We stand at a critical point in India’s economic history, when jobs finally appear to have started shifting away from agriculture. During the past few decades, even as policymakers celebrated a decline in the share of agriculture in jobs, the number of workers in agriculture was still rising as the population was growing; it grew from 10 crore in 1951 to 26 crore in the 2011 census. For perspective, 25 lakh people employed in agriculture in the US produce enough for the US and more: They have had a trade surplus in agriculture for at least the last 50 years. If Indian farmers’ per person output were similar to the US, India would need only 88 lakh farmers (just 3.3 per cent of the 26 crore it has now).
The fact that India has too many people cultivating too little land has been observed at least as far back as 1880, by the First Indian Famine Commission (FIFC), which observed that worker to land ratios in India were even then more than 10 times Britain’s. This was, of course, after nearly a century of de-industrialisation in India. Eminent thinkers including B.R. Ambedkar (1918) have regularly highlighted the problem of disguised unemployment in agriculture (a 2008 paper by J. Krishnamurthy is a useful compendium of such research).
Recent annual surveys however suggest a sharp drop in agricultural employment: From 52.2 per cent of total workers in 2012 to 45.7 per cent in 2015. Two successive monsoon failures are a likely explanation. As growth in the value of output of agriculture (which is a good proxy for income growth for all people in agriculture) slowed from a 14 per cent annualised rate for nearly a decade till 2014 to below 5 per cent each in the following two years, workers had to start moving away. However, where an average monsoon in 2016 was expected to revive agricultural income growth, it may not (see “Paradox of Plenty”, IE, August 22, 2016). Higher volume growth has driven food prices down. It appears unlikely that the decline in agricultural employment can be reversed.
At the root of this stagnation in agricultural incomes lies slowing demand for food while agricultural productivity rises sharply, with access to markets (roads), information (phones) and automation (electricity, machines) improving substantially in the last decade. After all, when the world worries about robots taking away manufacturing jobs and neural networks (that is, artificial intelligence) replacing service jobs (of analysts, lawyers, and even doctors), how long can half our workers stay meaningfully employed growing food? That so little of our agricultural produce is processed further worsens the problem, as quality/processing is necessary to export the surpluses.
This creates other problems: With nearly half of India’s households seeing anaemic income growth, demand for mass-consumption products like soaps and detergents weakens. These households are also unlikely to add rooms to their houses: As nearly 60 per cent of construction jobs are in rural house construction, this exacerbates the problem. After all, landless manual labour moving out of agriculture is most easily redeployed in construction. As farmers are generally the largest buyers of farmland (a form of saving they prefer), this is also likely to depress land prices, hurting wealth perceptions.
This is very likely the common theme behind the spate of protests by agricultural communities in the last two years — for example, the Jats in Haryana, the Patels in Gujarat and the Marathas in Maharashtra. They were all protesting for government jobs: Given that nearly two-thirds of all salaried jobs in rural India are with the government or government-owned organisations, this is not surprising. Worryingly, though, these protests seem likely to spread as agriculture sheds workers.
One solution that is perhaps politically expedient (and has been popular in the past) is fiscal largesse that inflates food prices (for example, through aggressive growth in agricultural credit, farm loan waivers, and directed subsidies). This supports agricultural income growth and also reduces income inequality. But this medicine has other side-effects: High inflation, high fiscal deficits, high interest rates and a weak currency.
As it is clear that the transition can only be delayed, not avoided, a good doctor would not recommend this path. Adolescence, for many, is a painful period of transition, but an essential one.
That brings us to the other solution: Create non-agricultural jobs. This is not new. The FIFC in 1880 suggested “the complete remedy for this condition (of famine) . will be found only in the development of industries other than agriculture”. But how, is the question: Only 3.5 per cent of rural households currently have private salaried jobs (as per the Socio-Economic Caste Census), that is, only slightly above 6 million jobs.
However, if, say, by 2040, agriculture in India employs 8 crore people (that is, per person output is still one-tenth of the US’ current level), agriculture could be shedding 6-7 million workers every year. This is over and above the 10-12 million people entering the workforce every year. This poses a formidable policy challenge for the government. It cannot rely on corporate India (at least its current avatar).
The challenge, contrary to popular concerns, is not a lack of options or demand. Not only can India be a competitive agricultural exporter (with its year-long sunshine, fertile land, good rainfall and cheap labour), there is significant growth potential (and therefore jobs) in food processing, and production of consumer appliances (where Indian per capita consumption levels can increase significantly). Even labour-intensive exports like leather (apparently a middle-school dropout can be trained in three months), where India’s total exports are less than the procurement by one large US retailer, can create jobs. Indeed, industrial revolutions in the past have been built on cheap labour exiting agriculture.
The challenge, instead, as when one is thirsty during a tough trek, and staring down a cliff seeing that water source a thousand feet below, is how to get there. We know the end-point: But what must the government do?
One can be tempted to intervene directly, but given the limitations of state capacity, such efforts have failed miserably in the past. A wiser approach might be to provide the enabling environment — good law and order, roads, electricity, telephones (and data!), cheaper but market-priced credit, fair markets and easy government interfaces, and wait for the entrepreneurial instincts of our people to do the rest. Providing these is not easy, and certainly not on the scale that these are needed. But what are the alternatives?