Agribusiness Management · Foundations · also AEC-501, Advanced Micro Economic Analysis (AEC-601), Price theory
Consumer, producer and market theory at master's depth, with agricultural applications such as price floors and support prices. The core of the economic-theory unit in agricultural economics PhD entrance papers.
In the exams: agricultural master's entrance examinations, agricultural economics (heavy); agricultural economics PhD entrance and fellowship examinations, economic theory (heavy); agricultural scientist eligibility tests, economic theory (heavy)
Microeconomic theory studies how individual decision makers - households, farms, firms, traders and the owners of land, labour and capital - make choices under scarcity, and how their choices interact in markets to set prices and quantities. It is often called price theory because prices tie the pieces together. Agricultural economics grew out of this theory, and almost every applied topic in the discipline - price policy, input subsidies, farm size, market power in value chains, food security, land rent - is an application of it.
This manual covers the course usually taught as the first micro-economics paper of a Master's in agricultural economics or agribusiness (for example AEC-501 in the ICAR syllabus, and the core micro course of most European, American, African and Asian programmes), with the more advanced material of a PhD course (general equilibrium, welfare economics and market failure) in the last chapter. It moves from the consumer to the producer, market structures, factor markets and the whole economy, with agricultural examples throughout.
The subject is examined heavily: it is central to ICAR AIEEA-PG Group 06, the whole of Unit 1 (Economic Theory) of the ICAR AICE-JRF/SRF agricultural economics paper and the economic theory part of ASRB NET, and economics papers in UGC-NET and graduate entrance examinations elsewhere test the same material. Questions are definitions and properties, one-step numericals, and diagram reasoning expressed in words (who bears a tax, why a price floor creates a surplus).
Economics begins with scarcity: wants are unlimited but resources - land, labour, capital, water, entrepreneurship - are limited and have alternative uses, so every choice has an opportunity cost, defined as the value of the best alternative given up. A field planted to sugarcane costs the wheat it could have grown.
The production possibility frontier (PPF) (or transformation curve) shows the maximum combinations of two goods an economy, region or farm can produce with given resources and technology when the resources are fully and efficiently used. Points on it are efficient, points inside show idle or misused resources, and points outside are unattainable. The slope of the frontier is the marginal rate of transformation (MRT): how much of one good must be given up to get one more unit of the other. It measures opportunity cost directly.
If resources are equally suited to both goods, the frontier is a straight line and opportunity cost is constant. If a farm can produce either 100 tonnes of wheat or 60 tonnes of pulses along a straight line, a tonne of pulses costs 100/60 = 1.67 tonnes of wheat. Usually resources are specialised - some land is better for rice, some for pulses - so the frontier is concave to the origin (bowed outward) and opportunity cost rises as more of one good is produced. This is the law of increasing opportunity cost.
The frontier shifts outward with more resources (new irrigation, more labour) or better technology (high-yielding varieties). A technical change that affects only one crop rotates the frontier outward on that crop's axis.
Every society must answer what to produce, how to produce it and for whom; farming adds when to produce, because it runs on seasons and biological lags. A market economy answers these questions mainly through prices, a command economy through planning, and every real economy is mixed - agricultural markets visibly so, with private trade alongside public procurement, support prices, subsidies and tariffs. Economics uses models, and the ceteris paribus assumption (other things equal) lets us study one relationship at a time.
Positive economics describes and predicts: it deals with what is, was or will be, and its statements can in principle be tested against evidence. 'A 10 per cent rise in the procurement price of wheat will raise the area under wheat in the next season' is a positive statement; it may be wrong, but data can check it. Normative economics prescribes: it deals with what ought to be and depends on value judgements. 'The government should raise the procurement price of wheat to protect farm incomes' is normative; evidence can inform it but cannot settle it, because it rests on a judgement about whose welfare counts and how much. Welfare economics tries to discipline normative reasoning, for example with the Pareto criterion, which needs only the weak value judgement that making someone better off and nobody worse off is an improvement.
Robbins defined economics as the study of human behaviour as a relationship between ends and scarce means that have alternative uses, replacing older definitions based on wealth (Adam Smith) or material welfare (Marshall).
Demand is the quantity of a good buyers are willing and able to buy at each price over a period, other things equal. The law of demand says quantity demanded falls as price rises. Other determinants are income, prices of substitutes and complements, tastes, expectations, population and, for farm goods, season and festivals. A change in the good's own price moves the buyer along the demand curve (a change in quantity demanded); a change in any other determinant shifts the whole curve (a change in demand).
Supply is the quantity sellers are willing and able to offer at each price. It depends on input prices, technology, prices of competing and joint products, weather, the number of sellers and policy. For many crops supply responds to expected rather than current price, because the planting decision is made months before harvest.
Market equilibrium is the price at which quantity demanded equals quantity supplied. If demand is Qd = 200 - 4P and supply is Qs = -40 + 2P, equilibrium is at 200 - 4P = -40 + 2P, so P = 40 and Q = 40. Comparative statics follow directly. A good monsoon shifts supply right: price falls and quantity rises. A rise in consumer incomes shifts demand for a normal good such as milk right: price and quantity both rise. When both curves shift, one variable's direction is certain and the other depends on which shift is larger.
Elasticity measures responsiveness in percentage terms, so it does not depend on units. Price elasticity of demand is the percentage change in quantity demanded divided by the percentage change in price; it is negative for normal downward-sloping demand, and many texts quote its absolute value. Demand is elastic if the absolute value exceeds 1, inelastic if it is below 1 and unit elastic at 1. The point formula is (dQ/dP)(P/Q). For a linear demand Q = 100 - 2P at P = 20, Q = 60 and elasticity is -2 x 20/60 = -0.67. Along a straight-line demand curve, elasticity changes: it is unitary at the midpoint, elastic above it and inelastic below it. The arc (midpoint) formula uses average price and quantity.
Total revenue rises with a price cut when demand is elastic, falls when it is inelastic, and is at its maximum where elasticity is -1. Because demand for staples is inelastic, a bumper harvest can reduce farmers' total revenue (King's law). The reciprocal of price elasticity is price flexibility; with elasticity of -0.3, a 10 per cent larger crop requires a price fall of roughly 10/0.3 = 33 per cent.
Income elasticity is the percentage change in quantity demanded divided by the percentage change in income. It is positive for normal goods, negative for inferior goods, between 0 and 1 for necessities and above 1 for luxuries. Engel's law states that as income rises, the share of income spent on food falls, so food as a whole has an income elasticity below 1. Fruit, dairy and meat typically have higher income elasticities than cereals, which is why diets diversify with growth; actual values differ by country and income group and must be estimated.
Cross-price elasticity is the percentage change in the quantity of one good divided by the percentage change in the price of another. It is positive for substitutes and negative for complements.
Price elasticity of supply is the percentage change in quantity supplied divided by the percentage change in price. For the supply curve Qs = -40 + 2P at P = 40, it is 2 x 40/40 = 2. Supply is more elastic in the long run, when area, herds and orchards can change. A linear supply curve through the origin has elasticity exactly 1 at every point; one that cuts the price axis is elastic, and one that cuts the quantity axis is inelastic.
Demand elasticities are linked by restrictions that come from the budget constraint and from rational choice, and advanced papers test them. Engel aggregation: the budget-share-weighted sum of income elasticities equals 1. If food takes 60 per cent of the budget with income elasticity 0.5, the remaining goods, with a 40 per cent share, must have an average income elasticity of (1 - 0.6 x 0.5)/0.4 = 1.75. Homogeneity: demand is homogeneous of degree zero in prices and income (doubling all prices and income changes nothing), so for each good the own-price elasticity, the cross-price elasticities and the income elasticity sum to zero. Cournot aggregation: the share-weighted sum of all goods' elasticities with respect to one price equals minus that good's budget share.
The classical and Marshallian approach treats utility as measurable in units ('utils'), with the marginal utility of money assumed constant. Total utility is the satisfaction from consuming a quantity; marginal utility (MU) is the extra satisfaction from one more unit. The law of diminishing marginal utility says MU falls as consumption of a good rises, other consumption constant: the third litre of milk a day adds less to a household's satisfaction than the first.
With many goods, the consumer is in equilibrium when the marginal utility per rupee, dollar or shilling is the same for every good: MUx/Px = MUy/Py = ... = the marginal utility of money. This is the law of equi-marginal utility (Gossen's second law). If MUx/Px is 6 and MUy/Py is 4, a unit of money moved from Y to X raises total utility, so the consumer buys more X and less Y until the ratios are equal.
The approach's weaknesses - utility cannot be measured in units, the marginal utility of money is not constant, and income and substitution effects cannot be separated - led to the ordinal approach.
Ordinal utility theory, developed by Pareto and brought into English textbooks by Hicks and Allen, needs only that the consumer can rank bundles. Standard assumptions on preferences are completeness (any two bundles can be ranked), transitivity (if A is preferred to B and B to C, then A is preferred to C), non-satiation or monotonicity (more is preferred to less) and convexity (averages are preferred to extremes). A utility function then represents the ranking, and any increasing transformation of it represents the same preferences - which is why only the ordering matters.
An indifference curve joins bundles that give the same satisfaction. Indifference curves slope downward (by non-satiation), are convex to the origin (by convexity), never intersect (by transitivity) and higher curves represent higher satisfaction. The absolute slope is the marginal rate of substitution (MRS): the amount of Y the consumer will give up for one more unit of X while staying equally well off. MRS equals MUx/MUy and diminishes as X is substituted for Y. For U = xy at x = 10 and y = 20, MRS = y/x = 2. Perfect substitutes give straight-line indifference curves; perfect complements give L-shaped ones.
The budget line shows the bundles the consumer can just afford: Px x + Py y = M. With M = 600, Px = 20 and Py = 30 the intercepts are 30 units of X and 20 of Y and the slope is -Px/Py = -2/3. A rise in income shifts the line out in parallel; a change in one price rotates it about the other intercept.
The consumer maximises utility by reaching the highest indifference curve the budget allows. At an interior optimum the indifference curve is tangent to the budget line, so MRS = Px/Py, which is the ordinal counterpart of the equi-marginal rule; the second-order condition is that indifference curves are convex. If preferences do not give tangency - for example with perfect substitutes - the consumer ends at a corner and spends everything on one good.
With Cobb-Douglas utility U = x^a y^(1-a), the consumer spends the fixed share a of income on X, so x = aM/Px and y = (1-a)M/Py. With a = 0.4, M = 1000, Px = 10 and Py = 20, the demands are x = 40 and y = 30.
Varying income with prices fixed traces the income-consumption curve; from it comes the Engel curve, which plots quantity (or expenditure) against income and slopes upward for a normal good and downward for an inferior one. Varying one price with income fixed traces the price-consumption curve, and from it the ordinary (Marshallian) demand curve.
When the price of a good falls, two things happen. The good becomes cheaper relative to others, so the consumer substitutes towards it (the substitution effect); and the consumer's real income rises, so demand changes as it would with more income (the income effect). The substitution effect of a price fall is always to increase quantity demanded, because indifference curves are convex. The income effect raises quantity for a normal good and lowers it for an inferior good.
Hicks and Slutsky separate the two effects differently. Hicks compensates the consumer so that the original level of utility is kept: the substitution effect is the move along the original indifference curve to the new price ratio. Slutsky compensates so that the original bundle is just affordable: the consumer's money income is adjusted by the quantity originally bought times the price change. Slutsky's compensation is observable and slightly over-compensates, so the Slutsky substitution effect is at least as large as the Hicks one.
A worked case. With U = xy, income 120 and Py = 1, demand for X is x = M/(2Px). When Px falls from 4 to 2, x rises from 15 to 30. Slutsky compensation takes away 15 x 2 = 30, leaving 90; at the new price the consumer buys 90/4 = 22.5, so the Slutsky substitution effect is 7.5 and the income effect is 7.5. For Hicks, original utility is 15 x 60 = 900; at the new prices the cheapest way to reach 900 gives x = square root of (900 x 1/2) = 21.2, so the Hicks substitution effect is about 6.2 and the income effect about 8.8. The total change, 15, is the same in both decompositions.
If the good is inferior, the income effect works against the substitution effect. Usually the substitution effect is larger and demand still slopes downward. If the good is so strongly inferior, and takes such a large share of the budget, that the income effect outweighs the substitution effect, quantity demanded rises when price rises: this is a Giffen good. The classic story is of very poor households for whom a staple is both inferior and a large part of spending; when its price rises they can no longer afford dearer foods and buy more of the staple. Every Giffen good is inferior, but most inferior goods are not Giffen.
The Slutsky equation states the decomposition precisely: dx/dPx = (dx/dPx) compensated - x (dx/dM). In elasticity form, the uncompensated own-price elasticity equals the compensated elasticity minus the budget share times the income elasticity. With a compensated elasticity of -0.3, a budget share of 0.4 and an income elasticity of 0.5, the ordinary elasticity is -0.3 - 0.4 x 0.5 = -0.5. For a farm household that sells what it produces, a higher price also raises income, so the income effect can even be positive (the farm-household model).
The primal problem maximises utility subject to the budget and gives Marshallian (ordinary) demand x(P, M) and the indirect utility function V(P, M). The dual problem minimises the expenditure needed to reach a given utility level and gives Hicksian (compensated) demand h(P, U) and the expenditure function E(P, U). The two are linked: E(P, V(P, M)) = M and h(P, U) = x(P, E(P, U)).
Shephard's lemma says the derivative of the expenditure function with respect to a price is the Hicksian demand for that good. Roy's identity recovers Marshallian demand from the indirect utility function. Because the matrix of compensated price effects (the Slutsky substitution matrix) comes from second derivatives of the expenditure function, it is symmetric and negative semi-definite: own compensated effects are never positive, and the compensated effect of Py on x equals the compensated effect of Px on y. Hicksian demand therefore always slopes downward.
Samuelson's revealed preference theory derives demand from observed choices instead of assumed utility. If a consumer buys bundle A when bundle B was also affordable, A is directly revealed preferred to B. The weak axiom of revealed preference (WARP) says: if A is revealed preferred to B, then B must never be revealed preferred to A - whenever B is chosen, A must be unaffordable. The strong axiom (SARP) extends the rule to chains of choices (Houthakker).
A check shows how the axiom is applied. At prices (1, 2) a household buys (4 kg rice, 3 kg pulses), costing 10, when (6, 1) would have cost 8: (4, 3) is revealed preferred to (6, 1). At prices (2, 1) it buys (6, 1), costing 13, when (4, 3) would have cost 11: now (6, 1) is revealed preferred to (4, 3). The two statements contradict each other, so the choices violate WARP and no stable, well-behaved preference ordering explains them.
Revealed preference gives the law of demand without indifference curves: WARP implies that the Slutsky-compensated substitution effect is non-positive.
Consumer surplus is the difference between what consumers would be willing to pay for a quantity and what they actually pay. On a demand diagram it is the area below the demand curve and above the price, up to the quantity bought. For Q = 100 - 2P (choke price 50) at P = 20, consumers buy 60 and consumer surplus is 0.5 x 60 x (50 - 20) = 900. Producer surplus is the area above the supply curve and below the price: what sellers receive over the minimum they would accept. For Qs = -40 + 2P at P = 40 (supply starts at a price of 20) it is 0.5 x 40 x 20 = 400. Their sum, total surplus, is the usual yardstick of market efficiency, and a deadweight loss is a fall in total surplus that no one receives as a gain.
Marshall's measure is exact only if the marginal utility of money is constant (no income effect). Compensating variation (CV) is the income change that would restore the consumer to the original utility after the price change; equivalent variation (EV) is the income change that, at the old prices, would have the same effect on utility as the price change. For a normal good, when price rises, EV is smaller than the Marshallian loss, which is smaller than CV. With U = xy, income 120, Py = 1 and Px rising from 2 to 4, CV is about 49.7, EV about 35.1 and the Marshallian loss about 41.6. When income effects are small the three are close (Willig), which justifies Marshallian surplus in most appraisals.
A price ceiling is a legal maximum price. It binds only if set below equilibrium. With Qd = 200 - 4P and Qs = -40 + 2P, a ceiling of 35 raises quantity demanded to 60 while quantity supplied falls to 30, a shortage of 30. The standard predictions follow: shortages, queues, rationing by waiting or favouritism, black markets at prices above the ceiling, a fall in quality, and reduced supply in the long run because producers earn less. Consumers who obtain the good gain; those who cannot lose; producers lose; and there is a deadweight loss on the units no longer traded.
Food security programmes therefore usually pair a low issue price with public procurement and rationing, so the cost appears as a budget subsidy rather than a shortage.
A price floor is a legal minimum price and binds only above equilibrium. In agriculture it takes the form of a minimum support price, intervention price or guaranteed price. With the same curves, a floor of 45 cuts quantity demanded to 20 and raises quantity supplied to 50, a surplus of 30. The surplus must be dealt with or the floor collapses. Four instruments are common across the world. First, the government buys the surplus at the floor (public procurement or intervention buying); here that costs 30 x 45 = 1350 and leaves stocks to store, distribute or export, often with a subsidy. Second, it restricts supply by quotas or acreage limits so that the floor becomes the market price. Third, it pays a deficiency payment: farmers sell everything at whatever price clears the market, and the government pays the gap between a target price and the market price. Here producers supply 50 at the target of 45, the market clears 50 units at 37.5, and the payment is (45 - 37.5) x 50 = 375; consumers gain because they pay 37.5, not 45. Fourth, it subsidises demand, for example by distributing the grain cheaply.
India's minimum support prices with procurement, the European Union's past intervention buying and quotas, and past United States deficiency payments are versions of these; details vary by country and change over time. They differ in who pays (consumers or taxpayers), in deadweight loss, and in side effects such as surplus stocks and distorted crop choice.
A production function states the maximum output obtainable from each combination of inputs with given technology: Q = f(L, K, ...). In the short run at least one input is fixed (land, a milking parlour, a tractor); in the long run all inputs are variable. With one variable input, total product (TP) is output, average product AP = TP/L and marginal product MP = dTP/dL. The law of variable proportions (diminishing marginal returns) says that as more of a variable input is added to fixed inputs, its marginal product eventually falls.
The classical production function gives three stages. Stage I runs from zero to maximum AP (where MP = AP); since MP exceeds AP, a rational producer would not stop there. Stage II runs from maximum AP to the point where MP = 0 (maximum TP); both AP and MP are positive and falling. Stage III lies beyond maximum TP, where MP is negative and adding input reduces output. Stage II is the rational stage; the exact point in it depends on prices, where the value of the marginal product equals the input price. For TP = 6L^2 - 0.2L^3, MP = 12L - 0.6L^2 peaks at L = 10, AP = 6L - 0.2L^2 peaks at L = 15, and MP reaches zero at L = 20, so Stage II lies between 15 and 20.
The elasticity of production is MP/AP, the percentage change in output for a 1 per cent change in the input. It exceeds 1 in Stage I, lies between 0 and 1 in Stage II and is negative in Stage III.
With two variable inputs, an isoquant joins input combinations that give the same output. Isoquants slope downward in the economic region, are convex to the origin, do not intersect, and higher isoquants mean more output. Their absolute slope is the marginal rate of technical substitution, MRTS(LK) = MPL/MPK, which falls as labour replaces capital along an isoquant. For Q = L^0.5 K^0.5 the MRTS is K/L; at L = 16 and K = 4 it is 0.25. Ridge lines join the points where isoquants turn vertical or horizontal (MPK = 0 or MPL = 0) and bound the economic region, which is the two-input equivalent of Stage II.
The elasticity of substitution (sigma) measures how easily one input replaces another: the percentage change in the input ratio K/L divided by the percentage change in the MRTS. Sigma is zero for fixed proportions (Leontief), infinite for perfect substitutes, and 1 for Cobb-Douglas. The constant elasticity of substitution (CES) function, Q = A[dK^-rho + (1-d)L^-rho]^(-v/rho), has sigma = 1/(1 + rho); with rho = 0.25, sigma = 0.8.
Returns to scale describe what happens when all inputs rise in the same proportion. If output rises by the same proportion the function shows constant returns; by more, increasing returns; by less, decreasing returns. For Cobb-Douglas Q = AL^a K^b, returns to scale equal a + b; Q = AL^0.6 K^0.3 shows decreasing returns (0.9), and each exponent is the output elasticity of that input. A function that is homogeneous of degree one satisfies Euler's theorem: L x MPL + K x MPK = Q, so paying each input its marginal product exactly exhausts output - the basis of marginal productivity theory of distribution.
A producer who wants a given output at least cost chooses the point where the isoquant is tangent to an isocost line (wL + rK = C). The condition is MRTS = w/r, or MPL/w = MPK/r: the last rupee spent on each input adds the same output. With Q = L^0.5 K^0.5, w = 4 and r = 16, the condition K/L = w/r gives K = 0.25L; producing 40 units needs 0.5L = 40, so L = 80, K = 20 and cost is 4 x 80 + 16 x 20 = 640.
The expansion path joins the least-cost points as output grows with input prices fixed. For a homothetic production function, such as Cobb-Douglas, the expansion path is a straight line from the origin: the input ratio does not change with scale. An isocline joins points of equal MRTS; the expansion path is the isocline for the ruling price ratio. When the relative price of labour rises, as it has with rural wage growth in many countries, the least-cost point moves along the isoquant towards capital - one economic force behind mechanisation of harvesting.
An economic cost includes explicit (paid-out) costs and implicit costs - the opportunity cost of owner-supplied inputs such as family labour, owned land and the farmer's own capital. Accounting profit subtracts only explicit costs; economic profit subtracts both. Normal profit is the return just sufficient to keep the entrepreneur in the business and is part of economic cost, so zero economic profit means normal profit is being earned.
In the short run, total cost (TC) = total fixed cost (TFC) + total variable cost (TVC). Average fixed cost (AFC = TFC/Q) falls continuously as output rises. Average variable cost (AVC), average total cost (ATC = AFC + AVC) and marginal cost (MC = dTC/dQ) are U-shaped because of the law of variable proportions: MC is at its minimum where MP is at its maximum, and AVC at its minimum where AP is at its maximum. MC cuts AVC and ATC at their minimum points, and the minimum of ATC lies to the right of the minimum of AVC because AFC keeps falling. If TFC is 500 and AVC is 20 at an output of 50, ATC is 500/50 + 20 = 30.
For TC = 100 + 10Q - 2Q^2 + 0.2Q^3, AVC = 10 - 2Q + 0.2Q^2 reaches its minimum at Q = 5, where AVC = 5 and MC = 10 - 4Q + 0.6Q^2 also equals 5. The minimum of ATC lies further right, at about Q = 8.5 with ATC about 19.2. Those two points are, as the next chapter shows, the shutdown and break-even points of a competitive firm.
In the long run every input, including plant size, is variable. The long-run average cost curve (LAC) is the lower boundary of all these SACs - the envelope curve (Viner and Wong). It touches each SAC at one point. Only at the minimum point of LAC does it touch an SAC at that SAC's minimum; to the left of it (falling LAC, economies of scale), LAC is tangent to SACs on their falling parts, so the optimal plant for such an output is used below its own minimum-cost output; to the right (rising LAC, diseconomies of scale), the tangency is on the rising parts of the SACs.
Long-run marginal cost (LMC) cuts LAC at its minimum, and at each output SMC equals LMC at the tangency point. LAC is U-shaped in the textbook because economies of scale are followed by diseconomies (managerial limits, supervision of hired labour). Farm LAC curves are often L-shaped. With fixed input prices, increasing returns to scale imply economies of scale, but economies can also be pecuniary, such as bulk discounts.
Economies of scope exist when producing two outputs together costs less than producing them separately: C(q1, q2) < C(q1, 0) + C(0, q2). Mixed crop-livestock farms rely on them.
The cost function C(w, r, Q) gives the minimum cost of producing Q at input prices w and r. Shephard's lemma in production states that the derivative of the cost function with respect to an input price is the conditional (cost-minimising) demand for that input. A cost function is non-decreasing and concave in input prices and homogeneous of degree one in them.
Perfect competition assumes many buyers and sellers, a homogeneous product, free entry and exit, perfect information and perfectly mobile resources. Each firm is a price taker facing a horizontal demand curve at the market price, so price = average revenue = marginal revenue. Many farm commodity markets come close. The firm maximises profit where MR = MC, with MC rising. With TC = 50 + 2Q + 0.5Q^2, MC = 2 + Q; at a price of 12 the firm produces 10, earning revenue 120 and total cost 120, so economic profit is zero; at a price of 14 it produces 12 and earns a profit of 168 - 146 = 22.
In the short run the firm continues to produce, even at a loss, as long as price covers average variable cost, because the fixed cost must be paid anyway. The shutdown point is the minimum of AVC; the break-even point is the minimum of ATC. The firm's short-run supply curve is the part of its MC curve above minimum AVC, and the market short-run supply is the horizontal sum of firms' supply curves. For TC = 100 + 10Q - 2Q^2 + 0.2Q^3 the shutdown price is 5. A farmer harvests a sown crop if the price covers the remaining harvest and marketing costs, even if not rent and seed.
In the long run, economic profit attracts entry and losses cause exit, until price equals the minimum of long-run average cost. At long-run equilibrium, P = MR = SMC = SAC = LMC = LAC, each firm operates at the minimum-cost scale and earns only normal profit. If LAC = Q^2 - 10Q + 40, minimum LAC is 15 at Q = 5; with market demand Q = 1000 - 20P, the market buys 700 units at P = 15, so the industry has 700/5 = 140 firms.
The long-run industry supply curve is horizontal in a constant-cost industry (entry does not change input prices), rising in an increasing-cost industry (entry bids up scarce inputs such as land, water or skilled labour, which is typical in agriculture) and falling in a decreasing-cost industry (external economies). It gives allocative efficiency (P = MC) and productive efficiency (minimum average cost), the benchmark for other structures.
A specific (per-unit) tax on sellers shifts the supply curve up by the tax. With Qd = 200 - 4P, Qs = -40 + 2P and a tax of 6, the new equilibrium is 200 - 4P = -40 + 2(P - 6), so consumers pay 42, producers receive 36 and 32 units are traded. Consumers bear 2 of the 6 and producers 4. Tax revenue is 6 x 32 = 192 and the deadweight loss is 0.5 x 6 x (40 - 32) = 24. The legal incidence (who writes the cheque) does not matter: a tax of 6 on buyers gives the same prices net of tax. The economic incidence depends on elasticities: the consumer share of the tax is Es/(Es - Ed), so the side of the market that is less elastic bears more. Here, at the original equilibrium, Ed = -4 and Es = 2, so consumers bear 2/6 = one-third.
A per-unit subsidy to producers shifts supply down by the subsidy. With a subsidy of 6, the market price falls to 38, producers receive 44, and 48 units are traded. The subsidy costs 6 x 48 = 288, consumer surplus rises by 88 and producer surplus by 176, and the deadweight loss is 288 - 264 = 24, from producing units whose cost exceeds their value to buyers. Input subsidies raise output but also encourage overuse of the subsidised input, such as fertiliser or pumping electricity.
A monopoly is a single seller of a product without close substitutes, protected by barriers to entry: control of a key resource, legal protection (patents on a seed trait, licences), economies of scale large enough to make one firm cheapest (natural monopoly), or network effects. The monopolist faces the downward-sloping market demand curve, so marginal revenue is below price: MR = P(1 + 1/e), where e is the price elasticity. For linear demand P = a - bQ, MR = a - 2bQ, falling twice as fast as demand. The monopolist maximises profit where MR = MC and charges the price on the demand curve at that output. It never operates on the inelastic part of demand, because there MR is negative.
With P = 100 - 2Q and MC = 20, MR = 100 - 4Q = 20 gives Q = 20 and P = 60, with profit (60 - 20) x 20 = 800. Under competition, P = MC = 20 and Q = 40. The deadweight loss is 0.5 x (60 - 20) x (40 - 20) = 400. The Lerner index of monopoly power is (P - MC)/P = 1/|e| at the profit-maximising point; here it is 40/60 = 0.67, matching an elasticity of 1.5 at that point. With elasticity -3 and MC = 40, the markup rule P = MC/(1 - 1/|e|) gives P = 60.
A monopolist has no supply curve. Natural monopolies are usually regulated, often with average-cost pricing, because marginal-cost pricing would leave them with a loss.
Price discrimination is charging different prices for the same good for reasons not related to cost. It needs market power, the ability to separate buyers, and no resale between them. First-degree discrimination charges each buyer his or her reservation price, captures all consumer surplus and gives the efficient output. Second-degree discrimination charges by quantity or version, as in block tariffs for irrigation electricity or bulk discounts on inputs. Third-degree discrimination charges different groups different prices: domestic versus export markets, urban versus rural buyers, or fresh versus processing uses of milk.
Under third-degree discrimination the firm sets MR equal in every market and equal to MC, so the market with less elastic demand gets the higher price. With MC = 30, a domestic market with elasticity -2 and an export market with elasticity -4, the prices are 30/(1 - 1/2) = 60 at home and 30/(1 - 1/4) = 40 abroad. Selling abroad more cheaply than at home in this way is one economic explanation of dumping.
Chamberlin's (and, independently, Joan Robinson's imperfect competition) model has many sellers, free entry, and differentiated products - differences in brand, packaging, location, quality or service. Each firm faces a downward-sloping but fairly elastic demand curve. In the short run a firm may earn profit, like a monopolist, by setting MR = MC. In the long run, entry of close substitutes shifts each firm's demand curve left until it is tangent to the average cost curve: price equals average cost and economic profit is zero (Chamberlin's tangency solution).
Because the tangency is on the falling part of the average cost curve, each firm produces less than the output that minimises average cost. The difference is called excess capacity, and price exceeds marginal cost, so the outcome is neither productively nor allocatively efficient. Against this, consumers gain from variety. Branded packaged foods and farm input retail are examples; much agribusiness strategy consists of turning an undifferentiated commodity into a differentiated product.
An oligopoly is a market with a few sellers whose decisions affect each other: fertiliser and farm machinery manufacture, commercial seed, grain export trading and large-scale meat packing are concentrated in several countries. Because each firm's best choice depends on what its rivals do, oligopoly is analysed with game theory. A Nash equilibrium is a set of strategies, one for each player, such that no player can gain by changing strategy while the others keep theirs. A dominant strategy is best whatever the rivals do. The prisoners' dilemma shows why oligopolists find it hard to keep a cartel: each firm's dominant strategy is to cheat (produce more or price lower), so the Nash equilibrium leaves both worse off than if they had cooperated. Repeated interaction and the threat of punishment can sustain cooperation.
Sweezy's kinked demand curve model explains price rigidity: rivals are assumed to match a price cut but not a price rise, so demand is elastic above the current price and inelastic below it. MR has a gap at the kink, and marginal cost can move within that gap without changing price or output.
In the Cournot model, firms choose quantities simultaneously, each taking the other's output as given. Each firm's reaction (best-response) function gives its profit-maximising output for each output of the rival; the Cournot-Nash equilibrium is where the reaction functions cross. With market demand P = 120 - Q and constant marginal cost 30 for both firms, each firm produces (120 - 30)/3 = 30, total output is 60 and price is 60, and each firm earns 900. A monopolist would produce 45 at a price of 75; competition would produce 90 at a price of 30. Cournot output lies between the two. With n identical firms, total Cournot output is n/(n + 1) of the competitive output, so with four firms it is 72 and price 48, and as n grows the outcome approaches perfect competition.
In the Stackelberg model one firm, the leader, chooses output first, knowing the follower will respond along its reaction function. With the same demand and costs, the leader produces the monopoly output of 45, the follower 22.5, total output 67.5 and price 52.5. The leader earns 1012.5 and the follower 506.25: there is a first-mover advantage, and total output is higher and price lower than under Cournot.
In the Bertrand model, firms choose prices simultaneously and sell identical products. Any price above marginal cost invites a rival to undercut slightly and take the whole market, so with two firms and equal constant marginal costs the only Nash equilibrium is P = MC - the competitive outcome with just two firms. This is the Bertrand paradox. It is resolved by product differentiation (then prices exceed MC), capacity constraints (Edgeworth's model, in which prices may cycle), or repeated interaction. The comparison of models teaches that the strategic variable matters: competition in quantities (Cournot) gives prices above MC; competition in prices with homogeneous goods gives P = MC.
A cartel coordinates output to maximise joint profit, behaving like a multi-plant monopoly that sets MR equal to each member's MC; but each member has an incentive to exceed its quota. Competition law in most countries prohibits such agreements, with exemptions for farmer cooperatives that differ by jurisdiction. Market concentration is measured by concentration ratios (the share of the largest four or eight firms) and by the Herfindahl-Hirschman Index, the sum of squared market shares.
The demand for an input - land, labour, water, fertiliser, machine services - is a derived demand: it exists because of the demand for the output it helps produce. A firm hires an input up to the point where the extra revenue it brings equals its extra cost. The extra revenue from one more unit of labour is the marginal revenue product, MRP = MPL x MR. In a competitive output market MR = P, so MRP equals the value of the marginal product, VMP = P x MPL. A competitive firm hiring in a competitive labour market employs labour until VMP = w. If MPL = 20 - 0.5L, output price is 10 and the wage is 100, then 200 - 5L = 100 gives L = 20.
The firm's input demand curve is its VMP (or MRP) curve over the relevant range, and the market demand for an input is less elastic when the input has few substitutes, when it is a small share of total cost, and when demand for the final product is inelastic - Marshall's rules of derived demand, refined by Hicks. Under output monopoly, MRP lies below VMP and less is hired.
A monopsony is a single buyer of an input or a product. A sugar mill that is the only buyer of cane within a feasible transport distance, a single processor collecting milk in a remote area, or the only large employer of harvest labour in a district each has monopsony power. The monopsonist faces an upward-sloping supply curve, so the marginal expenditure (marginal factor cost) of hiring one more unit exceeds its price, because the higher price must be paid on all units. With labour supply w = 20 + 2L, marginal expenditure is 20 + 4L; with MRP = 200 - 2L, the monopsonist hires where 200 - 2L = 20 + 4L, so L = 30, and pays the wage from the supply curve, w = 80, although the MRP of the last worker is 140. A competitive market would employ 45 at a wage of 110. The gap between MRP and the wage is monopsonistic exploitation (Joan Robinson).
One striking result follows: a minimum wage or minimum purchase price set between the monopsony price and the competitive price can raise both the price and the quantity bought, because it makes the marginal expenditure curve flat up to the supply curve. When a monopsony buyer faces a monopoly seller (a single processor dealing with a strong cooperative), the result is bilateral monopoly and the price is indeterminate within a range set by bargaining power.
Ricardo explained land rent by differences in fertility and location: as population grows, cultivation extends to poorer land, and the better land earns a surplus over the marginal (no-rent) land. In Ricardo's view rent is price-determined, not price-determining. Modern theory generalises the idea. Transfer earnings are the minimum payment a factor needs to stay in its present use (its opportunity cost). Economic rent is any payment above transfer earnings. On a supply-and-demand diagram for a factor, economic rent is the area above the supply curve and below the price. With labour supply w = 20 + 2L and an equilibrium of 30 workers at a wage of 80, economic rent is 0.5 x 30 x 60 = 900. If supply is perfectly inelastic, as the total supply of land broadly is, the whole payment is economic rent; if it is perfectly elastic, the whole payment is transfer earnings.
Marshall's quasi-rent is the short-run return to factors that are fixed for the time being, such as machines, orchards, or a cold store: total revenue minus total variable cost. A cold store that earns revenue of 5000 with variable costs of 3000 and fixed costs of 1200 earns a quasi-rent of 2000 and an economic profit of 800. Quasi-rent can exceed or fall short of fixed cost; in the long run, when the fixed factor can be varied or leave, quasi-rent disappears into normal returns or leads to disinvestment.
Partial equilibrium studies one market with others held fixed. General equilibrium studies all markets together, recognising that a change in one - a drought raising cereal prices - affects demand for other foods, farm wages, land rents and the incomes that determine demand in every market. Walras' law states that the value of excess demands across all markets sums to zero, so if all markets but one clear, the last clears too, and only relative prices are determined. Arrow and Debreu proved existence under conditions including convexity; computable general equilibrium models are the applied form.
An allocation is Pareto efficient (Pareto optimal) if no one can be made better off without making someone worse off. Pareto efficiency has three parts. Efficiency in exchange: every consumer has the same MRS between any two goods; in the Edgeworth box, this holds along the contract curve where indifference curves are tangent. Efficiency in production: every producer has the same MRTS between any two inputs, giving the production contract curve and, from it, the production possibility frontier. Overall (product-mix) efficiency: the MRS common to consumers equals the MRT along the frontier. In an exchange economy of two people with identical preferences U = xy, one owning 10 units of rice and the other 10 units of wheat, the competitive equilibrium has a price ratio of 1 and each person consuming 5 of each.
The first fundamental theorem states that, under stated conditions (complete markets, no externalities, perfect information, price-taking behaviour and locally non-satiated preferences), every competitive equilibrium is Pareto efficient. It formalises Adam Smith's invisible hand. The second fundamental theorem states that, with convex preferences and technologies, any Pareto efficient allocation can be reached as a competitive equilibrium after a suitable lump-sum redistribution of initial endowments. Together they suggest a division of labour: use markets for efficiency and lump-sum transfers for equity.
Pareto efficiency says nothing about fairness: there are infinitely many efficient allocations, including very unequal ones. A social welfare function (utilitarian, Rawlsian or general Bergson-Samuelson) ranks them. Arrow's impossibility theorem shows that no rule for turning individual rankings into a social ranking can satisfy a small set of reasonable conditions (unrestricted domain, Pareto principle, independence of irrelevant alternatives and non-dictatorship) at once. Policy analysis therefore often uses compensation tests: the Kaldor criterion asks whether gainers could compensate losers and still be better off; the Hicks criterion asks whether losers could not bribe gainers to prevent the change; Scitovsky showed both can give contradictory answers and proposed a double test. Cost-benefit analysis applies the Kaldor-Hicks idea. The theory of the second best (Lipsey and Lancaster) warns that when one distortion cannot be removed, removing others may not raise welfare - a caution for piecemeal agricultural policy reform.
Markets fail to reach Pareto efficiency when the theorems' conditions are broken. An externality is a cost or benefit that falls on others without payment. Pesticide drift, nutrient run-off and stubble burning are negative externalities; pollination services are a positive one. With a negative externality, the market produces where marginal private cost equals marginal benefit, too much compared with the social optimum where marginal social cost equals marginal benefit. If demand is P = 100 - Q, marginal private cost is 10 + Q and the marginal external cost is 4, the market produces 45 while the optimum is 43; the deadweight loss is 0.5 x 4 x 2 = 4. A Pigouvian tax equal to the marginal external cost at the optimum (here 4) corrects it. Coase showed that if property rights are clear and transaction costs low, bargaining between parties reaches the efficient outcome whoever holds the right, although the distribution differs; high transaction costs limit this.
Public goods are non-rival (one person's use does not reduce another's) and non-excludable (people cannot easily be kept from benefiting). Agricultural research, weather forecasts and pest surveillance are examples. Free riding leads markets to under-supply them. The efficient quantity is where the vertical sum of individuals' marginal benefits equals marginal cost (Samuelson's condition): if two farmers value a drainage channel at 50 - Q and 30 - Q and marginal cost is 40, the efficient size is where 80 - 2Q = 40, so Q = 20. Common-pool resources (groundwater, fisheries, village grazing land) are rival but hard to exclude from, leading to over-use - the tragedy of the commons - which communities can sometimes govern through local rules (Ostrom).
Asymmetric information is another source of failure. Adverse selection arises when one side knows more before a contract: Akerlof's 'market for lemons' shows how buyers who cannot judge quality may drive good-quality sellers out, as in seed or used-machinery markets and crop insurance. Moral hazard arises after the contract when one party's actions are hidden: an insured farmer may take less care, and a borrower may take more risk. Responses include signalling (certification, grades), screening (deductibles, collateral) and contract design such as sharecropping.
Government intervention can also fail, through deadweight losses and rent-seeking, so each intervention should be judged against realistic alternatives.
Use this subject as the grammar for every other agricultural economics course. When you meet a policy (a support price, a fertiliser subsidy, a tariff) draw the supply-demand diagram and ask who gains, who loses and how large the deadweight loss is. Practise the standard numericals until they take under a minute each: point and arc elasticity, Cobb-Douglas demand, Slutsky decomposition, stages of production, least-cost combination, shutdown price, monopoly output and deadweight loss, Cournot and Stackelberg outputs, monopsony wage. For the PhD paper, add duality results, the welfare theorems and compensation criteria, and be able to state the assumptions behind each result, because examiners test the conditions as often as the results.
An analyst in an agribusiness firm, bank, cooperative or government department uses these tools weekly. Price elasticities decide whether a price cut on a packaged product will raise revenue; cost curves and the shutdown rule decide whether to keep a processing line running in a lean season; the markup rule and Lerner index frame pricing across segments; the incidence formula predicts how much of a new levy or subsidy will reach farmers. A credit officer uses break-even and shutdown logic to judge whether a borrower's enterprise will survive a price fall. A procurement manager facing a few rival buyers reasons like a Cournot or Stackelberg player when deciding capacity and purchase volumes.
A senior manager or policy adviser uses micro theory to structure decisions about markets rather than single transactions: whether a support price should be delivered by procurement or by deficiency payment, and what each costs the budget and consumers; whether a merger among input suppliers or processors will create market power over farmers; whether a statutory minimum price for cane or milk can raise both price and volume in a monopsony setting; how to price across domestic and export markets without inviting anti-dumping action. The value at this level is knowing which model fits the market in front of you, and knowing the conditions under which the textbook result stops holding.
A researcher with long experience uses the theory as the basis for estimation and for judging policy. Demand systems are estimated with homogeneity, adding-up and symmetry restrictions; welfare effects of food price shocks are measured with compensating and equivalent variation; farm-household models combine consumer and producer theory where markets are missing; market power in value chains is tested with conduct parameters and concentration measures; externalities from agriculture are valued for Pigouvian or payment-for-ecosystem-services design; computable general equilibrium models trace economy-wide effects of trade and price policy. The researcher's job is to state the assumptions a policy conclusion depends on, test them against data, and say plainly where the evidence is weak.
The manual gives no country-specific estimates of food demand elasticities, supply response elasticities or support-price levels, because these vary by country, commodity, income group and period and must be taken from current studies and official releases. It covers game theory only as far as oligopoly models need (no mixed strategies, extensive-form games or auction theory), treats choice under uncertainty and expected utility only in passing (see the production economics and risk manuals), and does not derive the existence proofs of general equilibrium, behavioural economics, or mechanism design. Diagrams are described in words because figures are not yet supported.
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. Every numerical answer was computed by script before it was written.
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.