Agribusiness Management · Agri-sector specialisation · also Fiscal economics, AICE-JRF Ag Econ Unit 3
How governments raise and spend money, who really bears a tax, and how budgets, fiscal policy and public debt serve development.
In the exams: agricultural economics PhD entrance and fellowship examinations, public finance (present)
Public finance studies how governments raise money, how they spend it, and what those choices do to efficiency, equity and growth. It asks why some goods must be supplied through the state at all, who really pays a tax once prices adjust, how large a government should be, how a budget is read and its deficit measured, how tax powers and spending duties are divided between national and local governments, and when public borrowing helps or harms.
For agriculture the subject is not abstract. In most developing countries farm income is lightly taxed directly but heavily affected by indirect taxes, export duties and input subsidies. Fertiliser, power, irrigation water and credit subsidies, minimum support prices and public procurement, direct income transfers to farmers, and crop insurance premium support are among the largest items in many budgets. Understanding incidence, deadweight loss and the fiscal cost of such programmes is a core skill for an agricultural economist, a bank officer appraising rural schemes, or a policy analyst in a ministry.
The subject is examined as Unit 3 (Public Finance and International Economics) of the ICAR AICE-JRF/SRF Agricultural Economics paper and at the same depth in ASRB NET, and as a unit of UGC-NET Economics. Applied questions on deficits, budget terms, fiscal federalism and subsidies appear in the Economic and Social Issues papers of NABARD Grade A and RBI Grade B. This manual covers the eight syllabus units in twelve chapters with worked numericals and examples from several countries. Figures for any particular country's current budget are deliberately not given, because they change every year; the student should read them from the latest official budget documents.
Public finance, also called fiscal economics or public economics, is the study of the revenue, expenditure, borrowing and budgeting of governments and of the economic effects of these activities. The word fiscal comes from the Latin fiscus, the treasury. Dalton defined it as the study of the income and expenditure of public authorities and the adjustment of one to the other; modern texts widen it to include any government action that changes the allocation of resources or the distribution of income, including regulation, tax expenditures and state enterprises.
The public sector includes the central or federal government, state or provincial governments, local bodies, and in a broad definition public enterprises and social security funds. National accounts usually separate general government (which supplies mostly non-market services financed by taxes) from public corporations (which sell output at economically significant prices). Cross-country comparisons of deficits therefore depend on which of these bodies are included.
Public and private finance are alike in that both deal with scarce means, both must reconcile income with spending, and both may borrow. The differences, which examiners ask about repeatedly, are these. First, adjustment of income and expenditure: a household generally fits its spending to its income, while a government first decides what it must spend and then finds the revenue, although this contrast is only a tendency, since modern governments also operate under budget constraints and fiscal rules. Second, compulsion: a government can raise money by compulsory taxation and, in its own currency, by borrowing from the central bank; a private agent can only earn, sell assets or borrow voluntarily. Third, objective: a firm maximises profit and a household its own welfare; a government is meant to maximise social welfare, which includes goods that yield no revenue. Fourth, time horizon: a government is in principle permanent and can borrow for projects whose benefits reach future generations, whereas an individual's horizon is finite. Fifth, a public budget is published and audited; private budgets are confidential. Sixth, the scale of borrowing: internal public debt is owed largely by citizens to citizens, which is why some economists argued it is not a burden in the way private debt is, a claim examined in the chapter on public debt. Seventh, a surplus is not in itself a sign of public success.
Richard Musgrave, in The Theory of Public Finance (1959), organised the functions of the fiscal system into three branches. The allocation branch corrects market failure: it provides public goods, deals with externalities and ensures that resources flow to uses the market undersupplies. The distribution branch adjusts the distribution of income and wealth to what society considers fair, through progressive taxes, transfers and subsidised services. The stabilisation branch uses taxes, spending and borrowing to keep employment high and prices stable. Musgrave's device was to treat the three as if run by separate managers, so that each could be analysed on its own, while admitting that every real tax or spending decision touches all three.
Developing-country texts add a fourth function, economic development or growth: mobilising savings for public investment, building infrastructure, and financing agricultural research, extension and irrigation. In agriculture all branches are visible at once. Public agricultural research is an allocation measure because new varieties are partly non-rival and hard to appropriate; a food subsidy for the poor is distributive; buffer stocks and procurement are partly a stabilisation device; and rural roads and canals are development spending.
A good is rival if one person's consumption reduces what is available to others, and excludable if non-payers can be kept from using it. A pure public good is non-rival and non-excludable: national defence, a flood-warning system, basic agricultural research results once published, or the control of a locust outbreak across a region. A pure private good, such as a bag of fertiliser, is rival and excludable. Between them lie club goods (non-rival up to congestion but excludable, such as a toll road or a cooperative cold store) and common-pool resources (rival but hard to exclude, such as groundwater, fisheries and village grazing land).
Paul Samuelson (1954) showed that the efficient quantity of a pure public good is where the sum of all consumers' marginal rates of substitution between the public good and a private good equals the marginal rate of transformation. Put simply, because everyone consumes the same unit, the demand curves for a public good are added vertically, not horizontally as for private goods. If three farmers value one more kilometre of drainage channel at 20, 15 and 10, its marginal social benefit is 45.
Because a public good is non-excludable, each person has an incentive to understate his valuation and let others pay: the free-rider problem. Voluntary contributions therefore supply too little. Erik Lindahl proposed a benchmark in which each person pays a tax price equal to his marginal benefit; with those personalised prices everyone would demand the same quantity, and that quantity satisfies the Samuelson condition. The Lindahl solution is efficient and follows the benefit principle, but it depends on people truthfully revealing preferences, which is exactly what free riding prevents. In practice governments decide through voting and representation, which is why public choice theory (Buchanan and Tullock) studies how voters, politicians and bureaucrats actually decide, and why government failure is taken as seriously as market failure.
Charles Tiebout (1956) argued that for local public goods, people reveal preferences by choosing where to live, a mechanism he called voting with one's feet. This idea underlies much of fiscal federalism.
An externality is a cost or benefit that falls on others and is not priced. Pigou proposed a corrective tax equal to the marginal external cost at the efficient output, or a subsidy equal to the marginal external benefit. Ronald Coase showed that where property rights are clear and transaction costs are low, bargaining can reach the efficient outcome whatever the initial assignment of rights, so taxation is not the only remedy. Agriculture is full of externalities: pesticide drift and nitrate run-off (negative), stubble burning and air pollution (negative), groundwater depletion by pumping under a flat or free power tariff (negative, and a common-pool problem), and pollination, soil conservation on a hillside or disease control in livestock (positive).
Musgrave also defined merit goods, whose consumption society wishes to encourage beyond what individuals would choose, such as primary education, vaccination and nutrition for children, and demerit goods such as tobacco and liquor. Merit goods are often private in the technical sense but provided or subsidised by the state on paternalistic or equity grounds. Subsidised school meals are usually justified this way, not as public goods.
Hugh Dalton called the principle of maximum social advantage the most fundamental principle in public finance. Every tax imposes a sacrifice on those who pay; every item of public expenditure yields a benefit. The best fiscal system is one that secures the maximum social advantage from the combined operation of taxes and spending. Dalton's rule has two parts. On the spending side, public expenditure should be distributed among uses so that the marginal social benefit of the last rupee, dollar or euro is equal in every use. On the revenue side, taxes should be distributed so that the marginal social sacrifice of the last unit collected is equal from every source. The size of the budget is then set where the marginal social benefit of public expenditure equals the marginal social sacrifice of taxation. Beyond that point a further unit of spending adds less benefit than the sacrifice of raising it.
Diagrammatically, a falling marginal social benefit curve (MSB) and a rising marginal social sacrifice curve (MSS) intersect at the optimum size of the public sector. MSB falls because urgent needs are met first; MSS rises because each extra unit of tax cuts into income of higher marginal utility.
Dalton gave objective tests by which a fiscal system can be judged. Economic welfare should improve through the effect of the budget on production (greater productive power, better organisation of production and improved composition of output) and through its effect on distribution (reduced inequality and reduced fluctuations in income). Other tests he mentioned include preservation of the community (defence and internal order) and the maintenance of economic stability. Modern theory restates it as maximising a social welfare function subject to the resource constraint.
A numerical illustration: suppose a district council ranks six successive crores of expenditure (or any currency unit) with marginal social benefits of 40, 32, 25, 19, 14 and 10 units of welfare, and the marginal social sacrifices of raising each crore are 8, 11, 14, 19, 25 and 32. Spending should rise to four crores, where MSB and MSS are both 19. A fifth crore would add 14 of benefit but cost 25 of sacrifice.
The principle is a normative ideal and has well-known weaknesses. Benefits and sacrifices are subjective and not directly measurable, and interpersonal comparison of utility is needed to add them across people, which many economists reject. Many benefits of public spending are indirect and long delayed, such as research, education or irrigation. The principle is static, while the budget is a dynamic instrument used for stabilisation. Political decisions are made by coalitions and interest groups, not by a planner equating margins. And when taxes distort, the marginal cost of public funds exceeds one, so the MSS curve must include deadweight loss.
Despite these limits the principle disciplines thinking. Cost-benefit appraisal of public projects, the marginal cost of public funds used by the World Bank and national planning bodies, and the practice of comparing the return on an irrigation rupee with that on a fertiliser-subsidy rupee are all practical descendants of Dalton's rule. Studies of Indian public agricultural spending by Fan, Hazell and Thorat used this logic and found that research and rural roads usually gave higher growth and poverty returns than input subsidies, though such rankings depend on period and method.
Public revenue in the broad sense includes all receipts of government: taxes, fees, licence fees, fines and penalties, special assessments, prices of goods and services sold by public enterprises, surpluses of public enterprises, grants and gifts, and borrowing. Many texts separate public revenue proper, which does not create a liability, from public receipts, which include borrowing. A tax is a compulsory payment to government with no direct quid pro quo: the taxpayer receives no specific service in return for the payment. A fee is paid for a specific service, a licence fee for permission to do something; a special assessment is a charge on property that gains from a particular public improvement, such as a betterment levy on land whose value rises when a canal is built.
Taxes are classified as direct (the person on whom it is levied is intended to bear it, such as income and wealth taxes) or indirect (levied on one person but intended to be passed on, such as excise, sales tax, customs duties and value added tax). Taxes may be specific (a fixed amount per unit, such as a duty per tonne) or ad valorem (a percentage of value). Their rate structure may be proportional, progressive or regressive, discussed below. Budget documents also separate revenue receipts (tax and non-tax revenue) from capital receipts (borrowing, recovery of loans and disinvestment proceeds). Tax buoyancy is the percentage change in tax revenue divided by the percentage change in GDP, including discretionary changes in rates and base; tax elasticity excludes those changes. If revenue grows 15 percent while nominal GDP grows 10 percent, buoyancy is 1.5, and the tax-to-GDP ratio rises.
Adam Smith, in The Wealth of Nations (1776), set out four canons of taxation: equality (people should contribute in proportion to their abilities), certainty (the amount, time and manner of payment should be clear, not arbitrary), convenience (tax should be collected at the time and in the manner most convenient to the payer, such as land revenue after harvest), and economy (the cost of collection should be small relative to the yield). Later writers added productivity, elasticity, flexibility, simplicity and diversity.
Two broad principles govern who should pay. The benefit principle says taxes should follow the benefits people receive from public services, as with a fuel tax earmarked for roads or a water rate for canal irrigation. It works well for quasi-private services but cannot finance redistribution, since the poor who gain most from transfers would pay most. The ability-to-pay principle says people with greater ability should pay more. Horizontal equity requires that people in equal positions pay equal tax; vertical equity requires that people with greater ability pay more, but does not by itself say how much more. Ability has been measured by income, consumption (Kaldor's expenditure tax) or wealth.
John Stuart Mill interpreted ability to pay as equal sacrifice. Three versions exist. Equal absolute sacrifice requires each taxpayer to lose the same total utility. Equal proportional sacrifice requires each to lose the same fraction of total utility. Equal marginal sacrifice, also called minimum aggregate sacrifice (Edgeworth and Pigou), requires the marginal utility of the last unit of income left to each person to be equal, which minimises the total sacrifice of society. With identical utility functions and diminishing marginal utility, equal marginal sacrifice implies levelling down the top incomes first, a maximally progressive tax, until the required revenue is raised.
Which rate structure the other two versions imply depends on how fast marginal utility falls. With logarithmic utility, equal absolute sacrifice gives exactly a proportional tax. For example, if two farmers have incomes of 100 and 400 and 100 must be raised, equal absolute sacrifice under log utility requires ln(100) minus ln(100 minus t1) to equal ln(400) minus ln(400 minus t2); this holds when both keep the same fraction of income, so t1 is 20 and t2 is 80. Under equal marginal sacrifice the whole 100 comes from the richer farmer, because his post-tax income of 300 would still exceed the other's 100. If marginal utility falls faster than in the log case, equal absolute sacrifice becomes progressive. The lesson examiners test is that the sacrifice principle alone does not fix the degree of progression; that depends on the assumed utility function.
A tax is progressive when the average tax rate rises with income, proportional when it is constant and regressive when it falls. The marginal rate, the tax on the last unit of income, exceeds the average rate in a progressive system. Consider a schedule with no tax on the first 250, 10 percent from 250 to 500, 20 percent from 500 to 1000 and 30 percent above 1000. At an income of 1200 the tax is 0 plus 25 plus 100 plus 60, which is 185; the average rate is 15.4 percent and the marginal rate 30 percent. At 800 the tax is 85 and the average rate 10.6 percent; at 400 it is 15 and 3.75 percent. The rising average rate confirms progression.
Local measures of progression include average rate progression (the change in average rate per unit change of income) and liability progression, the elasticity of tax liability with respect to income, equal to the marginal rate divided by the average rate. At 1200 in the schedule above, liability progression is 0.30 divided by 0.154, about 1.95; any value above one indicates progression. Residual progression, the elasticity of post-tax to pre-tax income, is below one for a progressive tax. Global measures compare distributions: the Kakwani index is the concentration coefficient of tax payments minus the Gini coefficient of pre-tax income; a positive value means the tax is progressive. The Reynolds-Smolensky index measures redistributive effect as the pre-tax Gini minus the post-tax concentration coefficient, and in the absence of re-ranking it approximately equals t over (1 minus t) times the Kakwani index, where t is the average tax rate. A highly progressive tax that raises little revenue therefore redistributes little. Indirect taxes on necessities are usually regressive with respect to current income.
The impact of a tax is on the person who pays it to the treasury; the incidence is on the person who finally bears its burden, after prices have adjusted. Shifting is the process by which the burden passes from one to another. Forward shifting passes it to buyers through higher prices; backward shifting passes it to suppliers of inputs, including workers and landowners, through lower prices paid. Musgrave distinguished absolute incidence (the effect of one tax, with spending unchanged), differential incidence (the effect of replacing one tax with another of equal yield) and balanced-budget incidence (the combined effect of a tax and the spending it finances). Economic incidence is independent of statutory incidence in a competitive market: whether a per-unit tax is collected from buyers or from sellers, the final prices paid and received are the same. This is one of the most frequently tested results in the subject.
Partial-equilibrium incidence in a single competitive market depends on elasticities. For a small per-unit tax, the share borne by buyers is the elasticity of supply divided by the sum of the elasticity of supply and the absolute elasticity of demand; the share borne by sellers is the absolute elasticity of demand divided by the same sum. The side of the market that is less elastic bears more. If demand is perfectly inelastic, buyers bear it all; if supply is perfectly inelastic, sellers bear it all; if demand is perfectly elastic, sellers bear it all.
Let demand be Qd = 100 minus 2P and supply be Qs = minus 20 plus 4P. Before tax, equilibrium is where 100 minus 2P equals minus 20 plus 4P, giving P = 20 and Q = 60. Now impose a tax of 6 per unit on sellers. Sellers now need Pc minus 6 to supply, so supply becomes minus 20 plus 4(Pc minus 6). Setting it equal to demand gives 6Pc = 144, so the consumer price Pc is 24, the producer price Ps is 18 and quantity falls to 52. Buyers bear 4 of the 6 (two thirds) and sellers 2 (one third). The ratio matches the slopes: demand has slope 2 in quantity per unit price and supply slope 4, so buyers bear 4 over 6. Revenue is 6 times 52, which is 312. The deadweight loss, the triangle between the curves over the lost quantity, is one half times 6 times (60 minus 52), which is 24.
If the tax is doubled to 12, the consumer price becomes 28, quantity 44 and the deadweight loss one half times 12 times 16, which is 96: four times as large. Deadweight loss rises with the square of the tax rate, which is why a broad base with a low rate is usually more efficient than a narrow base with a high rate.
Excess burden, or deadweight loss, is the loss of welfare beyond the revenue collected, caused because the tax changes behaviour. A lump-sum tax, which does not depend on anything the taxpayer can change, has no excess burden; nearly every real tax does. For a small ad valorem tax t on a good with perfectly elastic supply, the Harberger approximation gives deadweight loss as one half times t squared times the absolute demand elasticity times the value of expenditure. With t of 10 percent, demand elasticity of 1.2 and expenditure of 50,000, the loss is 0.5 times 0.01 times 1.2 times 50,000, which is 300.
Frank Ramsey (1927) asked how commodity taxes should be set to raise a given revenue with the least total excess burden. With independent demands, the answer is the inverse elasticity rule: tax rates should be inversely proportional to the demand elasticities, so that the proportional reduction in quantity is equal across goods. Since necessities, including staple foods, have inelastic demand, the Ramsey rule points to taxing them more heavily, which conflicts with equity. That conflict between efficiency and equity is why many countries exempt or zero-rate basic food under VAT, and why Diamond and Mirrlees showed that distributional weights must be added to the Ramsey formula. The Diamond-Mirrlees production efficiency theorem also says intermediate inputs should generally not be taxed, an argument for VAT with input credit.
Partial-equilibrium analysis ignores effects on other markets. Arnold Harberger's two-sector model (1962) showed that a tax on capital in the corporate sector drives capital into the non-corporate sector, lowering the return to capital everywhere, so under his assumptions capital as a whole bears roughly the full burden, not only corporate shareholders. In open economies with mobile capital, much of the burden can shift to immobile labour and land.
Land in fixed supply is the classic case of full backward shifting. A tax on land rent cannot be passed on, because the supply of land does not change; the landowner bears it all. Moreover, the tax is capitalised: the price of land falls by the present value of the future tax stream. A land tax of 50 per hectare per year at a 5 percent discount rate reduces the price of a hectare by 50 divided by 0.05, which is 1,000. Henry George used the fixed supply of land to argue for a single tax on land rent, because it has no excess burden. Taxes on marketed farm output, by contrast, are shared between farmers and buyers according to elasticities, and a tax on an export crop in a small country facing a given world price falls wholly on domestic producers and on factors specific to that crop.
Direct taxation of farm income is difficult almost everywhere in the developing world. Farms are numerous and small, records are rare, output is partly consumed at home, income fluctuates with weather and prices, and in many countries farmers are politically powerful. Administrative cost per unit collected is high, violating Smith's canon of economy. As a result, agriculture has historically been taxed through land taxes, marketing arrangements and trade policy rather than through personal income tax.
Land revenue was the main source of public finance in pre-modern agrarian states and remained important under colonial rule in South Asia. Japan's Meiji land tax reform of 1873 converted payments in kind into a money tax based on assessed land value and financed much of early industrialisation, a case often cited in development texts. Nicholas Kaldor argued in the 1960s that a tax on the potential output of land, assessed on its fertility rather than actual output, is the most efficient way to tax agriculture: it does not penalise extra effort, it falls on a factor in fixed supply, and it encourages idle land to be put to use. Many land taxes eroded because assessments were not revised with inflation.
Where direct taxes are weak, governments have often taxed farmers implicitly by holding domestic farm prices below world prices. Instruments include export taxes, export bans, compulsory procurement at below-market prices, overvalued exchange rates, and state marketing boards that paid farmers less than the export price. The World Bank study led by Krueger, Schiff and Valdes (1988, 1991) covering eighteen developing countries found that indirect taxation through exchange-rate and industrial protection policies generally outweighed direct sectoral price measures in depressing farm incentives. Robert Bates's Markets and States in Tropical Africa (1981) argued that marketing boards in several African countries were used to extract surplus from export-crop farmers for urban and state purposes. Later World Bank work led by Kym Anderson showed that many developing countries moved from taxing to supporting agriculture as they grew richer. The general lesson is the development paradox: poor countries, where farmers are the majority, have tended to tax agriculture; rich countries, where farmers are few, subsidise it.
The incidence of an export tax in a small country follows from the incidence chapter: the world price is given, so the tax lowers the domestic producer price by the full amount and the burden falls on growers and on land and labour specific to the crop. Consumers of the exportable at home gain from the lower domestic price. Market fees and cesses on produce sold in regulated markets follow the same elasticity rule: with a demand elasticity of minus 0.3 and a short-run supply elasticity of arrivals of 0.2, buyers bear 0.2 divided by 0.5, or 40 percent, and farmers 60 percent.
In India, agricultural income is exempt from central income tax under the Income Tax Act, 1961, and the Constitution assigns taxes on agricultural income to the states (State List, Entry 46). Few states levy such a tax in a meaningful way, though some have taxed plantation income. For individuals who also have non-agricultural income above the exemption limit, agricultural income is taken into account for determining the rate applied to non-agricultural income, the so-called partial integration method, which reduces the use of the exemption to lower tax on other income.
China abolished its centuries-old agricultural tax with effect from 1 January 2006, part of a shift toward supporting rural incomes. In the United States, farm income is taxed under the federal income tax like other business income, with some special provisions such as income averaging for farmers. In the European Union, farm income is taxed under national income tax systems, several of which offer simplified or flat-rate regimes for small farmers, including flat-rate VAT schemes that compensate farmers for input VAT. In many African and Latin American countries, presumptive taxes, market cesses and local produce levies play a larger role than income tax. Students should check current rules in the relevant national law before quoting any rate.
Adolph Wagner, observing nineteenth-century Germany, stated the law of increasing state activity: as per capita income rises, the share of public expenditure in national income rises. Explanations include the growing complexity of an industrial economy, which needs more regulation and administration; the income elasticity of demand for education, health and culture exceeding one; and the need to finance large indivisible investments. Empirical support is mixed.
Peacock and Wiseman, studying the United Kingdom from 1890 to 1955, argued that spending grows not smoothly but in steps. In normal times, tolerable levels of taxation limit spending. A social upheaval, such as war or depression, forces spending up and makes higher taxes acceptable: the displacement effect. After the crisis, spending does not fall back to the old level, because the tolerable burden has been permanently raised, the inspection effect has revealed new needs, and functions have been concentrated at the centre (the concentration effect). Other explanations include Baumol's cost disease (slow productivity growth in labour-intensive services) and budget-maximising bureaucrats (Niskanen).
Public expenditure can be classified in several ways. Economically, it is divided into revenue (current) expenditure, which does not create assets, such as salaries, interest, subsidies and maintenance, and capital expenditure, which creates assets or reduces liabilities, such as irrigation works, roads, loans to public enterprises and repayment of debt. Functionally, it is divided by purpose: general services (administration, defence, police), social services (education, health), and economic services (agriculture, irrigation, energy, transport). The UN Classification of the Functions of Government (COFOG) standardises this for international comparison. Another classification is into transfer and non-transfer (exhaustive) expenditure: transfers such as pensions, interest and direct benefit payments move purchasing power without using up resources, while purchases of goods and services absorb resources directly. Older Indian budgets classified spending as plan and non-plan, a distinction dropped from the central budget in 2017-18. Spending may also be classified as developmental (economic and social services) or non-developmental (administration, defence, interest), a split common in Indian state finance studies.
Public expenditure affects production through the ability to work, save and invest (health, education and infrastructure raise ability), the willingness to work and save (transfers may reduce willingness if poorly designed), and the diversion of resources among sectors. It affects distribution through progressive spending on services used most by the poor and through transfers. And it affects stability through the multiplier: in a simple Keynesian model with marginal propensity to consume c, the government expenditure multiplier is 1 divided by (1 minus c), the tax multiplier is minus c divided by (1 minus c), and a balanced increase in spending and lump-sum taxes has a multiplier of one (Haavelmo's balanced budget theorem). With c of 0.8, the spending multiplier is 5 and the tax multiplier minus 4. Leakages and crowding out reduce these values in practice.
Public investment can crowd in private investment when it builds complementary infrastructure. Where public spending is financed by borrowing that raises interest rates, it may crowd out private investment instead.
Agricultural support takes many forms: output price support (a guaranteed or minimum price backed by public procurement or deficiency payments), input subsidies (on fertiliser, seed, power, irrigation water, credit and insurance premiums), direct payments (per hectare, per head of livestock or per farm, coupled or decoupled from current production), general services (research, extension, inspection, infrastructure, marketing and storage) and consumer food subsidies. The OECD measures support to producers with the Producer Support Estimate (PSE), the annual monetary value of gross transfers from consumers and taxpayers to farmers arising from policy, expressed also as a percentage of gross farm receipts. The General Services Support Estimate covers collective services. The WTO Agreement on Agriculture classifies domestic support into boxes: the amber box for trade-distorting support subject to reduction commitments, the blue box for payments under production-limiting programmes, and the green box for minimally distorting support such as research, extension, and decoupled income support. Article 6.2 exempts some investment and input subsidies to low-income or resource-poor producers in developing countries.
Consider fertiliser supplied at a constant cost of 20 per unit (perfectly elastic supply) with farm demand Q = 200 minus 5P. At 20 farmers buy 100 units. A subsidy of 4 per unit lowers the price to 16 and raises use to 120. The fiscal cost is 4 times 120, which is 480. Farmers' gain in surplus is the area between the two prices under the demand curve, 4 times (100 plus 120) divided by 2, which is 440. The difference, 40, is the deadweight loss: fertiliser whose value to farmers is less than its cost is being used. If the tax that finances the subsidy itself has an excess burden, the true social cost is higher still.
Input subsidies are also criticised because they tend to favour larger farmers who buy more inputs, distort input ratios (for example, subsidising one nutrient more than others encourages imbalanced application), promote overuse of groundwater when power is free or flat-rated, and are hard to withdraw once capitalised into land values and expectations. Supporters argue that where credit and insurance markets are missing, well-targeted subsidies can speed adoption of productive technology. Malawi's input subsidy programme, begun in the 2005-06 season, was widely studied: early evaluations credited it with higher maize output, while later ones raised concerns about targeting, displacement of commercial purchases, fiscal cost and sustainability. The evidence from Africa's so-called smart subsidies is mixed and depends on design.
A minimum support price backed by procurement guarantees a floor price. Its fiscal cost includes the cost of buying, storing and distributing stocks, minus what is recovered on sale; when stocks exceed buffer and distribution needs, carrying costs and losses rise. In India, minimum support prices are announced on the recommendation of the Commission for Agricultural Costs and Prices, and the Food Corporation of India and state agencies procure mainly rice and wheat for the public distribution system. The United States has used deficiency payments, marketing loans, counter-cyclical programmes and subsidised crop insurance. The European Union's Common Agricultural Policy moved from intervention buying and export refunds toward direct payments, and the 2003 reform introduced the Single Payment Scheme that decoupled most payments from current production; later reforms linked payments to environmental conditions and, from 2023, eco-schemes.
Direct income transfers are an alternative to price and input subsidies. India's PM-KISAN scheme, launched in 2019, pays a fixed sum per year to eligible landholding farm families in three instalments; the amount at launch was 6,000 rupees a year. Such transfers distort less than input or price subsidies because they change no relative price, which is why decoupled transfers fit the WTO green box. Their limitations are targeting (tenants and landless labourers who do not own land may be excluded) and the lack of any direct link to productivity or resource conservation.
A government budget is a statement of estimated receipts and expenditure for a financial year, together with the policy behind them. It is at once an accounting record, a legal authorisation from the legislature, and an instrument of economic policy. In India the Constitution calls it the Annual Financial Statement (Article 112), and money can be withdrawn from the Consolidated Fund of India only under an appropriation made by law. Most budgets separate a revenue account (revenue receipts and revenue expenditure) from a capital account (capital receipts and capital expenditure). Revenue receipts are tax revenue and non-tax revenue such as interest, dividends, fees and profits of enterprises; capital receipts are borrowings and other liabilities, recovery of loans and disinvestment proceeds.
Budgets may be balanced, surplus or deficit. Classical economists favoured an annually balanced budget. Abba Lerner's functional finance argued that the budget should be judged by its effect on the economy, not by whether it balances, and that deficits are appropriate when unemployment is high. The cyclically balanced budget, balanced over the business cycle rather than every year, sits between these views. Automatic stabilisers, such as progressive income tax, raise the deficit in a downturn without any new decision.
Several deficit concepts are used, and examiners test their definitions and arithmetic. Revenue deficit is revenue expenditure minus revenue receipts: government is borrowing to pay for current consumption. Fiscal deficit is total expenditure minus total receipts other than borrowing (revenue receipts plus non-debt capital receipts, which are recovery of loans and disinvestment); it equals the government's total borrowing requirement. Primary deficit is fiscal deficit minus interest payments; it shows how much of the borrowing is for current decisions rather than the cost of past debt. Effective revenue deficit, an Indian measure introduced in the 2011-12 budget, is revenue deficit minus grants to states and other bodies for the creation of capital assets. Budget deficit, an older Indian concept financed by ad hoc treasury bills, was dropped after automatic monetisation ended in 1997. Monetised deficit is the part of the deficit financed by the central bank's net credit to government.
Worked example: revenue receipts 2,000; non-debt capital receipts 100; revenue expenditure 2,600, of which interest is 700 and grants for creation of capital assets 200; capital expenditure 500; GDP 25,000. Total expenditure is 3,100. Fiscal deficit is 3,100 minus 2,100, which is 1,000, or 4 percent of GDP. Revenue deficit is 600, which is 60 percent of the fiscal deficit: most borrowing finances current spending. Primary deficit is 1,000 minus 700, which is 300, or 1.2 percent of GDP. Effective revenue deficit is 600 minus 200, which is 400.
Traditional line-item budgeting lists inputs (salaries, travel, supplies) and makes control easy but says little about results. Performance budgeting, developed in the United States after the Hoover Commission (1949), classifies spending by programmes and activities with measures of workload and output. Planning-programming-budgeting systems (PPBS) of the 1960s linked spending to long-term objectives. Zero-based budgeting, associated with Peter Pyhrr in the 1970s, requires every programme to be justified from zero each year rather than from the previous year's base. Outcome budgeting links spending to measurable outcomes; India's central government has published an outcome budget since 2005. Medium-term expenditure frameworks set multi-year ceilings.
Fiscal rules limit deficits or debt. The European Union's Maastricht criteria set reference values of 3 percent of GDP for the government deficit and 60 percent for gross debt. India's Fiscal Responsibility and Budget Management Act, 2003 set targets for eliminating the revenue deficit and reducing the fiscal deficit; the targets and dates have been amended and suspended several times, including during the pandemic, so students should consult the current Act and the Medium-Term Fiscal Policy Statement.
Fiscal federalism studies how functions and revenue sources should be divided among levels of government. Musgrave assigned stabilisation and distribution mainly to the central government, because local governments cannot control monetary policy, face leakage of stimulus to other regions, and cannot redistribute when rich taxpayers can move away. Allocation of local public goods is best done locally. Wallace Oates's decentralisation theorem (1972) states that where a public good's benefits are confined to a jurisdiction and there are no cost savings from central provision, it is at least as efficient, and usually more so, for each local government to provide the level its residents prefer than for the centre to provide a uniform level. Tiebout's voting with one's feet strengthens the case where people are mobile.
Taxes suited to local governments are those on immobile bases, such as land and property, and user charges, following the benefit principle. Taxes on mobile bases, progressive income taxes, and taxes whose base is unevenly distributed across regions, such as natural-resource royalties or customs, are better assigned to the centre. The result, in most federations, is vertical fiscal imbalance: the centre raises more revenue than it spends, and sub-national governments spend more than they raise. Horizontal fiscal imbalance refers to differences in fiscal capacity and needs between regions at the same level.
Transfers close these gaps. An unconditional or general-purpose grant can be spent as the recipient wishes and acts like an increase in the community's income. A conditional or specific-purpose grant must be spent on a given function. A matching grant pays a share of each unit spent on a function, which lowers its local price; a one-to-one matching grant halves the price, and a grant that pays 75 percent leaves the local government paying 25 percent. Matching grants have both an income and a substitution effect and so stimulate spending on the aided function more than an equal-sized lump-sum grant, which has only an income effect. They are justified when local spending creates benefits that spill over to other regions. Empirical studies often find that lump-sum grants raise local public spending much more than an equal rise in private income would, a result known as the flypaper effect: money sticks where it hits. Equalisation grants aim to let every region provide comparable services at comparable tax effort.
In India, the Finance Commission under Article 280 of the Constitution is appointed every five years to recommend the share of states in the divisible pool of central taxes and its distribution among states, together with grants-in-aid under Article 275. The Fourteenth Finance Commission raised the states' share to 42 percent, and the Fifteenth, chaired by N. K. Singh, recommended 41 percent for 2021-26 after the reorganisation of Jammu and Kashmir. Distribution among states uses criteria such as population, area, income distance and forest cover. The Goods and Services Tax, introduced from 1 July 2017 through the 101st Constitutional Amendment, created a dual GST levied concurrently by the centre and states and a GST Council (Article 279A) for joint decisions. Agriculture, land and water are State subjects, so most farm spending is done by states, supported by centrally sponsored schemes, which are conditional grants.
Australia's Commonwealth Grants Commission recommends distribution of GST revenue among states on the principle of horizontal fiscal equalisation. Canada operates an equalisation programme financed by the federal government. The United States has no general equalisation programme; its federal grants are mostly categorical. China's 1994 tax-sharing reform recentralised revenue while leaving large spending responsibilities with local governments, which then relied on land sales and borrowing through financing vehicles, an example of how a vertical imbalance can shape local fiscal behaviour.
In developing countries fiscal policy has broader aims than short-run stabilisation. It is expected to mobilise resources for investment, raise the ratio of savings to income, allocate investment to priority sectors, reduce inequality, and keep inflation in check while doing so. The instruments are taxation, public expenditure, public borrowing and deficit financing, and the management of public enterprises. The Harrod-Domar framework, in which the growth rate equals the savings rate divided by the capital-output ratio, was used in early development planning to calculate how much public saving and investment was needed for a target growth rate.
Deficit financing in the older Indian sense meant financing expenditure by borrowing from the central bank, that is, by creating new money. Its advocates held that if it financed productive investment and the economy had idle resources, the extra output would absorb the new money without much inflation. Critics noted that farm supply is inelastic in the short run, so money creation raises food prices and acts as an inflation tax on the poor. Most countries have since restricted direct central-bank financing of deficits.
A developing country's tax effort is limited by a small modern sector, a large informal and agricultural sector, low literacy and weak administration, so the tax-to-GDP ratio is usually lower than in rich countries. Tax effort studies compare actual collections with the capacity predicted from income, trade openness and sector shares. Reform packages associated with the IMF and World Bank since the 1980s have favoured broad-based VAT, lower and fewer rates, fewer exemptions, and moderate corporate and personal income taxes. A VAT taxes only the value added at each stage because each trader deducts the tax paid on inputs. For a chain with sales values of 100, 150 and 220 at a 10 percent rate, VAT collects 10, 5 and 7, a total of 22, equal to 10 percent of the final value. A cascading turnover tax of 10 percent on each gross sale without credit collects 10, 15 and 22, a total of 47, and it encourages firms to integrate vertically simply to avoid tax.
Tax expenditures, such as exemptions for agricultural income, concessional rates on farm inputs or deductions for rural investment, are revenue forgone and should be evaluated with the same rigour as direct spending.
Fiscal policy shapes agriculture through public investment in irrigation, research and roads; through subsidies and procurement; through taxation of inputs and outputs; and through safety nets. Rural public works programmes, such as India's employment guarantee under the Mahatma Gandhi National Rural Employment Guarantee Act, 2005, which guarantees up to 100 days of wage employment a year to a rural household whose adult members volunteer for unskilled manual work, act both as transfers and as automatic stabilisers because demand for work rises in droughts. Conditional cash transfers such as Mexico's Progresa and Brazil's Bolsa Familia link payments to school attendance and health visits.
The composition of spending often matters more for development than its level. A shift from input subsidies toward public goods such as research, extension, irrigation maintenance and market infrastructure is a recurring recommendation of the World Bank's World Development Report 2008 (Agriculture for Development) and of national expert committees. Political economy resists such shifts, since subsidies have concentrated beneficiaries.
Public debt is the stock of government liabilities from past borrowing. It is classified as internal (owed to residents) or external (owed to non-residents); marketable (bonds and treasury bills) or non-marketable (small savings, provident funds); short-term, medium-term or long-term; productive or unproductive; and funded (long-term, with no fixed date or a distant one) or floating (short-term, such as treasury bills).
Whether public debt is a burden on future generations is a long debate. Lerner argued that internal debt is not a burden on the nation as a whole, since we owe it to ourselves: future taxes to pay interest are transferred to future bondholders. External debt is a real burden because servicing it sends resources abroad. Buchanan argued that the burden falls on future taxpayers who must pay taxes they did not choose. Modigliani argued that debt finance displaces private capital accumulation, so future generations inherit a smaller capital stock and lower income, which is a real burden even with internal debt when the economy is at full employment. Robert Barro revived the Ricardian equivalence proposition (1974): if households are forward-looking, linked across generations by bequests, and face no borrowing constraints, a debt-financed tax cut will be saved in anticipation of future taxes, so debt and taxes have equivalent effects on demand. Most empirical work finds only partial equivalence.
The debt-to-GDP ratio d evolves according to the equation d(next) = d times (1 plus r) divided by (1 plus g), plus the primary deficit ratio, where r is the nominal interest rate on debt and g the nominal growth rate of GDP. When r exceeds g, debt ratios rise unless the government runs a primary surplus; when g exceeds r, a stable ratio is consistent with a primary deficit. The primary balance that holds the debt ratio constant equals d times (r minus g) divided by (1 plus g). With debt at 50 percent of GDP, r of 8 percent and g of 6 percent, the required primary surplus is 0.5 times 0.02 divided by 1.06, about 0.94 percent of GDP. With a primary deficit of 2 percent instead, next year's ratio is 0.5 times 1.08 divided by 1.06 plus 0.02, which is 52.9 percent, a rise of 2.9 points. With d of 60 percent, r of 7 percent and g of 10 percent, the ratio can be held with a primary deficit of about 1.6 percent of GDP.
Evsey Domar (1944) showed that if the government runs a constant overall deficit of f percent of GDP and nominal income grows at g, the debt ratio converges to f divided by g. A deficit of 3 percent with 10 percent nominal growth gives a long-run ratio of 30 percent, and the interest burden stays bounded.
Methods of debt redemption include repayment from budget surpluses, a sinking fund (setting aside a sum each year to redeem debt at maturity), refunding (issuing new debt to repay maturing debt), conversion (replacing high-interest debt with lower-interest debt), capital levy (a one-time tax on wealth) and, in extremis, repudiation or restructuring. Inflation also erodes nominal debt at bondholders' expense.
Debt management covers the choice of maturity, currency and instrument to minimise cost subject to prudent risk. Short maturities and foreign-currency debt raise rollover and exchange-rate risk. For low-income countries, the Heavily Indebted Poor Countries (HIPC) Initiative launched in 1996 by the IMF and World Bank, and the Multilateral Debt Relief Initiative of 2005, cancelled debts conditional on reforms and poverty-reduction strategies. Within federations, sub-national borrowing is often controlled: in India, Article 293 requires a state that is indebted to the centre to obtain the centre's consent before borrowing, and states have their own fiscal responsibility laws. Debt waivers for farm loans, announced by several Indian states and the centre at various times, are a fiscal transfer financed ultimately by the budget; their effects on credit discipline are debated.
A student needs to be able to define each deficit and compute it from a set of budget figures, derive tax incidence from elasticities or from linear demand and supply, draw and explain the deadweight-loss triangle, state the Samuelson condition and Dalton's principle, and recall the named theories of public expenditure growth and public debt. Practise reading an actual budget at a glance, separating revenue from capital items, and explaining in two sentences why a fertiliser subsidy has a deadweight loss that a decoupled transfer does not. These skills answer most questions in the JRF, NET and bank-officer papers.
Five years into work in a bank, a ministry, a development agency or an agribusiness firm, public finance is used to read budget announcements and judge their effect on the business: a change in the GST rate on inputs, a new interest subvention scheme, an increase in a procurement price or a new direct transfer. A bank officer appraising a rural scheme uses subsidy and grant rules to assess repayment capacity. A company analyst uses incidence logic to estimate how much of a tax or cess on its product can be passed on to buyers. Knowing that incidence depends on elasticities, not on who writes the cheque, prevents costly pricing mistakes.
A senior official or executive with twenty-five years of experience is asked to design or defend fiscal programmes: whether to replace an input subsidy with a direct transfer, how to target a price support, how to finance a state irrigation programme within a fiscal responsibility limit, or how to present an industry's case before a Finance Commission or a tax council. The work requires estimating fiscal cost, deadweight loss and distributional effects, anticipating political economy, and knowing international experience with decoupling, input vouchers and cash transfers, so that recommendations survive both a finance ministry and a farmers' organisation.
A researcher with a doctorate and long experience works on questions that remain open: the incidence of input subsidies and price supports across farm sizes and between farmers and landowners; the marginal cost of public funds in economies with large informal sectors; the returns to different kinds of public agricultural spending; the effects of intergovernmental transfers on state agricultural investment; and debt sustainability under climate shocks. Methods include computable general equilibrium models, household survey microsimulation, panel econometrics of fiscal data and randomised or quasi-experimental evaluation of transfer programmes. Such work shapes Finance Commission formulas, subsidy reform and international support estimates.
The manual gives no current budget, deficit, debt or subsidy figures for any country, because they change every year; read them from the latest official budget documents. It does not state current tax rates, current PM-KISAN or MSP amounts beyond the launch-year PM-KISAN figure, or the Sixteenth Finance Commission's recommendations, which were not verified. Optimal income taxation (Mirrlees) and the theory of tax evasion are only mentioned, not developed. International taxation, carbon taxes and the fiscal treatment of climate adaptation are not covered. The chapter on agricultural taxation gives principles and a few well-documented cases; state-level agricultural income tax laws in India and farm tax regimes in other countries should be checked in the current law.
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.