Agribusiness Management · Foundations · also ABM 503, Applied Agribusiness Economics, Micro Economics (NIAM)
Economic theory put to work for business decisions: demand, production, cost and pricing, with a short macro block. Named explicitly in agribusiness master's entrance syllabi.
In the exams: agricultural master's entrance examinations, agribusiness (heavy); agribusiness management PhD entrance and fellowship examinations (heavy); agricultural master's entrance examinations, agricultural economics (heavy)
Managerial economics is economic theory put to work on the decisions a manager actually has to take: what price to charge, how much to produce, which inputs to buy, whether to invest, how to respond to a rival and how to read the economy the firm sells into. What makes it managerial is the question it asks. Economics asks how markets allocate resources; managerial economics asks what this firm should do next, given what it knows and what it cannot know.
Agribusiness gives the subject some of its sharpest cases. Farm output depends on weather and biology, so supply arrives in lumps and with a lag. Demand for staple foods is price-inelastic, so a good harvest can lower farm revenue. Processors of milk, sugarcane, cotton and meat turn one raw material into several joint products whose costs cannot be separated. Many input and output markets have a few large buyers or sellers facing many small farmers, so pricing power and strategic behaviour matter. Cooperatives and farmer producer organisations have objectives that differ from those of an investor-owned firm.
The subject is examined in three ways. Entrance papers for agribusiness and agricultural economics Master's programmes, such as ICAR AIEEA-PG Group 19 Unit IX and Group 06 Unit II, test definitions, elasticity, cost curves, market structures and one-step numericals. PhD and recruitment papers, such as ICAR AICE-JRF/SRF (Agri-Business Management Unit 6) and ASRB NET, go further: optimisation with calculus, price discrimination rules, Lerner index, game solutions, forecasting methods and macro theory. This manual covers all three depths, uses agribusiness examples from several countries, and states where practice or policy differs by country.
Managerial economics applies economic reasoning to business decisions. Its core principles are few. The opportunity cost principle says that the cost of any choice is the value of the best alternative given up: a dairy farmer who feeds home-grown maize to cows gives up the price that maize would have fetched in the market. The incremental principle says that a decision is judged by the change it causes in revenue and cost, not by averages: a feed mill should accept an extra order if the extra revenue exceeds the extra cost, even if the price is below average total cost, provided fixed costs are unaffected. The marginal principle is the incremental principle for small changes: output should expand while marginal revenue exceeds marginal cost. The equi-marginal principle says that a limited resource is best allocated when the marginal return per unit is equal in every use, which is how a farm should split irrigation water or credit across crops. Accounting profit is revenue minus explicit costs. Economic profit also subtracts implicit costs, such as the owner's own labour, own land and own capital valued at their opportunity cost. Normal profit is the minimum return that keeps the owner in the business; it is counted as a cost, so zero economic profit means the owner earns exactly a normal return.
The traditional theory assumes the firm maximises profit. Because profits arrive over many years and are uncertain, modern finance restates the goal as maximising the value of the firm, the present value of expected future profits, which for a listed company is its share value. Several alternative theories arose because ownership and control are separated in large firms, so managers may pursue their own goals within limits set by shareholders. Baumol's sales revenue maximisation model says managers maximise total revenue subject to a minimum profit constraint, which leads to higher output and lower price than profit maximisation. Williamson's managerial discretion model says managers maximise a utility function that includes staff expenditure, managerial emoluments and discretionary investment. Simon argued that, because information and computing ability are limited, managers satisfice: they seek a satisfactory rather than an optimal result. The principal-agent problem describes the underlying issue: owners (principals) cannot perfectly observe managers (agents), so contracts, bonuses tied to profit and monitoring are used to align interests. Cooperatives and farmer producer organisations are a different case: the owners are also the suppliers or customers, so the usual objective is to maximise the returns members receive, through better prices or patronage refunds, rather than the surplus kept by the cooperative.
Most managerial problems are optimisation problems: maximise profit, minimise cost, or maximise output from a budget. The derivative of a function measures its slope, the rate at which the dependent variable changes as the independent variable changes by a small amount; in economics a marginal value is a derivative. Marginal revenue is dTR/dQ and marginal cost is dTC/dQ. To find a maximum or minimum, set the first derivative equal to zero (first-order condition) and check the sign of the second derivative (second-order condition): negative for a maximum, positive for a minimum. For profit, setting d(TR - TC)/dQ = 0 gives MR = MC, and the second-order condition requires MC to cut MR from below. For example, if TR = 100Q - Q squared and TC = 200 + 20Q, then MR = 100 - 2Q and MC = 20, so Q = 40, price = 100 - 40 = 60 and profit = 2,400 - 1,000 = 1,400. When a function has several inputs, the partial derivative measures the effect of one while holding the others constant, which is how marginal products are found. Constrained optimisation uses the Lagrange multiplier method: the objective and the constraint are combined into one function, and the multiplier measures how much the optimum improves if the constraint is relaxed by one unit, the shadow price of the constraint. Linear programming solves problems with linear objectives and many inequality constraints, such as least-cost feed formulation or farm plans.
Demand starts with consumer choice. Cardinal utility theory, associated with Marshall, assumed utility could be measured in units; ordinal theory, developed by Hicks and Allen, needs only that the consumer can rank bundles. An indifference curve joins all bundles of two goods that give the consumer the same satisfaction. Under the usual assumptions (completeness, transitivity, non-satiation and a diminishing marginal rate of substitution) indifference curves have four properties: they slope downward, because more of one good must be offset by less of the other; they are convex to the origin, because the marginal rate of substitution diminishes; they cannot intersect, because intersection would contradict transitivity; and a curve further from the origin represents a higher level of satisfaction. The marginal rate of substitution of X for Y (MRS) is the amount of Y the consumer will give up for one more unit of X while remaining equally satisfied; it equals the absolute slope of the indifference curve and the ratio of marginal utilities MUx/MUy.
The budget set is every bundle the consumer can afford. With income M and prices Px and Py, the budget line is Px X + Py Y = M. Its intercepts are M/Px and M/Py, and its slope is -Px/Py. The consumer is in equilibrium where the budget line touches the highest attainable indifference curve, so MRS = Px/Py, equivalently MUx/Px = MUy/Py: the last rupee, shilling or dollar spent on each good yields the same satisfaction. The total effect of a price change splits into a substitution effect, the change in quantity from the change in relative prices with real income held constant, which is always negative for a price rise, and an income effect, the change caused by the change in purchasing power. For a normal good both effects work in the same direction. For an inferior good the income effect works against the substitution effect. A Giffen good is an inferior good whose income effect is so strong that it outweighs the substitution effect, so quantity demanded rises when price rises; it is usually illustrated with a staple food that takes a large share of a very poor household's budget. Hicks holds real income constant at the original indifference curve; Slutsky holds it constant by letting the consumer just afford the original bundle.
Standard theory assumes consumers and managers are fully rational, have stable preferences and process all available information. Behavioural economics tests these assumptions against observed behaviour and builds psychology into the models. Herbert Simon's idea of bounded rationality says people have limited information, time and computing capacity, so they use rules of thumb. Kahneman and Tversky documented systematic biases in those rules: anchoring (estimates are pulled toward an initial number, such as last season's price), availability (events that are easy to recall, such as last year's drought, are judged more likely) and representativeness (judging probability by resemblance, ignoring base rates). Their prospect theory says people evaluate outcomes as gains and losses relative to a reference point rather than as final wealth, that losses weigh more heavily than equal gains (loss aversion; many experiments find losses weigh roughly twice as much), that people are risk-averse over gains and risk-seeking over losses, and that small probabilities are overweighted. Framing effects follow: the same choice described as a loss or as a forgone gain draws different decisions. Present bias, often modelled as hyperbolic discounting, means people give extra weight to the present, so they plan to save or buy inputs later and then do not. Field experiments in Kenya by Duflo, Kremer and Robinson found that offering farmers a small, time-limited discount to buy fertiliser right after harvest, when they had cash, raised fertiliser use, consistent with present bias.
Demand is the quantity buyers are willing and able to purchase at each price over a period. The law of demand says that, other things equal, quantity demanded falls as price rises. The demand function lists what else matters: the prices of substitutes and complements, consumer income, tastes and preferences, the size and composition of the population, expectations about future prices and income, availability of credit, and, for many firms, advertising. A change in the good's own price causes a movement along the demand curve; a change in any other determinant shifts the whole curve. Demand for an input is derived demand: farmers want fertiliser, tractors or farm labour because of the output they help produce, so input demand depends on output prices and on the input's productivity. In the same way, demand for farm products at the farm gate is derived from retail demand, after subtracting marketing and processing margins. Exceptions to the law of demand include Giffen goods, Veblen or prestige goods bought partly because they are expensive, and cases where buyers expect prices to rise further, as in speculative stocking of grain during a shortage.
Price elasticity of demand measures how responsive quantity demanded is to price: the percentage change in quantity divided by the percentage change in price. Because the law of demand makes it negative, it is often quoted in absolute value. Point elasticity is (dQ/dP) x (P/Q). Arc elasticity, used between two observed points, divides changes by the averages of the two prices and two quantities (the midpoint method) so that the answer does not depend on direction. For example, if the price of tomatoes falls from 20 to 16 per kg and sales rise from 100 to 120 tonnes, arc elasticity = (20/110) / (-4/18) = -0.82, inelastic. Demand is elastic when the absolute value exceeds one, unit elastic at one and inelastic below one. The link to revenue is what matters to managers: when demand is elastic a price cut raises total revenue; when it is inelastic a price cut lowers total revenue; at unit elasticity revenue is at its maximum. Formally MR = P(1 - 1/|e|), so with price 50 and elasticity -2, MR = 25. On a straight-line demand curve elasticity is not constant: it is elastic above the midpoint, unit elastic at the midpoint and inelastic below it. Staple foods taken as a group are price-inelastic. This is the basis of the farm problem, often called King's law: a bumper harvest pushes prices down more than proportionately, so total farm revenue can fall in a good year.
Income elasticity of demand is the percentage change in quantity divided by the percentage change in income. It is positive for normal goods and negative for inferior goods. Normal goods with income elasticity between zero and one are necessities; those above one are luxuries in the economist's sense. Engel's law, from Ernst Engel's nineteenth-century household budget studies, says that as household income rises the share spent on food falls, even though the absolute amount rises. In most countries income elasticities are lower for cereals than for fruit, vegetables, dairy, meat and processed foods, which is why diets diversify as incomes grow and why value-added food processing expands faster than staple production. Exact values differ by country, income group and period, so managers should use recent estimates for their own market. Cross elasticity of demand is the percentage change in quantity of good X divided by the percentage change in the price of good Y. It is positive for substitutes (two edible oils), negative for complements (tractors and diesel, or tea and sugar) and near zero for unrelated goods. Promotional or advertising elasticity is the percentage change in sales divided by the percentage change in advertising spending.
Demand forecasting estimates future sales so that production, procurement, inventory, staffing, finance and capacity can be planned. A seed company must decide months ahead how much hybrid seed to produce; a cold-storage operator must decide how much capacity to build; a dairy must plan milk procurement and processing across flush and lean seasons. Short-term forecasts, usually up to a year, guide production scheduling, purchasing and working capital. Long-term forecasts guide investment in plants, brands and distribution. Agricultural demand forecasting has special features: strong seasonality, weather-driven supply shocks that move prices and hence quantities, policy events such as procurement prices, export restrictions or subsidy changes, and festivals or religious calendars that shift consumption. A useful rule for long-run food demand projection combines population and income growth: growth in demand is approximately population growth plus income elasticity multiplied by growth in per capita income. With population growing at 1 percent a year, per capita income at 5 percent and an income elasticity of 0.4, demand grows at about 1 + 0.4 x 5 = 3 percent a year. Price changes and changes in tastes are ignored in this simple rule, which is its limitation.
Qualitative or judgemental methods are used when data are scarce, as for a new product, or when structural change makes history a poor guide. The survey of buyers' intentions asks customers what they plan to buy; it works well for industrial buyers and large institutional customers, such as asking poultry integrators about their feed requirements, and poorly for many small consumers who do not follow through on stated plans. The expert opinion method collects views of dealers, consultants and industry specialists. The Delphi method, developed at the RAND Corporation, refines expert opinion by collecting anonymous forecasts in rounds, feeding back the group summary each time, so that experts can revise their views without the pressure of face-to-face discussion; it is often used for long-range and technology forecasting. The sales force composite method adds up the estimates of salespeople for their territories; it uses people close to customers but may be biased if forecasts become targets. Market experiments, including test marketing in selected towns and controlled laboratory experiments, observe how buyers respond to price, packaging or promotion before a national launch.
Trend projection fits a line or curve to past sales against time, usually by least squares: a linear trend Y = a + bt adds a constant amount each period, while an exponential trend adds a constant percentage. Moving averages smooth out random variation by averaging the last n observations; they lag behind turning points. Simple exponential smoothing updates the forecast as a weighted average of the latest actual and the previous forecast: F(t+1) = alpha x A(t) + (1 - alpha) x F(t), where alpha lies between 0 and 1. With alpha = 0.3, last forecast 100 and actual 110, the next forecast is 103. A higher alpha reacts faster to change but follows noise more. Seasonal indices, often calculated by the ratio-to-moving-average method, adjust forecasts for regular within-year patterns, which are strong for milk, vegetables and fruit. Barometric methods use indicators: leading indicators change before the series of interest (for example, rainfall and reservoir levels before tractor and input sales in rain-fed regions), coincident indicators move with it and lagging indicators move after it. Econometric methods estimate a demand function by regression, relating quantity to price, income, prices of related goods and other variables; they explain as well as forecast, but face problems such as the identification problem (observed price-quantity points are intersections of shifting supply and demand curves) and multicollinearity. Forecast accuracy is measured by the mean absolute deviation (MAD), the mean squared error (MSE), which penalises large errors more heavily, and the mean absolute percentage error (MAPE), which allows comparison across products of different sizes.
A production function shows the maximum output obtainable from each combination of inputs with a given technology, Q = f(L, K, land, materials). In the short run at least one input is fixed, such as land or a milk processing plant; in the long run all inputs are variable. With one variable input, total product (TP) is total output, average product (AP) is TP divided by the input, and marginal product (MP) is the addition to TP from one more unit of the input. The law of variable proportions (diminishing returns) says that as more of a variable input is added to fixed inputs, MP eventually declines. Three stages follow. In stage I, AP is rising and MP exceeds AP; the fixed input is under-used, so a rational producer expands. Stage II runs from the point where AP is at its maximum (MP = AP) to the point where MP = 0 (TP at maximum); this is the rational stage. In stage III, MP is negative and TP falls, as when too much nitrogen lodges a crop or too many workers crowd a small packhouse. Within stage II, the exact profit-maximising level of a single variable input is where the value of its marginal product equals its price: VMP = Py x MPx = Px, equivalently MPx = Px/Py. If maize sells at 20 per kg and nitrogen costs 60 per kg, nitrogen should be applied until one more kg adds 3 kg of grain.
With two variable inputs, an isoquant joins all combinations that produce the same output. Isoquants slope downward, are convex to the origin and do not intersect. The marginal rate of technical substitution (MRTS) of labour for capital is the amount of capital that can be replaced by one more unit of labour with output unchanged; it equals the slope of the isoquant and the ratio MPL/MPK, and it diminishes along the isoquant. An isocost line shows all input combinations with the same total cost: wL + rK = C, slope -w/r. The least-cost combination for a given output is where the isoquant is tangent to the isocost line: MRTS = w/r, or MPL/w = MPK/r, meaning the last unit of money spent on each input adds the same output. If MPL/w is greater than MPK/r, the firm should shift spending toward labour. The line joining least-cost points as the budget grows is the expansion path. When input prices change, the firm substitutes toward the input that has become relatively cheaper, which is why rising farm wages drive mechanisation of harvesting and weeding in many countries.
Returns to scale describe what happens to output when all inputs are increased in the same proportion, a long-run concept. Increasing returns mean output rises more than proportionately, constant returns mean it rises in the same proportion and decreasing returns mean it rises less than proportionately. A function is homogeneous of degree n if multiplying every input by t multiplies output by t to the power n; n greater than, equal to or less than one indicates increasing, constant or decreasing returns. The Cobb-Douglas function Q = A L^a K^b is the workhorse of applied production economics. Its exponents are the output elasticities of each input, their sum a + b gives returns to scale, its elasticity of substitution between inputs equals one, and taking logarithms makes it linear, so it is easy to estimate by regression. Euler's theorem states that for a function homogeneous of degree one, if each input is paid its marginal product, total output is exactly exhausted. Returns to scale should not be confused with economies of scale, which refer to falling long-run average cost and can also arise from bulk buying of inputs, cheaper finance or marketing spread over a larger output.
Different decisions need different cost concepts. Explicit costs are cash payments; implicit costs are the opportunity costs of owner-supplied resources. Opportunity cost is the value of the best alternative forgone. Sunk costs have already been incurred and cannot be recovered, so they are irrelevant to future decisions: the money spent building a packhouse should not decide whether to keep running it. Incremental or differential costs are the changes in cost from a decision. Fixed costs do not vary with output in the short run (rent, insurance, depreciation by time, salaried staff); variable costs do (seed, feed, fertiliser, fuel, casual labour). In India, the Commission for Agricultural Costs and Prices uses a ladder of farm cost concepts: A1 covers paid-out costs such as hired labour, owned and hired bullock and machine labour, seed, fertiliser, manure, plant protection, irrigation, depreciation, land revenue and interest on working capital; A2 adds rent paid for leased-in land; B1 adds interest on owned fixed capital excluding land; B2 adds the rental value of owned land; C1 adds imputed family labour to B1; C2 adds imputed family labour to B2; C2* values labour at the higher of market or statutory wage; and C3 adds 10 percent of C2* for management.
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 = change in TC per unit change in output) are usually U-shaped, because of the law of variable proportions: MC = w/MPL, so MC falls while marginal product rises and rises when marginal product diminishes. MC cuts both AVC and ATC at their minimum points, and the minimum of AVC occurs at a lower output than the minimum of ATC, because AFC keeps falling. In the long run all inputs are variable and the firm chooses the plant size. The long-run average cost (LAC) curve is the envelope of short-run average cost curves; it touches each SAC at one point and, when U-shaped, touches the minimum of an SAC only at the minimum of the LAC itself. The falling part reflects economies of scale and the rising part diseconomies. The minimum efficient scale relative to market size is a key determinant of how many firms an industry can support, which is why a region may support only one or two sugar mills or milk-drying plants. Economies of scope exist when producing two products together costs less than producing them separately, as when a feed mill makes cattle and poultry feed on shared equipment.
Cost-volume-profit (break-even) analysis assumes linear revenue and cost over the relevant range. Contribution per unit is price minus average variable cost; it is what each unit contributes toward fixed costs and then profit. The break-even output is fixed cost divided by contribution per unit. A fruit-juice unit with fixed costs of 600,000 a year, a price of 50 per litre and variable cost of 30 per litre breaks even at 600,000 / 20 = 30,000 litres. The profit-volume (P/V) ratio, or contribution margin ratio, is contribution divided by price (here 0.4), and break-even sales revenue is fixed cost divided by the P/V ratio (1,500,000). The output needed for a target profit is (fixed cost + target profit) / contribution per unit. The margin of safety is actual or planned sales minus break-even sales, often expressed as a percentage of actual sales: at 40,000 litres it is 25 percent. The degree of operating leverage, total contribution divided by operating profit, measures how sharply profit responds to a change in sales; firms with high fixed costs, such as cold chains and processing plants, have high operating leverage, which raises both potential profit and risk. Economic theories of why profit exists include profit as a reward for bearing uncertainty that cannot be insured (Knight), profit as a reward for innovation that later competition erodes (Schumpeter), and monopoly profit arising from market power.
Supply is the quantity producers are willing and able to sell at each price. Perfect competition assumes many small sellers and buyers, a homogeneous product, free entry and exit, perfect information and perfect mobility of resources. Each firm is a price taker facing a horizontal demand curve, so price = average revenue = marginal revenue. The firm maximises profit where P = MC with MC rising. In the short run it may make profit, normal profit or loss. It continues to produce at a loss as long as price covers average variable cost, because the loss is then smaller than fixed cost; below the minimum of AVC, the shutdown point, it stops. The firm's short-run supply curve is therefore its MC curve above minimum AVC.
In the long run firms can change plant size and can enter or leave the industry. Economic profit attracts entry, which increases supply and lowers price; losses cause exit. Long-run equilibrium occurs when price equals minimum long-run average cost, so P = MR = SMC = SAC = LMC = LAC, and each firm earns only normal profit. The firm's long-run supply is its LMC above minimum LAC. The long-run industry supply curve depends on what happens to input prices as the industry expands. In a constant-cost industry input prices do not change, so long-run supply is horizontal at minimum LAC. In an increasing-cost industry expansion bids up scarce inputs such as suitable land or skilled labour, so long-run supply slopes upward; most agricultural industries are of this kind because good land is limited. In a decreasing-cost industry, external economies such as better infrastructure or specialised suppliers lower costs as the industry grows, giving a downward-sloping long-run supply curve. Price elasticity of supply, the percentage change in quantity supplied divided by the percentage change in price, is higher in the long run than in the short run, because producers have more time to adjust land, herds, orchards and machinery.
Farm supply responds to price with a lag, since planting decisions are made on expected prices and output arrives a season or more later. The cobweb model assumes supply this period depends on last period's price while demand depends on the current price. A shortage raises the price, farmers plant more, the resulting glut lowers the price, farmers plant less, and so on. The oscillation converges when supply is less elastic than demand (the supply curve is steeper than the demand curve in absolute slope), diverges when supply is more elastic than demand, and continues with constant amplitude when the two are equal. Marc Nerlove's supply response model replaces naive expectations with adaptive expectations and partial adjustment: farmers move only part of the way toward the desired area each year because of constraints on land, credit, labour and habit. The coefficient of adjustment lies between zero and one; the long-run elasticity equals the short-run elasticity divided by the coefficient of adjustment. A short-run area elasticity of 0.2 with a coefficient of adjustment of 0.4 implies a long-run elasticity of 0.5.
Market structure is classified by the number and size of sellers and buyers, product differentiation, barriers to entry and information. The main structures are perfect competition, monopolistic competition, oligopoly and monopoly on the selling side, and their mirror images on the buying side: monopsony (a single buyer) and oligopsony (a few buyers). Agricultural markets often combine many sellers with few buyers, as when a region's cane growers sell to one sugar mill or cattle producers sell to a few meatpackers, so buyer power is as important as seller power. The structure-conduct-performance (SCP) paradigm, associated with Mason and Bain, holds that structure (concentration, barriers, differentiation) shapes conduct (pricing, advertising, collusion, research) which determines performance (profitability, efficiency, progress). Concentration is measured by concentration ratios, the market share of the largest four or eight firms, and by the Herfindahl-Hirschman Index, the sum of squared market shares; thresholds used by competition authorities differ by country. Contestable market theory (Baumol, Panzar and Willig) showed that even a market with few firms can behave competitively if entry and exit are free and costless.
A monopolist is the only seller of a product with no close substitutes, so its demand curve is the market demand curve, sloping downward, and marginal revenue lies below price. It maximises profit where MR = MC and charges the price buyers will pay for that quantity. Because MR = P(1 - 1/|e|) and MC is positive, a profit-maximising monopolist always operates on the elastic part of demand. Price exceeds marginal cost, so output is lower and price higher than under competition, and a deadweight loss results. The Lerner index of monopoly power is (P - MC)/P, which at the profit maximum equals 1/|e|. The same logic gives the markup pricing rule P = MC / (1 - 1/|e|): with |e| = 4 and MC = 30, P = 40. A monopsonist buying an input, such as a single processor buying milk in an area, buys where the marginal expenditure on the input equals its marginal value, paying a price below the value of the marginal product. Monopolistic competition, analysed by Chamberlin, has many sellers of differentiated products (brands of packaged flour, spices, snacks) and free entry. Each firm has a downward-sloping demand curve and some pricing power in the short run, but entry erodes profits, so in long-run equilibrium the firm's demand curve is tangent to its average cost curve, price equals average cost and exceeds marginal cost, and output is below the level at minimum average cost. This gap is called excess capacity.
Oligopoly is a market with a few interdependent sellers, each of which must consider rivals' reactions; fertiliser, farm machinery, agrochemicals, seed and grain trading are concentrated in many countries. There is no single oligopoly model. Cournot firms choose quantities, taking rivals' output as given; the result lies between monopoly and competition. Bertrand firms choose prices; with identical products and costs, price is driven to marginal cost. In Stackelberg's model a leader sets output first and gains from moving first. Sweezy's kinked demand model assumes rivals match price cuts but ignore price rises, so demand is elastic above the current price and inelastic below it; the resulting gap in marginal revenue explains why prices stay rigid when costs change moderately. Cartels, such as agreements among producers of a commodity, aim to maximise joint profit by acting like a monopolist, but each member has an incentive to cheat by producing more, and most cartels are illegal under competition law. Price leadership occurs when one firm, often the dominant or lowest-cost firm, sets the price and others follow; barometric price leadership is by a firm respected for reading the market.
Game theory studies decisions where each player's payoff depends on the choices of others, which is the essence of oligopoly, bargaining and contract design. A game has players, the strategies available to each, the payoffs for every combination of strategies and the information players have. In the normal (strategic) form, a simultaneous game is shown as a payoff matrix. A dominant strategy is best for a player whatever the others do; if every player has one, the outcome is a dominant-strategy equilibrium. A Nash equilibrium, named after John Nash, is a set of strategies in which no player can gain by changing strategy while the others keep theirs; every dominant-strategy equilibrium is a Nash equilibrium, but not the reverse. In zero-sum games one player's gain is the other's loss. Von Neumann's minimax theorem shows each player can secure a value: a player choosing the maximin (the best of the worst outcomes) and the opponent choosing the minimax reach a saddle point when the two values coincide.
The prisoners' dilemma is the most used game in business. Two players each do better by defecting whatever the other does, but when both defect both are worse off than if both had cooperated. Two processors buying milk in the same district illustrate it: if both pay a moderate price each earns a good margin, but each is tempted to raise its price slightly to attract suppliers from the other; if both raise prices, margins fall for both. The same structure explains advertising wars and why cartel members cheat on quotas. In a one-shot game the dilemma leads to defection. In a repeated game with no known last period, cooperation can be sustained because a player who defects today can be punished tomorrow. Trigger strategies, such as grim trigger (cooperate until the other defects, then defect forever), and tit-for-tat (cooperate first, then copy the rival's last move) are examples; Axelrod's computer tournaments found tit-for-tat performed well. Cooperation is easier when players value the future highly (a low discount rate), when defections are quickly detected, when there are few players and when demand is stable.
In a sequential game players move in turn, and the game is shown as a tree (extensive form). It is solved by backward induction: start at the last decision, find the best move there, and work back to the first move. The resulting equilibrium, a subgame-perfect equilibrium, rules out threats that would not be carried out when the time came. For example, an incumbent grain trader may threaten a price war if a new trader enters, but if a price war would cost the incumbent more than sharing the market, the threat is not credible and the entrant should enter. A strategic move changes the game to the mover's advantage by limiting its own options. Building excess capacity, signing long-term supply contracts or publicly committing to match any competitor's price can make a threat credible, because they change the incumbent's own payoffs.
Joint products arise when one production process yields two or more outputs at the same time. Agribusiness is full of them: milk separates into cream and skimmed milk, wheat into grain and straw, seed cotton into lint and cottonseed, cane into sugar, molasses and bagasse, slaughtered cattle into meat, hides and offal, and oilseed crushing into oil and meal. When products come in fixed proportions, the cost of the joint process cannot be meaningfully split between them; any allocation of joint cost by weight or value is arbitrary and should not be used for pricing or output decisions. Instead, treat each batch of the process as the unit of output. The marginal revenue of a batch is the vertical sum of the marginal revenues of the joint products, and profit is maximised where this combined marginal revenue equals the marginal cost of the batch. If at that output the marginal revenue of one product is negative, the firm should not sell all of it: it sells that product only up to where its marginal revenue is zero and disposes of or stores the rest, and the output decision then uses the marginal revenue of the other product alone.
When the proportions can be varied, for example by choosing a wheat variety with more grain and less straw or a dairy product mix with more butter and less cheese from the same milk, the problem becomes one of product-product choice. The product transformation curve shows combinations of two products obtainable from given resources; it is usually concave to the origin because the marginal rate of product transformation rises as production shifts toward one product. An isorevenue line joins combinations with the same revenue, with slope equal to minus the ratio of product prices. The revenue-maximising combination for a given cost is where the isorevenue line is tangent to the transformation curve: the marginal rate of product transformation equals the price ratio. Products can be competitive (more of one means less of the other), complementary (more of one increases the other, as when a legume in rotation raises yields of the following cereal) or supplementary (one can be increased without reducing the other, such as backyard poultry using family labour that is otherwise idle).
Price discrimination is charging different prices for the same product when the price differences do not reflect differences in cost. It requires market power, the ability to identify or sort buyers with different willingness to pay, and prevention of resale between groups. First-degree or perfect discrimination charges each buyer their maximum willingness to pay, so the seller captures all consumer surplus; output rises to the competitive level, so there is no deadweight loss, but all surplus goes to the seller. It is approached in individual negotiation, such as a trader bargaining separately with each farmer. Second-degree discrimination charges different prices for different quantities or versions and lets buyers sort themselves: quantity discounts, block tariffs for electricity or water, and two-part tariffs with a fixed fee plus a per-unit charge. Third-degree discrimination divides buyers into identifiable groups with different elasticities of demand, such as domestic and export markets, farmers and industrial users of power, or students and adults. The seller sets output in each market so that MR1 = MR2 = MC, which means the higher price is charged in the market with less elastic demand. Because MR = P(1 - 1/|e|), if |e| is 2 in one market and 4 in the other and marginal cost is 12, prices are 24 and 16.
Much agricultural production spans more than one period: a dairy cow is reared for two or more years before she calves, fruit trees bear for decades, timber takes a generation and soil fertility built this year pays off later. In multi-period production inputs and outputs occur at different times, so they must be compared in present-value terms using an interest rate. Three classic decisions follow. The first is when to harvest a growing asset. For a single rotation, a stand of trees or a fattening animal should be harvested when the rate of growth of its value falls to the rate of interest, because after that point money in the bank grows faster. When land will be replanted, the Faustmann rule accounts for the delay to all future rotations, and so implies an earlier harvest than the single-rotation rule. The second is when to replace an asset, such as a laying flock, a cow or a tractor. The asset should be kept as long as its marginal net revenue for the next period exceeds the average annual net revenue (annuity equivalent) obtainable from the best replacement. The third is how much to invest now for returns later, the capital budgeting problem.
Capital budgeting is the evaluation of long-lived investments such as a cold store, a processing line, a dairy herd expansion or a drip irrigation system. The steps are to estimate incremental cash flows after tax, choose a discount rate reflecting the opportunity cost of capital and project risk, apply decision criteria and test sensitivity. The payback period is the time taken to recover the initial outlay from cash inflows; it is simple and favours liquidity but ignores the time value of money and flows after payback. Discounted methods are preferred. Net present value (NPV) is the present value of inflows minus the present value of outflows; accept if NPV is positive. An outlay of 100,000 returning 30,000 a year for five years has a payback of 3.33 years, and at 10 percent the present value of inflows is 30,000 x 3.7908 = 113,724, so NPV = 13,724. The internal rate of return (IRR) is the discount rate that makes NPV zero; accept if it exceeds the cost of capital. The benefit-cost ratio (profitability index) is the present value of benefits divided by the present value of costs; accept if it exceeds one. For independent projects NPV and IRR give the same accept or reject decision. For mutually exclusive projects of different size or timing they can rank differently; NPV is preferred because it measures the absolute addition to wealth and assumes reinvestment at the cost of capital rather than at the IRR. For public agricultural projects, financial analysis at market prices is complemented by economic analysis at shadow prices that remove taxes, subsidies and distortions, following the tradition of Gittinger's work for the World Bank.
Frank Knight distinguished risk, where the probabilities of outcomes are known or can be estimated, from uncertainty, where they cannot. Under risk the expected value is the probability-weighted average of outcomes, the variance or standard deviation measures spread, and the coefficient of variation (standard deviation divided by expected value) compares risk per unit of return across projects of different size. Under uncertainty several decision rules are taught. The maximax rule picks the alternative with the best possible outcome (optimistic). The maximin or Wald rule picks the alternative whose worst outcome is best (pessimistic). The minimax regret or Savage rule computes, for each state, the regret of each alternative (the best payoff in that state minus the alternative's payoff) and picks the alternative whose maximum regret is smallest. The Laplace rule treats all states as equally likely. The Hurwicz rule weights the best and worst outcome of each alternative by a coefficient of optimism. Expected utility theory explains why people pay to avoid risk: a risk-averse decision maker has a concave utility function, values a gamble at a certainty equivalent below its expected value, and would pay a risk premium equal to the difference. The expected value of perfect information, the expected payoff with perfect foresight minus the best expected payoff without it, caps what a manager should pay for a forecast. Risk is managed by diversification of enterprises and markets, crop and livestock insurance, forward and futures contracts, contract farming, holding liquid reserves, flexible plant and keeping leverage moderate.
Firms sell into an economy whose total spending, prices and interest rates they cannot control but must forecast. The circular flow model shows households supplying factor services to firms and receiving incomes, and firms supplying goods and services to households and receiving expenditure. In a two-sector economy income equals expenditure. Saving is a leakage and investment an injection, handled by financial markets; adding government brings taxes as a leakage and government spending as an injection; adding the rest of the world brings imports as a leakage and exports as an injection. The economy is in equilibrium when planned leakages equal planned injections. Gross domestic product (GDP) is the market value of all final goods and services produced within a country in a period. It can be measured by the product or value-added method (sum of value added at each stage, which avoids double counting of intermediate goods), the income method (wages, rent, interest, profit and mixed income of the self-employed) and the expenditure method, GDP = C + I + G + (X - M). Gross national product (GNP) adds net factor income from abroad. Net figures subtract depreciation (consumption of fixed capital). National income in the traditional sense is net national product at factor cost, that is, at market prices minus indirect taxes plus subsidies. Nominal GDP is at current prices; real GDP is at constant prices, and the GDP deflator equals nominal GDP divided by real GDP, times 100.
The classical quantity theory, in Fisher's equation of exchange MV = PT, relates the money stock M and its velocity V to the price level P and the volume of transactions T; if V and T are stable, prices rise in proportion to money. The Cambridge cash-balance version, M = kPY, treats k as the fraction of nominal income people wish to hold as money, turning the identity into a demand for money. Keynes's liquidity preference theory gives three motives for holding money: the transactions motive and the precautionary motive, both rising with income, and the speculative motive, which varies inversely with the interest rate because people hold money when they expect bond prices to fall. At very low interest rates the demand for money may become perfectly elastic, the liquidity trap, in which extra money does not lower interest rates further. The Baumol-Tobin inventory model shows transactions demand rising with income and the cost of converting bonds to cash, and falling with the interest rate, with an elasticity with respect to income of one half (the square-root rule). Friedman's modern quantity theory treats money as one asset among several, with demand depending on permanent income and the returns on alternative assets, and argues that money demand is stable.
Inflation is a sustained rise in the general price level, measured by consumer price indices, wholesale or producer price indices and the GDP deflator. Food has a large weight in consumer price indices of low- and middle-income countries, so harvests, monsoons and world commodity prices move headline inflation strongly. Demand-pull inflation arises when aggregate demand outruns the economy's capacity to supply, from rapid money growth, large fiscal deficits or booming exports. Cost-push inflation arises from rising costs, such as oil, wages or imported inputs, shifting aggregate supply leftward; together with stagnant output it produces stagflation. Structural inflation, emphasised in developing economies, stems from supply bottlenecks in agriculture, infrastructure and foreign exchange. The Fisher relation approximates the real interest rate as the nominal rate minus expected inflation: a nominal rate of 9 percent with inflation of 6 percent gives a real rate of about 3 percent (exactly 1.09/1.06 - 1 = 2.83 percent). The Phillips curve described an inverse relation between unemployment and inflation; Friedman and Phelps argued that once expectations adjust the long-run curve is vertical at the natural rate of unemployment. Central banks fight inflation by raising policy interest rates, open market sales of securities and higher reserve requirements; many now use inflation targeting, for example India's flexible target of 4 percent consumer price inflation within a band of 2 to 6 percent.
Business cycles are recurrent fluctuations in aggregate output, employment and prices around the long-run trend. The four phases are expansion (recovery and boom), peak, contraction (recession, or depression if deep and prolonged) and trough. Cycles are recurrent but not periodic: their length and depth vary. Explanations differ. Keynesian theory stresses swings in aggregate demand, especially investment, amplified by the multiplier, which equals 1/(1 - MPC) in the simplest model, so a marginal propensity to consume of 0.8 gives a multiplier of 5. Samuelson's multiplier-accelerator model shows how their interaction can generate cycles. Monetarist theory emphasises changes in money supply; real business cycle theory attributes fluctuations to technology and other supply shocks; Schumpeter linked long waves to clusters of innovation; and in farming, weather and animal-disease shocks and the cobweb-type cycles of supply add their own rhythm. Demand for durable goods and capital equipment such as tractors, cold stores and processing machinery is more cyclical than demand for staple foods, which is income-inelastic.
Learn to move between a diagram, an equation and a number. For every topic be able to draw the curve, write the condition (MR = MC, MRTS = w/r, MR1 = MR2 = MC, NPV > 0) and solve a small numerical in under two minutes, because entrance papers reward speed with negative marking. Build a one-page sheet of formulas: elasticities and the MR formula, break-even and P/V ratio, Lerner index and markup rule, Nerlove long-run elasticity, exponential smoothing, multiplier and Fisher relation. Practise explaining each with an agricultural example (a bumper onion crop, a milk cooperative, a sugar mill with molasses), since interview panels for agribusiness programmes ask for applications, not definitions.
A manager five years into a seed, feed, dairy or food company uses this subject weekly. Price changes are tested against estimated elasticities and competitor responses before launch. Monthly sales forecasts combine exponential smoothing with seasonal indices and field intelligence on rainfall and sowing. Product-line reviews separate contribution from allocated overhead, so that a line covering its variable cost is not dropped because of an arbitrary share of joint cost. New depots, cold rooms or machines are appraised by NPV with sensitivity to throughput and price. Procurement managers buying from farmers think about the buyer-side game: paying reliably and building repeat dealing reduces side-selling more cheaply than a price war.
At twenty-five years' experience the questions are strategic. Where is the industry on its long-run cost curve and how many plants of minimum efficient scale can the region support? Will a capacity expansion deter entry or start a price war, and is the threat credible? How do inflation, the interest-rate cycle and farm-income swings change demand for our products over the next five years? Senior managers also negotiate with regulators over pricing, procurement and competition issues, and must understand how authorities read concentration, price discrimination and dumping. They design farmer contracts and incentive schemes using what behavioural economics and game theory say about trust, timing and risk-sharing, and they decide how much risk to carry versus hedge or insure.
A researcher with long field experience works at the frontier: estimating demand systems and supply response with modern econometrics, measuring market power in farm-to-retail chains, testing behavioural models of farmer decisions through field experiments, modelling investment and replacement under uncertainty with real options and dynamic programming, and evaluating how policy (price supports, trade restrictions, competition law) changes firm conduct. Such a person advises governments and boards on whether observed margins reflect costs or market power, designs forecasting systems for commodity boards, and supervises doctoral work. The recurring discipline is to check that each assumption of the textbook model (price-taking, full rationality, known probabilities) holds in the market studied, and to say plainly when it does not.
This manual gives the standard theory and methods with worked examples but no data tables, because tables are not yet supported. It does not give country-specific elasticity estimates, current inflation or GDP figures, competition-law thresholds, or tax rules for depreciation and transfer pricing, since these vary by country and year; managers must take them from their own national sources. Game theory, real options, demand-system estimation (such as AIDS) and dynamic programming are introduced only at overview level. The macroeconomic block covers the four syllabus units (national income, money demand, inflation, business cycles) but not open-economy macroeconomics, exchange-rate determination or detailed monetary-policy operations, which belong to other subjects.
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.