: agencies buy whatever arrives, and the central pool absorbs it. That converts a price policy into an annual expenditure with three parts — the gap between the support price and the price at which the grain is sold, the cost of carrying the stock, and the power subsidy that made the water free.

All three are committed before the harvest, and all are the income of the households the original policy was meant to protect. That is what makes this bill the hardest: a subsidy reaching millions of households is not a line item anyone strikes out. It is also why the exit has been discussed for thirty years and attempted only in pieces: diversification, a direct income transfer, a cap on procurement. Each hits the same wall — no other crop has a buyer, so leaving rice and wheat means leaving the only crop with a floor under it.

The arithmetic of a policy that worked

None of the three bills is hidden. The water table is monitored, the soil is tested, the subsidy appears in the budget every year. What made them easy to defer is that none is paid by the person who received the benefit of the decision: the aquifer against the next decade's water, the soil against the next rotation's yield, the budget against a taxpayer in a different line.

The north-west became the country's grain store: wheat output rose far enough that a country importing wheat on concessional terms in the mid-1960s had stopped doing so within a decade. The buffer stock that absorbs a bad monsoon sits in warehouses in these two states, and the yields that fill it are still being harvested.

The bill is also real. A policy that raises output by making one crop unlosable will, over time, produce a landscape in which that crop is the only rational choice — and a landscape with one rational choice is expensive to leave. Not that the decision was wrong: that its costs were booked to three ledgers nobody was required to read at the time.

This is one of the Modern India entries in About India.