Agribusiness Management · Agri-sector specialisation · also AEC-506, AEC-607 Quantitative Development Policy Analysis, Agricultural & Food Policy (IIMA), AFEPA (SLU)
The part agriculture plays in economic development, the classic theories of that role, and how price, land, credit, research, trade and subsidy policy are made and judged.
In the exams: agricultural economics PhD entrance and fellowship examinations, agricultural development and policy (heavy); agricultural development bank officer recruitment, economic and social issues (heavy); agricultural scientist eligibility tests, agricultural development and policy (heavy)
Agricultural development and policy analysis asks two linked questions. First, what part does agriculture play as a poor economy becomes a rich one: why its share of output and employment falls, how it releases labour, food, savings and foreign exchange to the rest of the economy, and why growth that starts on farms usually does more for the poor than growth that starts elsewhere. Second, how do governments act on agriculture, through prices, land, credit, research, trade, subsidies, food distribution and natural-resource rules, and how do we judge whether those actions work. The first is answered by theory built from the 1950s to the 1980s (Lewis, Fei and Ranis, Schultz, Mellor, Hayami and Ruttan); the second by measurement tools (protection coefficients, support estimates, poverty indices) and models (household, multi-market, CGE and programming).
The subject is taught the world over under different names: agricultural development and policy in Indian agricultural universities, agricultural and food policy in management schools, agricultural, food and environmental policy analysis in European programmes, and agricultural policy and law in Chinese master's papers. India is the most frequent worked example because Indian examinations test it in most detail and India runs almost every instrument in the book, but each chapter also names how the same instrument works in China, the European Union, the United States, Latin America or Africa. Policy facts are given only where certain, and dated; check the current position before relying on any scheme detail.
In examinations, this subject is the heavy second unit of the ICAR JRF/SRF agricultural economics paper, a large block of NABARD's Economic and Social Issues and Agriculture and Rural Development papers, and one named component of the ASRB NET agricultural economics syllabus. Those papers ask for theories by author and date, the mechanics of instruments, the WTO boxes, MSP cost concepts, and short calculations of protection and poverty.
Every economy that has grown rich has gone through a structural transformation: agriculture's share of output falls first, its share of employment falls more slowly, labour moves to industry and services, and the country urbanises. On the demand side, Engel's law (Ernst Engel, 1857) says that as income rises the share spent on food falls, so the income elasticity of demand for farm products is below one. On the supply side, technical change raises output per farm worker, so fewer workers feed the population. Agriculture shrinks in relative size even while its absolute output grows.
Because the output share falls faster than the employment share, farm labour productivity stays below that elsewhere for a long time, which is why rural incomes lag. Peter Timmer described four phases: agriculture generates a surplus, the surplus is transferred to the rest of the economy, agriculture is integrated with the macro economy through factor and product markets, and finally, in rich countries, it becomes a small sector that is protected rather than taxed. The World Development Report 2008, Agriculture for Development, grouped countries as agriculture-based, transforming and urbanised, and argued that the policy agenda differs across the three.
Simon Kuznets set out agriculture's contributions to development, commonly listed as four. The product contribution is food and raw materials; if food lags, food prices and hence wages rise and industrial profits are squeezed. The market contribution is farm households' demand for manufactured goods and inputs. The factor contribution is the transfer of labour and capital (savings, taxes) out of agriculture. The foreign exchange contribution is export earnings, or imports saved, which pay for capital goods.
Bruce Johnston and John Mellor (American Economic Review, 1961) listed five roles: food for a growing domestic market, foreign exchange, labour for industry, capital for investment, and rural incomes that form a market for industry. Against 1950s industry-first thinking, they argued that neglecting agriculture would choke industrialisation through food shortages and a small home market.
Since most of the poor in low-income countries are rural, growth that originates in agriculture reaches them more directly. The World Development Report 2008 summarised cross-country evidence that GDP growth originating in agriculture is at least about twice as effective in reducing poverty as growth originating elsewhere, with the advantage largest in the poorest countries. Farm income spent on local goods and services also creates rural non-farm jobs, the growth linkage stressed by Mellor.
A striking pattern in policy is that poor countries have historically taxed agriculture while rich countries protect it. Poor countries taxed farmers directly through export taxes and marketing board margins, and indirectly through industrial protection and overvalued exchange rates that lowered the domestic price of tradable farm goods. Rich countries, with small farm populations, support farm prices and incomes. Kym Anderson and Yujiro Hayami called this the development paradox: as the farm sector shrinks, it becomes politically easier for farmers to organise (fewer members, larger per-head gains) and cheaper for the rest of society to pay for support (food is a small budget share). Chapter 9 gives the evidence and Chapter 12 the political economy. The key contrast for examinations is between the 1950s view of agriculture as a passive reservoir of labour and savings to be squeezed for industry, and the post-1960s view that its own productivity growth is a precondition for development.
W. Arthur Lewis published Economic Development with Unlimited Supplies of Labour in The Manchester School in 1954, and shared the 1979 Nobel Prize with Theodore Schultz. His economy has a traditional subsistence sector, mainly family farming, where labour is so abundant that its marginal product is very low or zero, and a modern capitalist sector that hires labour for profit. Traditional-sector workers share output and receive roughly the average product, which sets an institutional wage. The capitalist sector can hire as many workers as it wants at a wage somewhat above that level (Lewis suggested a margin of about 30 per cent for the higher cost of urban living and moving), so labour supply to it is perfectly elastic.
Capitalists pay this constant wage, earn a surplus equal to the area between the marginal product of labour curve and the wage, and reinvest it. The labour demand curve shifts right, more workers are absorbed at the same wage, and profits and savings rise as a share of national income. When surplus labour is exhausted, at the Lewis turning point, further withdrawal reduces farm output, the labour supply curve turns upward and real wages rise in both sectors. The model explains why savings rates rise in early development and why wages can stay flat for decades while output grows.
John Fei and Gustav Ranis formalised Lewis in A Theory of Economic Development (American Economic Review, 1961) and Development of the Labor Surplus Economy (1964). They model agriculture explicitly and track the agricultural surplus that feeds workers who move, in three phases. In phase one, redundant labour (marginal product zero) leaves; farm output does not fall, and the food those workers ate becomes a surplus, so the average agricultural surplus per migrant equals the institutional wage. The shortage point is where marginal product becomes positive. In phase two, marginal product is positive but below the institutional wage; output falls as workers leave, the surplus per migrant falls below the institutional wage, the terms of trade turn against industry and industrial wages must rise. Workers in phases one and two together are the disguised unemployed. Phase three begins at the commercialisation point, where marginal product equals the institutional wage; from then on wages in both sectors are set by marginal product.
To avoid stagnation in phase two, farm productivity must rise alongside industrial investment, shifting the farm production function up and pushing the shortage and commercialisation points out. This balanced growth argument gives a formal reason to invest in farm technology rather than simply squeeze agriculture.
Dale Jorgenson's neoclassical model (Economic Journal, 1961) rejects zero marginal product: farm labour always has a positive marginal product and the wage is set by it. Development depends on an agricultural surplus arising from technical progress in agriculture once food per head reaches a threshold set by population growth. Where Lewis stresses capital accumulation, Jorgenson stresses farm technical change; both conclude agriculture must not be neglected.
John Harris and Michael Todaro (American Economic Review, 1970), building on Todaro (1969), explained why rural people migrate to cities even when urban unemployment is high. Migrants compare the rural wage with the expected urban wage, which is the urban wage times the probability of finding a job, roughly the employment rate. Migration continues until the rural wage equals the expected urban wage. A policy that creates urban jobs can therefore raise urban unemployment by attracting more migrants; the lasting remedy is to raise rural incomes.
Critiques of Lewis: zero marginal product has weak support (Schultz used the fall in Indian farm output after the 1918-19 influenza epidemic to argue labour was not redundant); seasonal peaks make labour short at planting and harvest; capitalists may reinvest in labour-saving machinery; and modern-sector wages are often held above the predicted level by unions, minimum wages or efficiency-wage reasons.
Theodore W. Schultz's Transforming Traditional Agriculture (1964) argued that farmers in traditional agriculture are rational and allocate the resources they have efficiently given the technology they know. This is the poor but efficient hypothesis. Traditional agriculture is an equilibrium in which farmers have used the same factors for generations, so there are no big gains left from reallocating them; the marginal return to investing more in the traditional factors is low, which is why savings and investment are low. Poverty persists not because farmers are lazy or irrational but because the returns to their existing inputs are exhausted. W. David Hopper's study of Senapur village in Uttar Pradesh, which found that farmers' input allocations were close to what profit maximisation predicts, was an important piece of evidence Schultz used.
The policy conclusion is that agriculture is transformed by supplying new, high-payoff modern inputs, mainly improved seeds, fertiliser and the knowledge to use them, and by investing in farmers' human capital through schooling and extension, not by exhortation or collectivisation. Schultz stressed that the supply of these inputs is largely a public responsibility because private firms under-invest in research. He also criticised policies that kept farm prices low, arguing that farmers respond to price incentives. Critics note that the hypothesis says farmers are efficient given their constraints; missing credit, insurance and land markets can make those constraints binding, so poor but efficient does not mean markets need no attention.
John W. Mellor, in The Economics of Agricultural Development (1966), described three stages: traditional agriculture, with static technology; technologically dynamic agriculture with low capital intensity, where output grows through yield-raising, capital-saving technology such as high-yielding varieties and fertiliser that uses abundant labour, with research, extension, input supply and credit as priorities; and technologically dynamic agriculture with high capital intensity, reached when labour becomes scarce and machines replace it. In The New Economics of Growth (1976) he proposed an employment-oriented strategy for India: farm technology raises farm incomes, farmers spend much of the gain on labour-intensive rural goods and services, those linkages create rural non-farm jobs, and cheaper food keeps real wages affordable for industry.
Ester Boserup's The Conditions of Agricultural Growth (1965) reversed the Malthusian argument. Malthus held that population outruns food and is checked by famine. Boserup argued that rising population density induces intensification: fallow shortens from forest fallow to bush fallow, short fallow, annual and multiple cropping, with more labour per hectare, manuring, irrigation and terracing. Output per hectare rises though output per hour may fall at first. Population pressure is a cause of agricultural change, an idea close to induced innovation.
The induced innovation hypothesis says that the direction of technical change responds to relative factor prices: invention economises on the factor that has become relatively expensive. John Hicks stated it in The Theory of Wages (1932). Yujiro Hayami and Vernon Ruttan applied it to agriculture in Agricultural Development: An International Perspective (1971; revised 1985). Comparing the United States and Japan from about 1880, they showed that in the United States, with abundant land and scarce labour, land was cheap relative to wages and technical change was mainly mechanical, raising land per worker. In Japan, with scarce land and abundant labour, land was dear relative to wages and technical change was mainly biological and chemical (fertiliser-responsive varieties, irrigation), raising yield per hectare. Both grew fast by paths suited to their endowments.
The mechanism runs through farmers demanding technology that saves the scarce factor, public research responding, and input suppliers following profitable markets. The metaproduction function is the envelope of all production functions that current knowledge could make available; innovation lets farmers move along it. Induced institutional innovation extends the idea: changing endowments and technology create demand for new institutions (property rights, tenancy contracts, research systems), supplied when gains exceed the cost of change.
Critics note that relative prices themselves change with technology, making tests hard, and that public research may respond to politics rather than prices. The practical lesson is that research priorities should reflect factor scarcities, and that distorting factor prices, for example subsidising machinery in a labour-abundant economy, induces the wrong kind of technology.
The Green Revolution is the name given to the spread, from the mid-1960s, of high-yielding semi-dwarf varieties of wheat and rice, used with chemical fertiliser, assured irrigation and plant protection, first in Asia and Latin America. Semi-dwarf wheats came from the Mexican wheat programme begun in 1943, later CIMMYT, where Norman Borlaug's work led to the 1970 Nobel Peace Prize. IR8, the first semi-dwarf rice from the International Rice Research Institute in the Philippines, was released in 1966. The international centres were grouped under the CGIAR, formed in 1971. The varieties were fertiliser-responsive and resisted lodging, so their advantage depended on water and nutrients.
In India, the Intensive Agricultural District Programme began in 1960-61 with a package approach in selected districts, the Intensive Agricultural Areas Programme followed in 1964, and the High Yielding Varieties Programme started in 1966-67. The institutional package that made it work was created at the same time: the Food Corporation of India and the Agricultural Prices Commission in 1965 (the latter renamed the Commission for Agricultural Costs and Prices in 1985), assured procurement at announced prices, and expanded institutional credit and input supply. Wheat output rose fastest in Punjab, Haryana and western Uttar Pradesh. Robert Evenson and Douglas Gollin's 2003 review in Science attributed a large share of yield growth in developing countries between 1960 and 2000 to modern varieties, and found that the gains came later and were smaller in sub-Saharan Africa, where the crops, rainfed conditions and institutions were different.
The Green Revolution has been criticised on equity, regional, environmental and nutritional grounds. On interpersonal equity, early studies (for example Keith Griffin's The Political Economy of Agrarian Change, 1974) argued that larger farmers adopted first because they had better access to credit, irrigation and information, and could use the gains to buy out or evict tenants. Later evidence generally found that the technology was scale-neutral (a small plot yields as much per hectare as a large one) but not resource-neutral: benefits depended on access to water, credit and inputs, so small farmers adopted later but did adopt where those were supplied, and landless labourers gained where labour demand rose with double cropping. Poor food buyers gained from lower real cereal prices.
On regional inequality, irrigated areas gained much more than rainfed and dryland areas, and eastern India and Africa lagged. On the environment, the package encouraged rice-wheat monoculture, heavy groundwater pumping (made worse by free or flat-rate electricity), falling water tables in north-west India, soil nutrient imbalances from subsidised nitrogen, pesticide overuse and loss of crop and varietal diversity. Cheaper cereals also displaced pulses and coarse grains in some areas. The balanced view in the literature is that the Green Revolution averted large food deficits and reduced poverty, while leaving second-generation problems of sustainability and regional imbalance that policy must now address.
Indian agricultural growth is usually divided into phases: area expansion and institutional change before the mid-1960s; the Green Revolution phase to about 1980, concentrated in wheat and irrigated north-west India; broad-based growth in the 1980s as the technology spread to rice in the east and south; a slowdown from the mid-1990s to the early 2000s linked to falling public investment and slowing yields; and recovery from about 2004-05 led by horticulture, livestock and fisheries. Rates differ with the data series and base year, so quote trends and reasons rather than one figure.
Structural features: holdings are small and shrinking; the Agricultural Census 2015-16 put the average at about 1.08 hectares, with marginal and small holdings (below 2 hectares) about 86 per cent of all holdings. About half the net sown area is irrigated, so rainfall still drives output. Foodgrain output first exceeded 300 million tonnes around 2020-21. India has been the largest milk producer since the late 1990s, helped by Operation Flood (from 1970). Agriculture's share in gross value added is well under a fifth while it employs a little under half the workforce.
Shenggen Fan, Peter Hazell and Sukhadeo Thorat found that extra public spending on rural roads and agricultural research gave the highest growth and poverty returns per rupee, with input subsidies among the lowest; Ashok Gulati and Sudha Narayanan's The Subsidy Syndrome in Indian Agriculture (2003) argued that rising subsidies had crowded out public investment. The policy question has shifted from raising cereal output to raising farm incomes, diversifying, and managing water and soil.
Farm prices are unstable because supply depends on weather and is fixed once a crop is sown, while demand is price-inelastic, so a small change in quantity causes a large change in price. Producers respond to last season's price (the cobweb model), which can make cycles worse. Instability hurts poor producers who cannot insure and poor consumers who spend a large share of income on food. Price policy therefore has several, sometimes conflicting, objectives: give producers an assured price that encourages investment and adoption of technology; protect consumers from price spikes; stabilise prices within a band; maintain buffer stocks for food security; and shift the terms of trade between agriculture and industry. The food price dilemma, named by Timmer, Falcon and Pearson in Food Policy Analysis (1983), is that high prices help farmers and hurt consumers, and low prices do the reverse.
Instruments include support prices backed by government purchase (procurement), buffer stocks with release prices, deficiency or price-gap payments that pay the farmer the difference between a target and the market price without buying the crop, variable import levies and export restrictions, and market intervention schemes for perishables. Each has a fiscal cost and different effects on market prices. A support price with procurement raises the market price for everyone and requires storage and disposal; a deficiency payment leaves the market price to find its level, so consumers pay less while the budget pays the gap.
India announces minimum support prices on the recommendation of the Commission for Agricultural Costs and Prices, which was set up as the Agricultural Prices Commission in 1965. The CACP recommends MSPs for 22 mandated crops (kharif and rabi cereals, pulses and oilseeds, and cotton, jute and copra) and the fair and remunerative price for sugarcane; the final decision is taken by the Union Cabinet. The CACP considers cost of production, demand and supply, domestic and international prices, inter-crop parity, terms of trade and effects on consumers.
Cost of cultivation data come from the Comprehensive Scheme for studying cost of cultivation. The cost concepts are: A1, all actual paid-out costs by an owner-operator (hired human, animal and machine labour, seed, fertiliser, manure, pesticides, irrigation charges, depreciation on implements and farm buildings, land revenue, interest on working capital and miscellaneous expenses); A2, which is A1 plus rent paid for leased-in land; A2+FL, which is A2 plus the imputed value of family labour; B1, which is A1 plus interest on the value of owned fixed capital excluding land; B2, which is B1 plus the rental value of owned land and rent paid for leased-in land; C1, which is B1 plus imputed family labour; C2, which is B2 plus imputed family labour; C2*, where family labour is valued at the higher of the market or statutory wage; and C3, which is C2* plus 10 per cent for the farmer's management.
The National Commission on Farmers (2004-2006), chaired by M. S. Swaminathan, recommended that the MSP be at least 50 per cent above the weighted average cost of production, which farm groups read as C2. The Union Budget 2018-19 announced that MSPs would be kept at least one and a half times the cost of production, and the government applied this from kharif 2018 using A2+FL as the cost base. The gap between A2+FL and C2 is the imputed rent and interest on owned land and capital, so the two bases give very different MSPs.
An MSP is effective only where the government buys. In India effective procurement has been concentrated in wheat and rice in a few states, so many farmers of other crops or in other states sell below MSP. Open-ended procurement has at times built stocks well above buffer norms and tilted cropping toward rice and wheat. PM-AASHA (2018) combined a price support scheme for pulses and oilseeds, a price deficiency payment scheme and a pilot private procurement scheme. Madhya Pradesh's Bhavantar Bhugtan Yojana (2017) paid farmers the gap between the MSP or a modal price and the market price; traders could depress market prices during the scheme window, a general risk of deficiency payments.
The EU's Common Agricultural Policy (from 1962) used intervention prices, variable import levies and export refunds, producing surpluses in the 1970s and 1980s; the MacSharry reform of 1992 cut support prices and compensated farmers with area payments, and the 2003 reform decoupled most payments from production. The United States used target prices, marketing loans and deficiency payments; 1996 legislation introduced decoupled production flexibility contract payments, and the 2014 farm bill replaced direct payments with price loss coverage and agriculture risk coverage. China introduced minimum purchase prices for rice (2004) and wheat (2006) and later temporary stockpiling for maize, whose large stocks led to reforms from 2016 moving maize to market pricing with producer subsidies. In Africa, maize price support through Kenya's National Cereals and Produce Board and Zambia's Food Reserve Agency crowded out private trade.
Land is the main asset of rural households, and who owns it, who operates it and on what terms affects efficiency, equity, investment and access to credit. Land policy covers ownership distribution (redistributive reform), tenancy terms, consolidation of fragmented plots, land records and titles, land acquisition, and land-use regulation.
Two economic arguments are central. The first is the inverse farm size-productivity relationship, noted in India's Farm Management Studies of the 1950s and discussed by Amartya Sen in 1962: output per hectare tends to fall as farm size rises. Explanations include the lower supervision cost of family labour compared with hired labour, more intensive use of labour on small farms because family labour has a low opportunity cost, differences in soil quality, and measurement error. If the relation holds, redistribution to small farmers can raise both equity and output. Its strength has weakened in some places with mechanisation and non-farm jobs.
The second concerns tenancy. Alfred Marshall argued that sharecropping is inefficient because the tenant bears the full cost of extra effort but receives only a share of the extra output, so he under-applies inputs: Marshallian inefficiency. Steven Cheung (1969) argued that if the landlord can monitor and specify inputs, share tenancy can be as efficient as fixed rent or owner cultivation. Later theory explains sharecropping as a trade-off between incentives and risk sharing under imperfect information and missing credit and insurance markets. The policy implication is that insecure tenancy and high shares reduce investment, which is the case for tenancy reform.
Indian land reform after independence had four main components. The abolition of intermediaries (zamindari and similar tenures), largely carried out in the early 1950s, brought cultivators into direct relation with the state. Tenancy reform aimed at security of tenure, regulation of rent (often a ceiling of one-fourth to one-fifth of produce) and conferring ownership on tenants. Ceilings on landholdings, legislated by states in the 1960s and tightened under national guidelines in 1972, aimed to take surplus land above the ceiling and distribute it to the landless. Consolidation of holdings, most successful in Punjab, Haryana and western Uttar Pradesh, reduced fragmentation. Land is a state subject, so the results differ widely by state.
Results were mixed. Intermediary abolition was largely achieved. Ceilings yielded little surplus land relative to expectations because of benami transfers, family partitions and exemptions. Tenancy laws often drove tenancy underground, since landlords resumed land or used oral leases, leaving tenants with even less security. West Bengal's Operation Barga (1978), which recorded sharecroppers' names so they gained protection from eviction and a legal share, is a well-studied success; Abhijit Banerjee, Paul Gertler and Maitreesh Ghatak (Journal of Political Economy, 2002) found it raised productivity. Kerala's reforms of 1969-70 conferred ownership on a large number of tenants. Recent policy has turned to legalising and securing leasing: NITI Aayog's Model Agricultural Land Leasing Act of 2016 proposed recognising tenants to give them access to credit and insurance while protecting owners' title. Computerisation of land records under the Digital India Land Records Modernization Programme aims at conclusive titles.
East Asian reforms are the textbook successes. Japan's 1946 reform under the Allied occupation bought land from absentee and large landlords and sold it to tenants. Taiwan reduced rents to 37.5 per cent of the main crop (1949), sold public land (1951) and carried out land-to-the-tiller (1953), compensating landlords partly with shares in state enterprises. South Korea's reform around 1950 was similar. These produced egalitarian owner-operated smallholder farming credited with supporting rapid, equitable growth.
China collectivised land in the 1950s and from 1978 introduced the Household Responsibility System, under which collectively owned land is contracted to households who keep output above a quota. Justin Yifu Lin (American Economic Review, 1992) attributed about half of output growth in 1978-84 to this change in incentives.
Latin American reforms (Mexico's ejido, Bolivia 1953, Peru and Chile in the 1960s-70s) were large but often followed by reversals or low productivity where support services were lacking. Market-assisted land reform, promoted by the World Bank in the 1990s (Brazil's pilots, South Africa's willing-buyer willing-seller approach after 1994 with an early target of transferring 30 per cent of commercial farmland), relies on voluntary sales financed by grants and loans and has generally been slow. Zimbabwe's compulsory fast-track reform from 2000 brought severe short-run losses in commercial output. Redistribution works best with secure rights, credit, services and smallholder-friendly technology.
Farmers need credit for seasonal inputs, for investment in wells, machinery and livestock, and for consumption smoothing. Rural credit markets are prone to failure for identifiable reasons. Lenders cannot easily observe borrowers' riskiness (adverse selection) or effort and use of funds (moral hazard); enforcing repayment is costly; collateral is limited, especially for tenants without land titles; loans are small and borrowers scattered, so transaction costs per rupee are high; and farm incomes are covariant, since a drought hits all borrowers in a region at once. Joseph Stiglitz and Andrew Weiss (American Economic Review, 1981) showed that with imperfect information lenders may ration credit rather than raise interest rates, because higher rates attract riskier borrowers and so can reduce expected returns.
Informal lenders (moneylenders, traders, landlords, relatives) overcome information and enforcement problems through personal knowledge and interlinked contracts, for example tying credit to the sale of output or to labour services, but often at high interest rates that reflect risk, cost and sometimes market power. The policy question is whether and how the state should intervene to expand formal credit.
The All India Rural Credit Survey (1951-52; report 1954) found that cooperatives supplied only about 3 per cent of rural borrowing and moneylenders about 70 per cent, and concluded that cooperation has failed, but cooperation must succeed. The multi-agency approach that followed built three channels. Cooperative credit, with a short-term three-tier structure of primary agricultural credit societies, district central cooperative banks and state cooperative banks. Commercial banks, after nationalisation of 14 banks in 1969 (and six more in 1980), the lead bank scheme (1969) and priority sector targets that require 18 per cent of adjusted net bank credit to go to agriculture, with a sub-target for small and marginal farmers. And Regional Rural Banks, set up from 1975.
NABARD was established in 1982 as the apex development bank for agriculture and rural development, refinancing lenders and managing rural infrastructure funds. The Kisan Credit Card (1998) gave farmers a revolving credit line based on scale of finance per hectare. The SHG-bank linkage programme, piloted by NABARD in 1992, lends to women's self-help groups. Interest subvention lowers the effective rate on short-term crop loans, with an extra incentive for prompt repayment.
Debt surveys show the institutional share of rural debt rose sharply after 1969, but moneylenders remain important for small farmers and tenants. Debt waivers, such as the national scheme of 2008 and several state waivers since 2017, give short-run relief but are criticised for moral hazard, for benefiting those who already had formal loans, and for reducing later lending.
From the 1950s to the 1970s many governments and donors ran supply-led agricultural credit programmes through state development banks with subsidised interest rates and targets. The Ohio State University school (Dale Adams, Douglas Graham and J. D. Von Pischke, Undermining Rural Development with Cheap Credit, 1984) argued that such programmes failed: subsidised rates attracted larger, better-connected borrowers, caused rationing by influence, discouraged savings mobilisation, and produced high default and financially unsustainable institutions. Their prescription was financial markets that mobilise savings and charge cost-covering rates.
Microfinance offered a different response. The Grameen Bank in Bangladesh, founded by Muhammad Yunus (Nobel Peace Prize 2006), used group lending with joint liability, small frequent repayments and progressive loan sizes to lend to poor women without collateral. Critiques of microfinance note modest average effects on income in randomised evaluations, high effective interest rates, and the poor fit between weekly repayment and seasonal farm cash flows. Current thinking emphasises warehouse receipt finance, value-chain finance, index insurance linked to credit, and secure land and lease records as collateral.
Much agricultural knowledge is a public good: it is non-rival (one farmer's use of an improved practice does not reduce another's) and often non-excludable (a self-pollinated variety can be saved and replanted, a management practice copied). Private firms cannot capture enough of the benefit, so they under-invest. The private sector does invest where appropriability is high, as in hybrid seed (farmers must buy fresh seed each season), agrochemicals, machinery and, with patents, biotechnology traits. Public research is therefore concentrated on self-pollinated crops, natural resource management, livestock and basic science, and in countries and crops where markets are small.
Measured returns to agricultural research are very high. Zvi Griliches's study of hybrid corn in the United States (Journal of Political Economy, 1958) was the pioneering estimate of social returns. Julian Alston and colleagues' meta-analysis (2000) of hundreds of studies found rates of return that are typically far above the cost of public funds, while warning of methodological biases. The implication, repeated by many reviews, is that agricultural research is under-funded relative to its payoff. Research takes time: lags from investment to adoption and full impact run to a decade or more, and benefits persist and then depreciate, which is why short-horizon budgeting under-provides research.
The standard way to evaluate research is the economic surplus approach set out by Alston, Norton and Pardey in Science Under Scarcity (1995). A new technology lowers unit cost and shifts the supply curve down or right. The resulting change in consumer and producer surplus, computed each year, is the gross annual benefit. Its size depends on the size of the shift (the cost reduction per unit), the adoption rate over time, the quantity produced, and the elasticities of supply and demand. The distribution depends on the elasticities: with inelastic demand, as for staple foods in a closed economy, much of the benefit passes to consumers through lower prices; for a small open economy facing a fixed world price, the price does not fall, so domestic consumers gain nothing and producers capture all of the gain. Annual benefits are then set against research and extension costs and discounted to give net present value, benefit-cost ratio and internal rate of return. The same logic explains Cochrane's treadmill: early adopters gain, but as adoption spreads, prices fall and late adopters or non-adopters lose, pushing marginal farmers out.
The United States built its system on the land-grant universities (Morrill Act 1862), state agricultural experiment stations (Hatch Act 1887) and the Cooperative Extension Service (Smith-Lever Act 1914), linking teaching, research and extension; India's state agricultural universities from the 1960s were modelled on it. India's system is led by the Indian Council of Agricultural Research, founded in 1929 as the Imperial Council of Agricultural Research; Brazil's by Embrapa (1973), credited with making the Cerrado productive. The CGIAR centres (for example CIMMYT, IRRI, ICRISAT, IFPRI) supply international public goods: germplasm, breeding lines and policy research. Research intensity (public research spending as a share of agricultural GDP) is much higher in rich than poor countries.
Extension has moved through models: the training and visit system promoted by the World Bank from the 1970s (fixed schedules of visits to contact farmers), which proved costly and was largely abandoned; decentralised, demand-driven and pluralistic extension, such as India's Agricultural Technology Management Agency (ATMA) at district level from 2005; farmer field schools; and digital advisory services.
Intellectual property shapes private incentives. The UPOV Convention (1961) provides plant breeders' rights. The WTO TRIPS Agreement, Article 27.3(b), requires members to protect plant varieties by patents, an effective sui generis system, or both. India's sui generis law, the Protection of Plant Varieties and Farmers' Rights Act, 2001, protects breeders while explicitly preserving farmers' rights to save, use, exchange and sell farm-saved seed (not branded seed of a protected variety). Stronger protection increases private investment but raises seed prices and concerns over access.
Analysts compare domestic prices with an undistorted benchmark, usually the border price (the world price at the relevant point, adjusted for transport, handling and quality). The nominal protection coefficient (NPC) is domestic price divided by border price; above one means producers are protected, below one that they are taxed. The effective rate of protection (ERP) compares value added (output value minus tradable input costs) at domestic and border prices: ERP = (VA at domestic prices minus VA at border prices) / VA at border prices. It captures protection on output and taxes or subsidies on inputs together.
The OECD producer support estimate (PSE) is the annual value of gross transfers from consumers and taxpayers to producers, at the farm gate: market price support (price gap times quantity) plus budgetary payments. The percentage PSE is PSE divided by gross farm receipts including support. The Policy Analysis Matrix of Eric Monke and Scott Pearson (1989) sets revenues, tradable input costs, domestic factor costs and profits at private and social prices; from it come the effective protection coefficient and the domestic resource cost ratio, DRC = domestic factor cost at social prices / (revenue minus tradable inputs, at social prices). A DRC below one indicates comparative advantage.
Anne Krueger, Maurice Schiff and Alberto Valdes, in a World Bank study of 18 developing countries (1988), found agriculture was taxed on average, and that indirect taxation through industrial protection and overvalued exchange rates exceeded direct taxation through export taxes and marketing boards. Kym Anderson's distortions project (2009) found this bias declined substantially after the 1980s.
The Uruguay Round Agreement on Agriculture took effect in 1995 with three pillars. Market access required tariffication (converting quotas and other non-tariff barriers into tariffs), binding and reduction: developed countries by an average 36 per cent with at least 15 per cent per line over six years, developing countries by 24 per cent with at least 10 per cent per line over ten years; least-developed countries made no reduction commitments.
Domestic support is classified in boxes. The amber box is trade-distorting support measured by the aggregate measurement of support (AMS), which developed countries cut by 20 per cent over six years and developing countries by 13.3 per cent over ten. Market price support in the AMS is the gap between the administered price and a fixed external reference price based on 1986-88, times eligible production. Support below de minimis, 5 per cent of value of production for developed and 10 per cent for most developing countries (product-specific and non-product-specific counted separately), is exempt. The blue box covers direct payments under production-limiting programmes. The green box (Annex 2) covers publicly funded support with no or minimal trade distortion and no price support: research, extension, infrastructure, domestic food aid, public stockholding for food security, decoupled income support, disaster relief and environmental programmes. Article 6.2, the development box, exempts investment subsidies and input subsidies to low-income or resource-poor producers in developing countries.
Export subsidies were cut (developed countries 36 per cent in value, 21 per cent in volume), and the Nairobi Ministerial (2015) decided to eliminate them. Because procurement at administered prices counts in the AMS against a 1986-88 reference price, the Bali Ministerial (2013) adopted an interim peace clause for developing countries' public stockholding, which the General Council in 2014 extended until a permanent solution is agreed.
Input subsidies, on fertiliser, electricity, irrigation water, seed and credit, are the main form of support in many developing countries. They can speed adoption of new inputs when farmers are unfamiliar with them or credit-constrained, and are politically popular. Their problems are fiscal cost, capture by larger farmers and by suppliers, diversion and smuggling, overuse and environmental damage when prices are far below cost, and crowding out of public investment. India controls the maximum retail price of urea and has since April 2010 used a nutrient-based subsidy for phosphatic and potassic fertilisers, a fixed subsidy per kilogram of nutrient; the price gap this leaves between nitrogen and other nutrients is blamed for imbalanced application. Malawi's input subsidy programme, begun in the 2005/06 season using vouchers, is the leading African case: it raised maize output in its early years but became costly and suffered from targeting and timing problems. Smart subsidies are designed to be temporary, targeted, market-friendly (vouchers redeemable at private dealers) and linked to an exit strategy.
The trend in many countries is toward direct transfers that are decoupled from input use. India's PM-KISAN, announced in February 2019, pays Rs 6,000 a year in three instalments to landholding farmer families; Telangana's Rythu Bandhu (2018) paid a per-acre investment support. China abolished its agricultural tax nationwide in 2006 and introduced direct grain subsidies. Direct transfers avoid distorting input use but raise questions of targeting (owners versus tenants), and do little to fix underinvestment in public goods.
The World Food Summit of 1996 defined food security as existing when all people, at all times, have physical and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life. Four dimensions are used: availability (production, stocks and trade), access (the income and prices that let households obtain food), utilisation (diet quality, safe water, sanitation, health, so the body can use food) and stability (all three over time, against shocks). Nutrition security adds the emphasis on diet quality and care, and on micronutrient deficiencies, the hidden hunger that can coexist with enough calories.
Amartya Sen's Poverty and Famines (1981) changed the analysis of famine. Studying the Bengal famine of 1943 and others, he showed that famines can occur without a significant fall in food availability; people starve when their entitlements, the bundles of food they can command through production, trade, labour and transfers, collapse, for example when wages of rural labourers fall relative to food prices. Famine prevention therefore depends on protecting entitlements through employment and transfers, and on information and political accountability. Measures of food insecurity include the FAO prevalence of undernourishment, the Food Insecurity Experience Scale, and anthropometric measures of stunting, wasting and underweight in children.
India's public distribution system grew from wartime rationing into a system in which the Food Corporation of India procures grain at MSP, stores it and allocates it to states, which distribute it through fair price shops. The universal system gave way to the Revamped PDS (1992) and the Targeted PDS (1997), with different prices for households below and above the poverty line; Antyodaya Anna Yojana (2000) targeted the poorest. The National Food Security Act, 2013 made access a legal entitlement: it covers up to 75 per cent of the rural and 50 per cent of the urban population, gives priority households 5 kg of grain per person per month at 3, 2 and 1 rupees per kg for rice, wheat and coarse grains, and gives AAY households 35 kg per household per month; it also covers child nutrition schemes and maternity benefits. Since 1 January 2023 grain under the Act has been free. Studies documented large leakages in the 2000s that fell in reforming states such as Chhattisgarh.
Other models: the United States runs the Supplemental Nutrition Assistance Program (food stamps until 2008), an electronic benefit spent at retailers. Mexico's PROGRESA (1997) and Brazil's Bolsa Familia (2003, within Fome Zero) are conditional cash transfers linked to schooling and health visits; PROGRESA's randomised phase-in made it a model for impact evaluation. Ethiopia's Productive Safety Net Programme (2005) gives food or cash for public works. Egypt subsidises baladi bread. Design debates are cash versus in-kind (cash is cheaper and gives choice; in-kind is self-targeting and protects against food inflation), targeting errors of exclusion and inclusion, and the fiscal and market effects of large public procurement.
Absolute poverty is measured against a poverty line, the cost of a minimum basket. The Foster-Greer-Thorbecke (1984) family is P-alpha = (1/N) times the sum over the poor of ((z - y)/z) to the power alpha, where z is the line and y income. With alpha = 0 it is the headcount ratio; with alpha = 1 the poverty gap index, measuring depth; with alpha = 2 the squared poverty gap, measuring severity by weighting the poorest more. The World Bank's international extreme poverty line was 1.90 dollars a day (2011 PPP), 2.15 dollars (2017 PPP) from 2022, and 3.00 dollars (2021 PPP) from June 2025. In India, expert groups (Lakdawala, Tendulkar, Rangarajan) revised national lines; the Tendulkar method gave 21.9 per cent poverty for 2011-12.
Inequality is shown by the Lorenz curve (cumulative income share against cumulative population share) and the Gini coefficient, the area between the Lorenz curve and the equality line divided by the whole triangle below that line, from 0 (equality) to 1. Simon Kuznets (1955) proposed that inequality first rises then falls with income, the inverted-U, which later data support only weakly.
Montek Ahluwalia (1978) found that rural poverty in India fell with higher agricultural output per head; Martin Ravallion and Gaurav Datt (1996) found that primary and tertiary sector growth reduced poverty far more than manufacturing growth. Public works (Maharashtra's Employment Guarantee Scheme of the 1970s and MGNREGA of 2005, guaranteeing 100 days of wage work per rural household per year) act as self-targeting safety nets and raise rural wages.
Groundwater is a common-pool resource: one farmer's pumping lowers the water table for neighbours, but excluding others is costly. Garrett Hardin's tragedy of the commons (1968) predicts over-extraction. Elinor Ostrom's Governing the Commons (1990), for which she shared the 2009 Nobel Prize, showed that communities often manage commons sustainably through local rules, monitoring and graduated sanctions.
In India, many states supply farm power free or at flat rates by pump horsepower, so the marginal cost of pumping is near zero, encouraging over-extraction, paddy in dry regions and utility losses. Gujarat's Jyotigram Yojana (from 2003) separated farm and non-farm feeders, giving households continuous power while rationing farm power to fixed hours. Punjab's Preservation of Subsoil Water Act, 2009 prohibits transplanting paddy before a notified date to cut evaporation losses; the later harvest left less time before wheat sowing, a factor in residue burning. Solar pumps (PM-KUSUM, 2019) can worsen over-extraction unless surplus power can be sold to the grid.
Canal water charges in many developing countries are levied per hectare by crop, not by volume, and are too low to cover maintenance, creating a circle of poor service and low willingness to pay. Volumetric pricing encourages efficiency but needs measurement. Participatory irrigation management hands minors to water user associations; Andhra Pradesh's Farmers' Management of Irrigation Systems Act (1997) was an early large example. Head-reach farmers take more than tail-enders, which rotation rules such as warabandi try to correct. Formal water markets (Australia's Murray-Darling Basin, Chile) move water to higher-value uses where rights are transferable.
Agriculture produces negative externalities (nitrate leaching, eutrophication from phosphorus runoff, pesticide residues, soil erosion, greenhouse gas emissions from enteric fermentation, rice paddies, fertiliser and land clearing, and air pollution from residue burning) and positive ones (landscape, biodiversity, watershed services, carbon storage). Where the externality is not priced, output of the polluting activity exceeds the social optimum. Policy instruments are regulation (standards, bans, the EU Nitrates Directive of 1991 with nitrate vulnerable zones), Pigouvian taxes equal to marginal external damage at the optimum, removal of input subsidies that encourage overuse, tradable permits, and payments for ecosystem services, as in Costa Rica's national programme from 1997 that pays landholders for forest conservation. The EU links direct payments to environmental rules through cross-compliance and funds voluntary agri-environment schemes; US conservation programmes pay farmers to retire erodible land (Conservation Reserve Program, 1985).
Climate policy for agriculture combines adaptation (drought-tolerant varieties, irrigation efficiency, insurance, diversification) and mitigation (improved livestock feeding, alternate wetting and drying in rice, better nitrogen use, agroforestry). Because input subsidies, price support and free power jointly shape resource use, reforms are usually packaged rather than piecemeal.
Policy is often taught as a cycle: agenda setting, analysis of options, decision, implementation, monitoring and evaluation, and revision. Real policy is shaped by interests and institutions. Mancur Olson's The Logic of Collective Action (1965) explains why small groups with concentrated stakes (farmers in rich countries, industrial lobbies or urban consumers in poor ones) organise better than large groups with diffuse interests. Michael Lipton's Why Poor People Stay Poor (1977) argued that urban bias, the concentration of political power in cities, led developing countries to neglect and tax agriculture. Robert Bates's Markets and States in Tropical Africa (1981) showed how African governments used marketing boards and overvalued currencies to tax export crops and keep urban food cheap, rewarding political supporters with subsidised inputs and credit.
In India, agriculture is a state subject under the Seventh Schedule, while inter-state trade, food procurement, and many schemes involve the Union, so policy is negotiated across levels. The three farm laws enacted in 2020 to liberalise trade outside regulated markets, contract farming and stockholding limits were repealed in 2021 after prolonged farmer protests, a reminder that economic analysis must account for trust, consultation and the distribution of gains.
Implementation fails for predictable reasons: principal-agent problems between policymakers and the officials who deliver, weak targeting, leakage and corruption, delays, and the capture of benefits by better-off groups. Delivery reforms include direct benefit transfer to bank accounts, social audits (used in MGNREGA) and digitised beneficiary records.
Evaluation asks whether a policy caused an outcome. Ex ante appraisal uses cost-benefit analysis and simulation models; ex post evaluation compares outcomes with a counterfactual. Methods include randomised controlled trials (PROGRESA's phased roll-out is the classic case), difference-in-differences (comparing changes over time in treated and untreated groups), regression discontinuity (using an eligibility cut-off), instrumental variables and propensity score matching. A logical framework or theory of change sets out inputs, activities, outputs, outcomes and impact with indicators and assumptions. Good evaluation also reports cost-effectiveness and who gains, and it asks about external validity, that is, whether a result from one region, season or delivery agency would hold elsewhere.
Most farms in developing countries are households that both produce and consume. The agricultural household model (Singh, Squire and Strauss, Agricultural Household Models, 1986) combines a production decision (maximise farm profit) with a consumption decision (maximise utility from goods, own food and leisure subject to a full-income constraint). When all markets exist and the household is a price-taker, the model is separable or recursive: production decisions depend only on prices and technology, and are made first; profit then enters full income and consumption is chosen. A key implication is the profit effect: a rise in the price of the staple the household produces raises its profit, so its own consumption of that staple can rise even though the price went up, unlike a pure consumer. When markets are missing or transaction costs create a price band (de Janvry, Fafchamps and Sadoulet, 1991), the household uses a shadow price, separability breaks, and supply response to price can be small. Chayanov's earlier theory of the peasant family farm (1925), balancing the drudgery of work against family consumption needs, is the ancestor of these models.
Partial equilibrium welfare analysis uses consumer and producer surplus to measure gains and losses and deadweight loss from a policy in one market. Multi-market models (Avishay Braverman and Jeffrey Hammer, 1986) extend this to a small set of linked markets (for example rice, wheat, fertiliser and labour), with cross-price elasticities and a budget account, to capture spillovers of a price or subsidy change while staying tractable.
Computable general equilibrium (CGE) models represent the whole economy: producers, households, government and the rest of the world, linked through markets that clear. They are calibrated to a social accounting matrix (SAM), a square table in which each account's row total (receipts) equals its column total (expenditures). Model results depend on elasticities and on closure rules, the choice of which variables are fixed and which adjust to balance savings and investment, the government budget and the foreign account. Widely used frameworks include the IFPRI standard model (Lofgren and colleagues, 2002) and the GTAP global trade model (Hertel, 1997), used to analyse trade liberalisation and WTO scenarios.
Mathematical programming models (linear programming of representative farms or regions) optimise an objective such as net returns subject to land, labour, water and capital constraints, and simulate responses to new technology or prices. Positive mathematical programming (Richard Howitt, 1995) calibrates such models to exactly reproduce observed cropping patterns, and is widely used in EU policy analysis. Each approach trades behavioural or technical detail for economy-wide scope.
Learn each theory by its mechanism, not only its author: be able to draw the Lewis labour supply curve, mark the Fei-Ranis shortage and commercialisation points, and show the US-Japan contrast in induced innovation. Memorise the CACP cost concepts in order and the WTO boxes with their examples, because these recur in every Indian agricultural economics paper. Practise the short calculations that appear often: NPC, ERP, %PSE, DRC, MSP from a cost base, the FGT indices and a two-group Gini, and the Harris-Todaro equilibrium. Keep a dated list of Indian schemes (1965, 1969, 1975, 1982, 1997, 1998, 2000, 2005, 2013, 2018, 2019) and one foreign counterpart for each.
In a bank, cooperative, processor or government department, you will use this subject to read the policy environment. A credit officer needs to know how interest subvention, priority sector targets and debt waivers change repayment behaviour. A procurement manager needs to know when MSP operations or export restrictions will move market prices. A programme officer designing an input or transfer scheme needs to anticipate targeting errors and leakages and to set up a baseline and comparison group so the scheme can be evaluated. The core practical skill is a simple incidence analysis: who gains, who pays, and how behaviour changes.
At senior level the task is advising on policy packages and anticipating their political economy. A senior official or chief economist weighs the fiscal cost of price support and input subsidies against public investment in roads, research and irrigation, using evidence that the latter give higher returns. They must judge WTO exposure of procurement programmes, the interaction of power, price and water policies in groundwater-stressed regions, and how to reform without losing farmer trust, a lesson from the repeal of India's 2020 farm laws. They design reforms in sequence, with compensation for losers and credible monitoring.
A researcher with long experience builds and critiques the quantitative tools: estimating supply response and household behaviour under missing markets, calibrating multi-market and CGE models with credible elasticities and closure rules, measuring returns to research with proper attribution and lags, and designing rigorous evaluations of land, credit, subsidy and food programmes. They also test the theories against new data, for example whether China or parts of India have passed a Lewis turning point, whether the inverse size-productivity relation survives measurement correction, and how climate change alters the case for public stockholding and input subsidies.
Gives policy facts only where certain and dated; current MSP levels, subsidy rates, scheme budgets and procurement figures change every year and are deliberately left out, as are exact growth rates by period, which differ by data series. Country coverage outside India, the EU, the US and China is at the level of named examples, not full policy histories. The PhD modelling chapter explains structure and assumptions but does not work through the algebra of a CGE or household model, estimation of supply response, or the numerical solution of programming models. Recent reforms after 2025 (e.g. any change to PM-KISAN, NFSA pricing, CAP after 2027 or the US farm bill) are not covered.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.