Wednesday, October 24, 2012

The young are getting away from agriculture

Who does agriculture in India? Here's some fascinating evidence, from the CMIE Household Survey for the quarter Apr-May-June 2012. This is a survey of 700,000 individuals in 150,000 households all across India, both urban and rural. Let's look at the share of the working population, in each age group, that's engaged in agriculture:

Age 15-20 19.69
Age 20-25 21.22
Age 25-30 24.70
Age 30-35 28.22
Age 35-40 30.91
Age 40-45 32.76
Age 45-50 34.75
Age 50-55 36.96
Age 55-60 40.02
Overall 31.31

As we see, in the overall dataset, 31.31 per cent of the working population is in agriculture. CMIE shows three categories of this -- `Small farmer', `Organised farmer' and `Agricultural labourer'. I have added up these three categories to make the table above.

That 31.31 per cent of the Indian workforce is in agriculture is fairly well known. What I had not thought about, previously, is the age structure. Will agriculture have a bigger share of young or old workers? We can envisage two competing effects. On one hand, if a family has underemployed young ones who are engaged in agriculture by default, then we'd see a lot of young people in agriculture. On the other hand, if families try hard to get their kids off the farm, and the growth in industry and services in India is successfully absorbing this workforce, then we should see a smaller share with the young.

The evidence above favours the latter story. The share of the overall workforce which is engaged in agriculture is 31.31%. But amongst the old (age 55-60), the share is higher at 40.02%. This share steadily drops as you get to the young. In the class of the working young (i.e. age 15-20 but a part of the working population), just 19.69% are in agriculture.

Perhaps there is greater malleability of human capital with the young: the old may not be able to easily pick up the skills required to participate in the modern world of services and industry. When the shift of a worker into services or industry is accompanied by migration, it adds up to a powerful engine of social and economic modernisation. It is a powerful mega-trend that is reshaping India today.

The agricultural workforce is greying. There are many divides between the old India and the new one. This evidence suggests one more: the old world of agriculture is disproportionately one of the old, while the new worlds of industry and services are disproportionately manned by the young.

This data helps us understand India's demographic dividend. Many people worry that services and manufacturing in India will not absorb the great surge of young people in India. If that was the case, there would be a lot more people in agriculture. Instead, we see only 20% of the young depending on agriculture.

The application of sound economic principles in the field of agriculture will give us a situation where no more than 5% of the workforce is required there. At present, agriculture is using up 31% of the workforce. This gives us a headroom of an additional 25% of the workforce which can move out. This movement would give a one-time improvement in GDP because the per-worker output in industry or services is greater than that seen in agriculture. But these effects are diminished with the young, where the alteration that's feasible is smaller: from 20% to 5%.

For an interesting comparison against China, in 2007, roughly 10% of the workforce was in agriculture in the age group from 16 till 35. By the time you got to the age group of 41-50 (in 2007), roughly 45% were in agriculture.  By 2012, China has reached a point where there is relatively little upside for GDP growth by getting workers out of agriculture. The Indian evidence for 2012 looks similar to China of 2004, so India is perhaps 10 years away from this loss of upside in GDP growth.


  1. I think we cannot assume that 'all those who are not in agriculture are in services or industry'. Some are unemployed and some in unskilled jobs like 'watchman' etc !

    1. I am reporting data from the CMIE household database about the structure of the working population. So there are no unemployed here.

      Yes, everyone who is not in agriculture is doing something else - ranging from watchmen to singers. The point here is not to discuss skill levels, it is to discuss the phenomenon of the age structure of the shift from agriculture.

      The marginal worker in agriculture generates less output than the marginal worker in industry or services, so people do gain from getting out of agriculture.

  2. If less young people are in agriculture, how do you think this will impact the future of food production and food security - say 20 years from now?

    1. Indian agriculture is phenomenally inefficient. By bringing good economic principles into agriculture policy, we can get a lot more output using a lot less workers.

    2. The US is food secure with less than 5% of its population working in the agricultural sector. All that India has to do is adopt the same levels of efficiency. Even if you don't believe in American GM foods, organic vs pesticide use, etc, it should still be easily possible and actually easier to achieve than in the US, because meat production (which in turn requires more grains) is higher in US vs India, so India has lower per capita needs.

      Food wastage estimates in India are about 25% as well.

  3. the key I guess is to reform agricultural land sales, which is horribly inefficient and illiquid right now.

    if this trend of youngsters shifting away from agriculture continues, one good thing is that it will lead to consolidation of farms - and that will bring in efficiency. But for that to happen, you need to be able to sell land easily. And I think if that happens, that will accelerate the shift away from agriculture.

  4. I agree with you completely.. the problem is educated people who understand the profits from Agriculture don't have land and people who have land don't understand the potential in next 2 decades..

  5. Ajay's point - 'By bringing good economic principles into agriculture policy, we can get a lot more output using a lot less workers.' is not only a wishful thinking, but also a red flag for inflation in food categories for the next decade. Good agricultural policies will not, while high prices (as atulastra says) and higher profitability will bring in investment and improvement in agricultural effeciency.

  6. Does anyone have numbers or projections on how the economic boom in the last two decades is likely to impact GDP downwards through a decline in health via pollution, etc? Urbanization does have its woes..visiting India after 18 years, I'm struct by the loss of farmland around urban areas, the degree of traffic and pollution (was nowhere near this bad in the early nineties), and on the Agra-Delhi road, the amount of communities living with the strong smell of chemicals in the air. Lets not even start with the other gdp decreasing climate heating woes of our coal fueled economy.

    We ought to be careful about what we wish for from Agriculture too: the march towards a more efficient agriculture in the US has inevitably led to erosion enhancing crop monocultures, fertilizer runoffs creating massive dead zones in the gulf of mexico, poverty and destitution for individual farmers, overly large planting and growing usage of corn, soya for biofuels and livestock at the cost of others. And the consequences of industrial livestock farming in the US and Europe are well known...

    Efficiency to the extent of feeding the population is a fine goal, but is it good for its own sake?

    1. The estimate of loss due to pollution is anywhere from 4 to 9% of GDP in China:

      China's costly pollution problem

      "According to the Environmental Performance Index, a bi-annual report put out by Yale University and Columbia University that ranks countries based on a number of environmental factors, India ranked last out of 132 countries.

      Yet, China wasn't far off, coming in at 128. Its overall index rank, which, along with air quality, included factors such as water, agriculture, and climate change was 116. (India came in at 125)."

  7. On "shortage" of land, see:

    A hectic pace of urbanisation will generate a 5% reduction in land under agriculture. But the gains that are feasible, in agricultural yields, are of the order of 200% to 500%. Indian policy approaches in agriculture are incredibly wrong. The first order story is to break with the status quo in agricultural policy. Everything else is small change compared with this.

  8. This shift of young population, even though in some sense positive outcome, there is a caveat. Like 'brain drain'which is very well discussed and debated when we talk about an engineer or MBA settling in OECD countries, there is 'brain drain'from agriculture to urban sectors. Young generation do not see their aspirations coming true in agriculture or minimum they need they cannot derive from agriculture. But Indian agriculture is not only about surplus labor which can be productively employed in non-agriculture sector. There is scope for many improvements, innovations in agriculture and they are necessary if we want to keep food security for forthcoming decades, in the times when impacts of green revolution are over, land productivity is on decline and large urban population will depend on agriculture for food and other commodities. Who can assimilate the necessary innovations, if not young generation? This observation, of falling young population share, must sound alarms. Not all, but interested young population should be working in agriculture. It is natural from perspective of economic development to have fall in agriculture employment, but greying of agriculture labor might not be entirely welcome outcome.

    1. Indeed! In Ancient Futures: Lessons from Ladakh for a Globalizing World, Norberg-Hodge documents the change in Ladakh (which is as good a historical experiment as any) in which the younger generation is more interested in the tourist trade than agriculture. Further, agricultural innovations like jersey cows are replacing Yaks and Dzo because of greater productivity. However the cows eat up rarer grass at the 11,000 feet level as they cannot live above 15,000 where there is more moisture and thus more grass and sedge. Ecologically critical location specific information is being lost in the move to a more efficient agriculture: it has been accompanied by population growth as the means of production are divorced from the carrying capacity of the rather barren land. And as youngsters move out of agriculture, I suspect future generations will be loosing valuable connection with the knowledge of their elders.

      I dont know what the health outcome story in Ladakh is, but on my trip there in September, to Leh and surroundings, it was clear that urbanization has come to stay in a big way. And, in what is a clear sign of things to come, the lack of electricity in Leh is compensated for by the massive use of diesel generators in a town recieving 300+ days of sunshine a year. (Ledeg tells me the story for solar is much better in the villages where the grid does not necessarily penetrate).

      Does anyone know where Health outcome statistics are available in India? I know economic data are available from CMIE...

  9. I don't agree with all points made by the author.As per latest data agriculture still accounts for 15% of GDP.As mentioned by the author 31% of people are engaged in it.This is not a very high no as a proportion of GDP share.The very nature of our agriculture is very labour intensive ,because of more consumption of rice and lately increased consumption of fruits and vegetables due to rising income.Farm mechanization has worked only for certain crops like wheat,maize etc,hence the US example for very low percentage of people working in agriculture.I am sure that the author is aware of several news stories published recently about severe shortages of workers in coffee plantations,coconut plantation,and recently in rice growing regions of south india.None of these can be automated through mechanization.Plus another factor has been rising fuel cost(inspite of Rs 10 diesel subsidy).The conditions under which farm mechanization happened in US were for a different place and a different time(when fossil fuel was dirt cheap).Even today in US vegetables are harvested manually and farmers in Southern California are having tough time finding workers though they are paying hourly wage rate (12-20)$ which is higher than most retailers(like walmart)and almost comparable to what US auto workers get.Even in India farm labour rates in southern and western states are Rs200-300 which in real terms is very high because most farm workers are local hence they do not need to spend on rents plus food cost is low as they have access to PDS.The only problem is the seasonal nature of agriculture work.This has largely been addressed by the NREGA scheme.Also unlike China property rights have existed historically in India.Even the poorest of poor in rural areas have better place to live than the slums were most of the urban poor end up.


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.