Economics
Saved & recent

Economics glossary

How people make choices, markets work, and economies grow. Each concept has a short explanation, an illustration, and links to the ideas behind it.

Economic foundations

Scarcity, tradeoffs, models and the logic of choice.

Scarcity

Resources and time constrain which wants can be satisfied together, so choosing one use excludes others.

feasible plan wish list study 2 h work 2 h leisure 2 h study 3 h work 3 h leisure 2 h six-hour limit 2 h over
Hypothetical day with six allocable hours. The wish list exceeds the resource constraint by two hours; scarcity is the constraint, not a temporary market shortage.

Opportunity cost

The cost of a choice includes the value of its best forgone alternative, including uses of time and owned resources.

Read first
Scarcity
one evening concert chosen paid shift best alternative 60 currency units forgone film at home
Hypothetical mutually exclusive alternatives. The paid shift is the best feasible forgone use of the evening; the values of all rejected options are not added.

Incentives

Changing rewards, penalties or constraints can change choices; responses depend on what people value and what alternatives they face.

Hypothetical 30-minute bus trip and 10-minute car trip, with time valued at 12 currency units/hour. Bus cost is 2 + 6 = 8; car cost rises from 4 + 2 = 6 to 11. Preferences and other trip features are held fixed.

Sunk costs

An unrecoverable past expense does not change the future benefits and costs of continuing a project.

Hypothetical currency amounts. The nonrefundable setup cost is sunk. Recoverable equipment resale value of 20 is available only if the project stops; continuing therefore gains 80 − 50 − 20 = 10 relative to stopping.

Positive and normative economics

Positive claims describe or predict outcomes; normative judgments evaluate what should happen using value judgments.

Read first
Economic models
Conceptual competitive model. A positive claim can be tested and can be false; normative assessments also require values. The model alone does not decide how fairness should be judged.

Production possibilities frontier

The frontier gives maximum feasible combinations of outputs with available resources and technology; its slope measures an output tradeoff.

Hypothetical nonnegative outputs with fixed resources and technology. The boundary shows productive efficiency; moving along it trades goods. Technology changes could move the boundary, but are not depicted.

Markets and prices

Supply, demand, responsiveness, gains from exchange and price interventions.

Supply

A supply schedule relates a good's price to the amount sellers offer over a specified period, holding other determinants fixed.

Read first
Ceteris paribus
Hypothetical supply Qs = −20 + 2P for P ≥ 10; quantities below zero are excluded. Technology and input prices are fixed. At price 30, sellers offer 40 units per period.

Market equilibrium

In the competitive model, equilibrium price makes quantity demanded equal quantity supplied; it is not a claim that the outcome is fair.

Read first
Demand, Supply
Hypothetical competitive market. D₀: Qd = 100 − 2P; D₁: Qd = 120 − 2P; unchanged supply Qs = −20 + 2P. Only demand shifts. Market clearing says nothing by itself about fairness.

Shortages and surpluses

At a given price, excess demand is quantity demanded above quantity supplied; excess supply is the reverse.

Hypothetical Qd = 100 − 2P and Qs = −20 + 2P. At P = 40, excess supply is 60 − 20 = 40; at P = 20, excess demand is 60 − 20 = 40 units per period. Each gap is a quantity read along one price, not a price difference.

Movement along a curve versus shift

An own-price change moves along a demand or supply curve; a changed underlying determinant changes the schedule itself.

Read first
Demand, Supply
Hypothetical demand P = 10 − Q. Left: own price falls, with other determinants fixed. Right: a preference change gives P = 13 − Q, increasing quantity demanded from 5 to 8 at P = 5. No supply change is assumed.

Price elasticity of demand

Demand elasticity measures percentage quantity response to a percentage price change; its magnitude differs from a curve's slope.

Read first
Demand
Hypothetical demand P = 20 − Q. Upper: (4/6) ÷ (−4/14) = −7/3. Lower, from (12, 8) to (16, 4): (4/14) ÷ (−4/6) = −3/7. Equal slopes do not imply equal percentage responses.

Price elasticity of supply

Supply elasticity measures percentage quantity response to percentage price change and depends on production flexibility and the time horizon.

Read first
Supply
Hypothetical local supply responses: QSR = 36 + 0.4P and QLR = 20 + 2P. At the common starting P = 10, Q = 40, point elasticity (dQ/dP)(P/Q) is 0.1 versus 0.5. Both axes use the same scale.

Income and cross-price elasticity

Income elasticity measures demand response to income; cross-price elasticity measures response to another good's price.

Read first
Demand
Hypothetical finite percentage changes using the stated initial values as the percentage bases. Each elasticity is ±5% / 10% = ±0.5. Classification describes demand responses over this change, with other factors fixed.

Elasticity and total revenue

Revenue equals price times quantity; a small price rise increases revenue on an inelastic demand segment and decreases it on an elastic segment.

Hypothetical P = 20 − Q gives TR = 20Q − Q², maximized at Q = 10 and P = 10. Above this price demand is elastic; below it demand is inelastic. Revenue excludes all production costs.

Consumer surplus

Consumer surplus measures willingness to pay above expenditure for units purchased, within the chosen demand model.

Hypothetical consumer surplus is ½ × 12 × (20 − 8) = 72 currency units per period. The shaded area includes only the 12 purchased units and treats the demand curve as marginal willingness to pay.

Producer surplus

Producer surplus is revenue above the minimum variable opportunity cost of supplying units; it need not equal profit after fixed costs.

Hypothetical producer surplus is ½ × 6 × (8 − 2) = 18 currency units per period. Fixed costs are excluded from this area; economic profit would subtract them as well.

Deadweight loss

Deadweight loss is forgone total surplus relative to a specified efficient benchmark, excluding transfers between participants.

Hypothetical competitive benchmark with no external effects. Restricting trade to Q = 6 leaves a 3-unit base and a marginal-value gap of 6 at that quantity: DWL = ½ × 3 × 6 = 9 currency units per period. Transfers are not counted as losses.

Price ceilings

A maximum legal price below competitive equilibrium is binding and produces excess demand in the basic model; rationing determines who buys.

Hypothetical Qd = 100 − 2P and Qs = −20 + 2P. A cap of 20 is below equilibrium 30: sellers offer 20 and buyers request 60. At most 20 units can trade without another source of supply; rationing determines buyers. A cap at or above 30 is nonbinding.

Price floors

A minimum legal price above competitive equilibrium is binding and produces excess supply in the basic model.

Hypothetical Qd = 100 − 2P and Qs = −20 + 2P. A floor of 40 is above equilibrium 30: sellers offer 60 and private buyers request 20. No government purchases are assumed, so offered output need not all sell. A floor at or below 30 is nonbinding.

Tax incidence

The economic burden of a per-unit tax depends on demand and supply responses, rather than merely on who legally remits it.

Hypothetical P demand = 20 − Q and P supply = 2 + Q. For t = 6, Q = (18 − 6)/2 = 6; buyer price is 14, seller price 8, and tax receipts 6 × 6 = 36 per period. Equal slopes yield equal burdens here; legal remittance does not determine these burdens.

Pareto efficiency

An allocation is Pareto efficient if no feasible change benefits someone without harming anyone else. This criterion does not rank fairness or resolve tradeoffs when a change creates winners and losers.

Hypothetical utility indices, with no interpersonal comparison implied. (4,4) Pareto-improves on (3,3). Every point on uA + uB = 10 is efficient, including (1,9) and (5,5); efficiency alone does not rank their fairness.

Consumer choice

Budgets, preferences and responses to changing prices and income.

Budget constraint

A budget set contains affordable consumption bundles at given income and prices; its boundary describes spending the available budget.

Read first
Scarcity
Hypothetical nonnegative bundles: 2x + 4y ≤ 40. The original slope is −1/2. Raising income to 60 gives intercepts (30,15); separately lowering px to 4/3 at income 40 gives (30,10). The two changes are alternatives, not combined.

Utility and preferences

Utility represents a person's ranking of alternatives; numerical utility differences need not measure happiness or compare wellbeing across people.

Hypothetical U(x,y) = xy for positive bundles. The strictly increasing transformation ln U preserves preferences while changing numerical gaps. Utility numbers are ranking devices here, not comparable measures of happiness across people.

Diminishing marginal utility

In a chosen utility representation, successive units may add less utility; declining additions can coexist with rising total utility.

Hypothetical U(q) = √q, with U(0) = 0. Discrete gains √q − √(q−1) shrink as q rises even though total utility rises. Curvature and marginal-utility numbers depend on the chosen utility representation.

Indifference curves and substitution

An indifference curve connects equally ranked bundles; its local tradeoff is the marginal rate of substitution.

Hypothetical U = xy for positive quantities. Higher curves rank above lower ones, and curves never cross. Along a curve, dy/dx = −y/x; the positive marginal rate of substitution is y/x, equal to 1 at (3,3).

Consumer optimum

A consumer optimum is the most preferred affordable bundle; smooth interior solutions equate willingness to trade with the price ratio.

Hypothetical positive prices and income 40. For U = xy, (10,5) spends 40 and MRS = 5/10 = px/py. For U = x + y, x gives twice the utility per currency unit, so (20,0) is optimal; a tangency condition is unnecessary at this corner.

Normal and inferior goods

At fixed prices, demand for a normal good rises with income and demand for an inferior good falls; these labels concern responses, not quality.

Hypothetical Engel curves for different goods: Qnormal = 2 + 0.2Y and Qinferior = 18 − 0.1Y, only over income Y = 20–80. The labels concern income responses over this range, not quality or responses at every income.

Income and substitution effects

A price change alters relative prices and purchasing power; separating these effects explains how the final consumption change arises.

Hypothetical U = xy, income 40 and py = 4; px falls from 2 to 1. Hicks compensation holds U₀ = 50: B = (√200, √12.5), with compensated income 20√2 ≈ 28.28. The substitution gain in x is 4.14 and the income gain is 5.86, adding to the total gain of 10. The final indifference curve U₁ = 100 passes through C.

Production and costs

Technology, output, costs and profitability.

Production function

A production function describes output attainable from specified inputs and technology; it separates technical possibilities from prices and profits.

Read first
Economic models
Hypothetical technology Q = √(KL). Physical capital K is machinery, not money. Holding K = 4 gives Q = 2√L; inputs and output use normalized physical units.

Marginal product and diminishing returns

Marginal product is extra output from another input unit; diminishing marginal returns can occur while total output still increases.

Hypothetical model Q = 2√L, K = 4 and L > 0. MPL = dQ/dL = 1/√L measures extra output per marginal labor unit. A smaller positive slope still raises total output.

Economies of scale

Economies of scale mean lower average cost at larger output when all inputs can adjust; they differ from changing one input with others fixed.

Conceptual cost envelope with fixed input prices. LRAC = 4 + 0.4 max(3−q,0)² + 0.4 max(q−5,0)². The shown plant has SRAC = LRAC + 0.3(q−4)², so it cannot lie below the envelope. Other plant choices form the rest of LRAC.

Economic versus accounting profit

Economic profit deducts explicit costs and implicit opportunity costs; positive accounting profit can coexist with zero or negative economic profit.

Hypothetical € thousands in one year: 100 − 70 = 30 accounting profit; 30 − 25 = 5 economic profit. The implicit 25 is the opportunity cost of owner resources, counted once. Zero economic profit would include their normal return.

Marginal revenue

Marginal revenue is the change in total revenue from additional sales; for a single-price seller, lowering price on existing units can reduce it below price.

Related
Monopoly
Hypothetical single-price seller: P = 20−q, TR = 20q−q² and MR = 20−2q, for 0 ≤ q ≤ 20. Selling more lowers the price on every unit. TR peaks at q = 10, where MR = 0.

Profit maximization

Profit maximization chooses feasible output with the greatest revenue minus cost; MR=MC is an interior condition, not by itself proof of a maximum.

Hypothetical feasible domain 0 ≤ q ≤ 10: TR = 20q−q², TC = 12+2q+q². Profit = −12+18q−2q² is strictly concave. At q = 4.5, MR = MC = 11 and profit = 28.5; endpoint profits are −12 and −32.

Shutdown versus exit

A firm can cease short-run production yet retain unavoidable fixed costs; long-run exit permits avoiding costs tied to staying in the industry.

Hypothetical price taker: TC = 4 + 5q − 2q² + q³/3; minimum AVC = 2 at q = 3. At P = 5, optimal q = 4 and profit ≈ 6.67. At P = 2.5, q ≈ 3.22 and loss ≈ 2.44, less than fixed cost 4. At P = 1.5, choose q = 0. Long-run exit can avoid costs that remain fixed today.

Returns to scale

Returns to scale compare the output response when all inputs increase in the same proportion with technology fixed. They describe production quantities; economies of scale describe average costs and additionally depend on input prices and cost minimization.

Hypothetical Q = KᵃLᵃ with fixed technology. Doubling both inputs gives output multiplier 2²ᵃ: 1.516, 2, or 2.639. All three have diminishing marginal product in either input alone because 0 < a < 1. Output multipliers are dimensionless.

Market structure

Competition, entry and market power.

Perfect competition

In the benchmark, many small buyers and sellers trade identical products, take price as given, and face free entry with relevant information available.

Related
Monopoly
Hypothetical market P demand = 20 − Q and P supply = 4 + Q. Market equilibrium is P = 12, Q = 8. Two identical firms with MC = 4 + 2q each produce q = 4, so their outputs sum to Q = 8. Assume operation is worthwhile; firm demand is horizontal while market demand slopes down.

Entry and exit

Profit attracts entry and losses encourage exit where barriers are absent; this can move long-run economic profit toward zero.

Hypothetical constant-cost industry with free entry and minimum average total cost 12. D changes from P = 20 − Q to 24 − Q; entry shifts S from P = 4 + Q to P = 4 + 2Q/3. Each firm has C(q) = 16 + 4q + q²: entry increases the number of identical firms from two to three. Equilibria are A(8,12), B(10,14), C(12,12). The final zero economic profit still includes normal compensation for owners’ resources.

Barriers to entry

Entry barriers impede potential competitors and can sustain market power through cost advantages, control of scarce inputs, or legal restrictions.

Read first
Entry and exit
Conceptual comparison, not a claim about any jurisdiction’s laws. Scale economies can disadvantage small entrants; control of scarce inputs or protected rights can restrict access. A cost all firms can meet on equivalent terms is not by itself proof of a barrier.

Monopoly

A sole seller with protected market access faces market demand and chooses output while accounting for how price changes with sales.

Hypothetical single-price monopoly with P = 20 − q and MC = 4. MR = 20 − 2q gives q = 8 and P = 12. The markup is 8 per unit; total profit requires subtracting any fixed costs. The efficient competitive benchmark is q = 16 when there are no external effects.

Natural monopoly

A market is a natural monopoly when one firm can supply relevant demand at lower total cost than multiple firms, often because of large fixed costs.

Hypothetical technology C(q) = 100 + 2q for each active firm. For every positive split Q = q₁ + q₂, two firms cost 200 + 2Q versus one firm’s 100 + 2Q. AC falls across the illustrated relevant demand range; at Q = 50, costs are 300 versus 200.

Monopolistic competition

Many sellers offer differentiated products, giving each some pricing power while entry limits persistent economic profit in the standard model.

Related
Oligopoly
Hypothetical D: P = 18 − 1.5q; MR = 18 − 3q; C(q) = 32 + 2q + 0.5q². Thus MC = 2 + q and ATC = 32/q + 2 + 0.5q. At q = 4, P = ATC = 12 and MR = MC = 6; both D and ATC have slope −1.5. This is a consistent free-entry long-run example, with minimum ATC at q = 8.

Oligopoly

When a few firms account for much supply, each firm's best action can depend on rivals' choices and expected responses.

Hypothetical simultaneous quantity competition, P = 20 − qA − qB, MC = 4, and no fixed costs. The illustrated (4,4) choice is not a Nash equilibrium: A can gain by changing output alone. Price competition or different assumptions can produce other oligopoly outcomes.

Price discrimination

A seller may charge different prices across buyers or purchase quantities for reasons beyond cost differences. Market power, information and limits on resale shape which forms are feasible; welfare effects depend on the scheme.

Hypothetical no-resale benchmark with P = 20 − Q and MC = 4; fixed costs excluded. A single price gives Q = 8, CS = 32, PS = 64. Perfect first-degree discrimination captures each buyer’s willingness to pay, serves Q = 16, gives CS = 0 and PS = 128. Its extra trades create 32 additional total surplus; other discrimination schemes need not have the same welfare effect.

Strategic interaction

Games, equilibria and repeated incentives.

Game theory

Game theory models strategic interaction by specifying players, feasible strategies, information, and payoffs from combinations of choices.

Read first
Incentives
Hypothetical one-period profits in thousands of currency units, with simultaneous choices and common knowledge of the payoff table. Rows are A’s prices, columns B’s prices, and every payoff pair is ordered (A, B).

Nash equilibrium

A Nash equilibrium is a strategy profile where no player can gain by changing their own strategy alone, holding the others' strategies fixed.

Read first
Game theory
Hypothetical payoff indices, higher preferred. Given B chooses X, A prefers X; given B chooses Y, A prefers Y, and symmetrically for B. Both (X,X) and (Y,Y) are Nash equilibria, although the latter has lower total payoff. Neither player has a dominant strategy.

Dominant strategies

A dominant strategy is best regardless of other players' choices; many games lack one even when they have a Nash equilibrium.

Read first
Game theory
Hypothetical ordered payoff indices (A, B), with simultaneous choices. Game 1: A’s top row is strictly better in both columns, 4 > 2 and 3 > 1. Game 2: A’s best response changes by column, so A has no dominant strategy. Only A’s comparisons are underlined.

Prisoner's dilemma

In a prisoner's dilemma, individually dominant actions lead to an outcome both players prefer less than mutual cooperation.

Hypothetical simultaneous game with payoff indices, higher preferred. Defection strictly dominates because 5 > 3 and 1 > 0 for each player. (D,D) is the unique Nash equilibrium, while (C,C) Pareto-dominates it. Arrows show unilateral temptations away from mutual cooperation.

Collusion and repeated games

Repeated interaction can make future retaliation affect today's incentives, potentially sustaining cooperation that is unstable in a one-shot game.

Related
Oligopoly
Hypothetical infinitely repeated prisoner’s dilemma, perfect monitoring, 0 ≤ δ < 1, and grim-trigger punishment. Cooperation requires 3/(1−δ) ≥ 5 + δ/(1−δ), equivalent to δ ≥ 1/2. Equality means indifference. The illustration is a model of incentives, not a statement about legality.

Measuring the economy

Output, prices, employment and accounting distinctions.

Stocks and flows

A stock is measured at an instant; a flow is measured over an interval. Flows change stocks, alongside possible valuation changes.

Hypothetical balance with no revaluations: €100m + (€30m/year − €20m/year) × 1 year = €110m. Stocks are measured at dates; rates become stock changes only after multiplying by elapsed time.

Circular flow of income

Production generates income used for expenditure; real resources and payments move in opposite directions between firms and households.

Conceptual closed economy with no government or foreign trade. Solid arrows show resources or goods; dashed arrows show payments or financial flows. Saving can finance investment, but each household need not spend its own current income. Other factor incomes are omitted.

Value added

Value added is output value less intermediate consumption. Adding value added across producers avoids counting purchased inputs repeatedly.

Hypothetical € per batch, with no product taxes or other intermediate inputs. Value added equals sales less purchased intermediate inputs: 30 + (50−30) + (80−50) = 80. Wages are paid from value added; they are not subtracted as intermediate consumption.

Gross domestic product

GDP measures production within an economy over a period. Production, expenditure and income approaches describe the same total in principle.

Hypothetical economy, €m in one year. Production uses value added at basic prices plus net product taxes; income includes compensation, gross operating surplus and net production/product taxes. Assume no other production taxes, mixed income, net exports or statistical discrepancy.

GDP expenditure components

Expenditure GDP is consumption plus investment plus government final purchases plus exports minus imports; transfers are not government purchases.

Hypothetical €m/year: GDP = C 60 + I 20 + G 25 + X 15 − M 20 = 100. I includes private fixed investment and inventories; G includes government consumption and investment, so no spending is counted twice. Transfers and existing financial-asset purchases are excluded. Imports remove foreign content already included in C, I, G or X.

Nominal and real values

Nominal values use current prices; real measures remove price changes to compare quantities or purchasing power.

Hypothetical two-good economy. Base quantities (2,1) and prices (€2,€6) give 10. Price-only case uses (€3,€9) and unchanged quantities; quantity-only case uses (4,2) at base prices. Real output uses fixed base prices, a simplification of chain-volume measures.

Price indices

A price index summarizes price changes using specified coverage and weights. Different baskets can yield different inflation measures.

Hypothetical fixed basket: 3 bread units and 1 milk unit. Prices change from (€2,€4) to (€3,€5). Base spending weights are 0.6 and 0.4; index = 100 × [0.6×1.5 + 0.4×1.25] = 140, also 100×14/10. Basket quantities stay fixed; normalization at 100 does not determine weights.

GDP deflator and consumer prices

The GDP deflator covers domestically produced final output, while consumer price indices cover household consumption; coverage and weights differ.

Conceptual coverage for a simplified CPI and GDP deflator. CPI covers household consumption, including imports; the deflator covers domestic final production, including exports and investment. Hypothetical compatible GDP values 120 and 100 imply deflator 100×120/100 = 120; CPI need not equal 120.

Inflation and the price level

Inflation is the percentage increase in a broad price index over a stated period. A lower positive inflation rate still means rising prices.

Read first
Price indices
Hypothetical annual index observations: 100 → 110 → 115.5. Inflation is 100×(current index / previous index − 1): 10%, then 5%. The decline is 5 percentage points, or 50% of the earlier rate; the price level still increases.

Disinflation and deflation

Disinflation means inflation slows; deflation means the general price level falls. Neither is defined by one product becoming cheaper.

Hypothetical broad price indices, all starting at 100. Disinflation: +10%, then +5%. Stable prices: 0%, then 0%. Deflation: −5%, then −5%. Each annual rate compounds from its own preceding level.

GDP and wellbeing

GDP records production, not a complete welfare score. Distribution, unpaid work, leisure and environmental costs require additional measures.

Hypothetical annual economy-wide comparisons. Real GDP/person is €30,000 in both. Leisure is hours/person/day; distribution is the bottom 50% share of total income; emissions are territorial tonnes CO₂/person/year. Separate scales and dimensions require explicit value judgments to combine.

Unemployment rate

Unemployment counts people without employment who meet availability and job-search criteria. The rate divides this group by the labor force.

Hypothetical population aged 15+, reference week. Unemployed people have no employment and meet availability and recent job-search criteria. Labor force = 60 employed + 10 unemployed = 70; unemployment rate = 10/70 ≈ 14.3%, not 10/100.

Labor-force participation and employment ratios

Participation divides the labor force by the working-age population; the employment ratio counts employed people over that population.

Hypothetical fixed population of 100 adults aged 15+, comparing reference weeks. Five unemployed people stop active search and move outside the labor force; no one gains a job. Participation falls 70%→65%; employment/population stays 60%; unemployment becomes 5/65 ≈ 7.7%.

Labor underutilization

Unemployment misses some unmet demand for work, including insufficient working hours and some available people outside the labor force.

Hypothetical adults aged 15+, one reference week, using ILO labor-underutilization concepts. Underemployed 8 are already among employed 60. Potential labor force 5 is outside the labor force. LU1 = 10/70 = 14.3%; LU2 = (8+10)/70 = 25.7%; LU3 = (10+5)/75 = 20%; LU4 = (8+10+5)/75 = 30.7%.

Frictional, structural and cyclical unemployment

Job search, persistent skill or location mismatch, and weak aggregate activity offer different explanations for unemployment; their measured contributions are uncertain.

Conceptual mechanisms, not an empirical partition: search delays, persistent skill or location mismatch, and weak aggregate demand can overlap. The examples do not identify measured shares or imply a universal natural unemployment rate.

Saving and investment

Saving is income not consumed; national-account investment adds produced capital or inventories. Purchasing an existing financial asset is not new real investment.

Conceptual asset exchange above; hypothetical closed economy without government below. In national accounts, S = Y−C = 100−80 = 20 = I, including inventory investment. This is an ex post identity, not a claim that planned saving and planned investment always coincide. Existing-share purchases are financial transactions.

Government debt and deficits

A deficit is a period’s expenditure exceeding revenue; debt is outstanding borrowing at a date. Valuation and other adjustments can also change debt.

Hypothetical general-government gross debt, €bn. Opening debt 100 + annual deficit 10 − other stock-flow adjustments 3 = closing debt 107. The adjustments can reflect financial-asset transactions, valuation or classification changes. Debt is a dated stock; the deficit covers the year.

Money and finance

Money, balance sheets, time value, financial claims and fragility.

Assets, liabilities and equity

A balance sheet records assets, obligations and residual equity at one date; each financial claim is another entity’s liability or equity claim.

Hypothetical nonbank loan at origination. The loan asset and matching debt cancel when the two parties are consolidated; lending itself creates no combined equity.

Credit and debt contracts

Credit transfers purchasing power against promised repayment. Maturity, collateral and default risk shape the contract and its cost.

Hypothetical two-year amortizing loan, 5% nominal annual interest on opening principal, no fees. Repayments are contractual promises; collateral does not guarantee full recovery.

Bank lending and money creation

A bank loan typically creates a matching deposit. Lending is constrained by profitability, credit demand, capital, funding and regulation.

Hypothetical changes to initially balanced sheets. Bank A has or obtains the 40 reserves needed for settlement; credit demand, risk, funding, capital and regulation constrain lending.

Interest and compounding

Interest prices the use of funds over time; compounding earns or accrues interest on prior interest as well as principal.

Hypothetical 100 principal at a constant 10% nominal annual rate; no deposits, withdrawals, fees or inflation adjustment. Solid: annual compounding. Dashed: simple interest. Balances are currency units; the plotted vertical scale starts at 90.

Present value and discounting

Discounting expresses future cash flows in today’s units using a rate matched to their timing and risk assumptions.

Hypothetical nominal payments discounted at 10% per year. This is a valuation under stated timing and rate assumptions, not a guarantee that the promised cash will be paid.

Nominal and real interest rates

A real interest rate adjusts a nominal rate for inflation. Expected inflation informs decisions; realized inflation determines the ex post purchasing-power return.

Hypothetical realized annual nominal return of 10% and inflation of 5%. The exact real return is 4.76%; subtraction gives the 5% approximation. Expected inflation would instead produce an ex ante estimate.

Yield curve

A yield curve compares borrowing yields across maturities for otherwise comparable claims; expectations and term premiums both influence its shape.

Illustrative alternative curves, not market observations or a forecast. Maturity varies while issuer, currency and credit risk are held comparable; inversion does not guarantee a recession.

Equity and debt financing

Equity is a residual ownership claim; debt creates contractual repayment obligations. Their cash-flow rights and loss exposure differ.

Hypothetical liquidation waterfall with one senior debt claim, limited liability and no other costs or claims. Residual equity is not a promised dividend; issuing equity is financing, not sales revenue.

Risk and return

Financial return combines income and value changes; risk concerns uncertain outcomes. Higher expected compensation for risk never guarantees higher realized return.

Hypothetical probability distributions for one period; bar height is outcome probability, not return size. Expected arithmetic returns are weighted averages, not guaranteed or compound growth rates.

Liquidity and solvency

Liquidity is the ability to meet payments when due; solvency concerns whether assets cover obligations. An institution can face one problem without the other.

Hypothetical asset values in common currency units; the 20 due now is part of total liabilities of 80. No immediate borrowing or sale of illiquid assets is assumed. Equity equals total assets minus liabilities.

Leverage

Borrowing increases asset exposure relative to equity, magnifying gains and losses on the equity cushion.

Hypothetical fixed debt, no interest or other flows. Initial assets/equity leverage is 1× versus 5×. A 20% asset loss would exhaust the leveraged equity; further losses imply negative accounting equity.

Bank runs and financial safeguards

A rush to withdraw short-term funding can force a bank to sell assets quickly; guarantees and emergency lending can reduce runs but create design tradeoffs.

Hypothetical one-bank run; a sale at 80% of carrying value bridges the cash gap. Deposit protection and central-bank assistance depend on jurisdiction and eligibility; liquidity support does not by itself repair insolvency.

Central banks

Central banks issue settlement money and conduct monetary policy; mandates, supervision roles and emergency-lending powers vary by jurisdiction.

Simplified conventional two-tier payment system. Households normally hold commercial-bank deposits and cash, rather than reserve accounts. Central-bank mandates, coin issuance and account-access arrangements vary.

Financial crises and amplification

Losses, funding stress and reduced lending can reinforce one another, turning financial fragility into lower spending and production.

Conceptual feedback mechanism, not an empirical timeline. Stronger capital, liquidity, borrower balance sheets or policy responses can weaken or interrupt these links.

Fluctuations and stabilization

Short-run models, demand and supply shocks, fiscal and monetary mechanisms.

Business cycles

Economic activity fluctuates around a changing longer-run path; expansions and contractions vary in duration and cause.

Hypothetical index, not observed GDP. The vertical axis starts at 90. Dashed trend is an estimate; recession dating and definitions vary across jurisdictions.

Potential output and the output gap

Potential output is an estimated sustainable level of production. The output gap compares actual output with that uncertain benchmark.

Read first
Business cycles
Hypothetical paths; dashed potential estimates are uncertain, not physical capacity ceilings. Gap = 100 × (actual − potential) / potential. Output above an estimate would give a positive gap.

Aggregate demand

An aggregate-demand curve relates planned total spending on domestic output to the price level under specified monetary and other conditions.

Illustrative AD₀: P = 160 − Y; AD₁: P = 180 − Y. Interest-rate, wealth and external-demand channels underpin aggregate demand; this is not a single-product substitution story. Other determinants are held fixed.

Short-run and long-run aggregate supply

Short-run supply captures output responses when some costs or prices adjust slowly; long-run supply links sustainable output to productive capacity.

Illustrative SRAS₀: P = 40 + Y, SRAS₁: P = 20 + Y. Vertical long-run supply marks model potential, not a physical maximum. A movement along SRAS is not permanent productivity growth.

Aggregate demand and supply shocks

A demand shock changes spending; a supply shock changes production conditions. They can move output similarly while moving the price level differently.

Illustrative baseline AD: P = 160 − Y; SRAS: P = 40 + Y. Dashed changes are AD intercept −20 or SRAS intercept +20. These are one-time price-level differences, not an assumption of continuing inflation.

Consumption and investment spending

Household consumption depends on income, wealth and expectations; firms’ investment depends on expected demand, financing costs and capacity.

Hypothetical first-round examples: marginal propensity to consume 0.75; new machinery spending is an assumed firm response. These are separate shocks. Real investment adds productive capital; purchasing existing securities does not.

Fiscal policy

Taxation, purchases and transfers influence activity and distribution. Effects depend on economic slack, financing, implementation and monetary responses.

Hypothetical first-round comparison, assuming prompt domestic purchases and no offsetting taxes in the period. Financing, imports, crowding out and later rounds can alter total output effects; transfers are not themselves GDP purchases.

Automatic stabilizers

Some taxes and transfers respond to income and employment without new legislation, cushioning changes in household income and aggregate demand.

Read first
Fiscal policy
Hypothetical schedule values, unchanged across scenarios; average taxes are progressive here. The government balance worsens by 24 without a new discretionary decision. This shows household income cushioning, not a complete multiplier model.

Fiscal multiplier

A fiscal multiplier measures output change relative to a fiscal impulse. Its size depends on leakages, capacity, financing and policy responses.

Hypothetical fixed-price closed economy with spare capacity, consumption propensity 0.5, no taxes or imports and no interest-rate response. Saving is the leakage. Taxes, imports and capacity limits change this special-case multiplier.

Crowding out and crowding in

Public spending can displace private activity through financing or resource pressure, or support it through demand and infrastructure; the result is contextual.

Conceptual competing channels. Monetary policy, financial openness, available resources and project quality affect the result; no one-for-one displacement or equal response at all levels of employment is assumed.

Public debt dynamics

Debt relative to GDP evolves with primary balances, interest costs, nominal growth and other stock-flow adjustments; no single debt ratio determines sustainability.

Hypothetical annual arithmetic, no valuation or other stock-flow adjustments. Primary surplus excludes interest and is 1% of current GDP: 1.03. Debt becomes 60 + 3 − 1.03 = 61.97; debt/GDP = 61.97/103.

Monetary policy transmission

Central-bank decisions influence financial conditions, spending and prices through several channels, with variable lags and uncertain strength.

Conceptual tightening under an interest-rate-based monetary regime. Market lending rates include risk and other spreads; currency and expectation responses are conditional, and inflation is not controlled instantly.

Quantitative easing

Central-bank asset purchases exchange securities for central-bank money and can affect yields and financial conditions; effects depend on market conditions and expectations.

Hypothetical purchase at carrying value through the seller’s commercial bank. The first two rows show asset and liability changes; the seller row shows two offsetting asset changes. Reserves are bank assets, not household deposits or a required onward loan.

Expectations and policy credibility

Beliefs about future income, prices and policy shape current choices. Credibility affects these beliefs but does not make policy announcements automatically effective.

Hypothetical expectation rules with zero term premium; the two-year nominal yield is approximated by the average expected annual short rate. Announcements, beliefs and subsequently realized rates are different quantities.

Phillips curve

Phillips-curve models link inflation to labor-market slack, expected inflation and supply shocks; they do not imply a permanent fixed inflation–unemployment menu.

Illustrative π = expected π − 0.5(u − 5), with rates in percent and no cost shock. The 5% benchmark is a model assumption, not a known constant; a positive cost shock would also shift this relation upward.

Policy lags and tradeoffs

Recognizing a shock, deciding policy and transmitting its effects take time; uncertainty and competing objectives complicate stabilization.

Conceptual sequence, not a calibrated time scale or ranking of policy speeds. Recognition, decisions, implementation and transmission depend on institutions and the shock. Bands are illustrative uncertainty, not estimated confidence intervals.

Long-run growth

Capital, skills, productivity and growth models.

Economic growth

Growth expands real production; growth per person depends on output growth relative to population growth and differs from broad development.

Hypothetical one-year changes, all levels indexed to 100 initially. Total real GDP grows 4% in both economies. Per-person growth is (1.04/1.01−1)×100 = 2.9703% in A and (1.04/1.03−1)×100 = 0.9709% in B. Subtracting population growth is only an approximation.

Capital accumulation and depreciation

Produced capital provides services over time; investment adds to its stock while depreciation reduces its productive value.

Hypothetical produced capital in constant-price units, one year: Knext = (1−δ)K + I = 0.9×100 + 20 = 110. Depreciation is 10% of opening capital; all investment is added at year-end, and revaluation is excluded. Land and financial claims are outside this capital measure.

Labor productivity

Labor productivity measures output per unit of labor input, such as an hour worked; it is not a measure of workers’ effort alone.

Hypothetical annual totals with matching coverage and constant-price output: Y/N = (Y/H)×(H/N). A: €50/hour×800 hours/person = €40,000/person. B: €40/hour×1,000 hours/person gives the same output/person. H/N is hours per resident, including nonworkers, not hours per worker.

Human capital

Skills, knowledge and health can raise productive capabilities. Education inputs are imperfect proxies for those capabilities.

Conceptual, hypothetical production comparison: equipment and four labor hours are fixed; greater task capability allows 4 versus 6 completed units. The difference is not an estimated causal return to schooling. Education, practice and health can contribute to skills and productive capability.

Total factor productivity

TFP summarizes output relative to measured factor inputs in a production model; it captures more than pure technology, including measurement and allocation effects.

Hypothetical normalization Y = A√(KL), with L = 4. A rises 1→1.5; at K = 4, output rises 4→6. This model uses Hicks-neutral A. A measured TFP residual can also reflect utilization, allocation and measurement, so the shift need not represent a new invention.

Growth accounting

Growth accounting decomposes output growth into weighted input growth and a productivity residual under explicit production and share assumptions.

Hypothetical Cobb–Douglas model Y = AK^(1/3)L^(2/3). Small-change approximation: gY ≈ (1/3)gK + (2/3)gL + gA. Capital growth 6% and labor growth 3% contribute 2 percentage points each; residual 1 point gives output growth ≈5%. Factor shares require model assumptions; residuals are not causal estimates.

Solow growth model

The Solow model combines saving, depreciation and diminishing returns to describe capital adjustment; sustained per-worker growth requires technological progress in its standard form.

Hypothetical continuous-time Solow model: f(k) = √k, s = 0.30, δ = 0.05/year, n = 0.025/year and no technological growth. dk/dt = 0.3√k−0.075k gives positive steady state k* = 16. Raising saving changes the steady-state level, not sustained per-worker growth in this baseline.

Conditional convergence

In growth models, economies farther below their own steady states can grow faster; this does not predict that all poorer economies automatically catch up.

Hypothetical continuous-time Solow paths, f(k)=√k and δ+n=0.075/year. Exact solution: k(t)=[s/0.075+(√k₀−s/0.075)e^(−0.075t/2)]². Left: s=0.3 for both, k*=16. Right: k₀=1 has s=0.15, k*=4; k₀=9 has s=0.3, k*=16. Simulations illustrate conditional convergence, not empirical proof.

International trade

Specialization, gains from exchange, distributional effects and trade barriers.

Comparative and absolute advantage

Absolute advantage means producing with fewer inputs; comparative advantage means giving up less of another good. Specialization can therefore benefit a producer that is less productive in both goods.

Hypothetical constant labor requirements with 12 labor hours available in each country. Opportunity cost uses ratios of hours per unit. A’s greater productivity in both goods does not eliminate B’s comparative advantage.

Gains from trade

Trade can expand feasible consumption by allowing specialization and exchange at mutually acceptable prices. Aggregate gains do not guarantee that every worker or household benefits.

Hypothetical constant returns, 12 labor hours per country, no trade costs. The 0.75 food/cloth price lies between opportunity costs 0.5 and 1. Exports and imports balance: 4 cloth × 0.75 = 3 food. Panel axis scales differ.

Trade and the distribution of gains

Opening trade changes domestic prices and returns to workers and owners. Consumers and exporting producers may gain while import-competing groups face losses and adjustment costs.

Hypothetical competitive small-country model: demand Q = 100 − 10P, supply Q = 10P, fixed world price 2, no trade costs. Blue shows the producer-surplus loss transferred to consumers; green is the net gain. Adjustment costs and distribution beyond these groups are omitted.

Tariffs

A tariff is a tax on imports. In a competitive small-country model it raises the domestic price, reduces imports, generates revenue and creates production and consumption losses.

Hypothetical competitive small-country model with fixed world price 2; Qd = 100 − 10P and Qs = 10P. Tariff revenue is 40 and deadweight loss 10 currency units per period. The tariff remains below the prohibitive level.

Import quotas and quota rents

An import quota limits quantity rather than directly taxing it. A binding quota raises scarcity rents, whose recipients depend on how import rights are allocated.

Read first
Tariffs
Hypothetical competitive markets; P is currency per unit. Fixed world price 2 and identical demand/supply to the tariff figure. With freely allocated licenses, rents go to license holders, which can include foreigners. Auctioning licenses can capture rents for government; equivalence depends on the stated conditions.

International finance

Currency prices, exchange-rate arrangements and external transaction accounts.

Exchange rates and currency appreciation

An exchange rate quotes one currency in another. Appreciation means a currency buys more foreign currency; whether the displayed number rises depends on the quotation convention.

Hypothetical currencies D and F, no fees or bid–ask spread. Each pair of reciprocal rates multiplies to one. A fall in D/F is domestic appreciation; the reciprocal F/D quote rises.

Real exchange rate

The real exchange rate compares the price of foreign and domestic goods after currency conversion. Nominal exchange rates and both countries' price levels determine it.

Hypothetical actual basket prices held fixed while the nominal quote e rises. Under q = eP*/P with e in domestic/foreign units, a higher q is real domestic depreciation. Price-index versions require an explicit normalization.

Purchasing power parity

PPP compares currencies by what they purchase. Statistical PPP conversion factors support real-volume comparisons; the theory that exchange rates equalize prices is limited by trade costs and nontraded goods.

Hypothetical one-basket comparison with identical quantities. Statistical PPPs use broad, weighted price comparisons. Trade costs, nontraded goods and other differences can keep market rates away from a basket’s PPP.

Fixed and floating exchange rates

Exchange-rate regimes specify how authorities manage currency values, from market floating to pegs and shared currencies. Stability commitments constrain policy and can require reserves or other adjustments.

Hypothetical flows per period. Private supply: e = 0.4Q + 0.4; demand rises from e = 3.6 − 0.4Q to e = 4.4 − 0.4Q. Solid supply is unchanged. Reserve sales defend against domestic depreciation and cannot continue without limit.

Balance of payments

The balance of payments records transactions between residents and nonresidents over a period. Current, capital and financial accounts are linked by double-entry accounting, with measured discrepancies recorded separately.

Hypothetical transaction, one common currency unit, no fees or valuation changes. FA uses net asset acquisition minus net liability incurrence. Measured errors and omissions are omitted here; residents’ foreign deposits are external assets.

Current account

The current account combines trade in goods and services with primary and secondary income. A trade deficit and a current-account deficit are different measures.

Hypothetical net balances, common currency units in one period. Primary income includes labor and investment income; secondary income includes current transfers. Purchases of financial assets belong in the financial account, not income.

Financial account and cross-border investment

The financial account records changes in foreign financial assets and liabilities through transactions. Direct investment, portfolio investment and reserves differ in form; they are distinct from the narrower capital account.

Hypothetical completed transactions in common currency units: assets +30, liabilities +50, CA −20 and KA 0. No valuation effects or residual foreign cash position. Equity investment and borrowing are financial-account entries, not capital transfers.

Development and distribution

Capabilities, poverty, inequality, redistribution and economic institutions.

Economic development

Development concerns improvements in people's capabilities and living conditions, including health, education and material security. Output growth can support development without fully measuring it.

Hypothetical comparison in common purchasing-power units (CU). The bars have separate scales for each indicator; they compare societies within a row, without assigning a cause.

Human Development Index

HDI combines normalized health, education and income dimensions using a geometric mean. It summarizes average attainment while leaving distribution and many aspects of wellbeing outside the headline measure.

Hypothetical inputs using UNDP’s 2025 technical-note goalposts; income is GNI per person in 2021 PPP dollars. Education averages the normalized expected and mean schooling indicators. Rounded values shown; raw units are never averaged.

Poverty and poverty thresholds

Poverty measures deprivation against a specified standard. Income or consumption poverty depends on the threshold, household adjustment, prices and whether the standard is absolute or relative.

Hypothetical per-person income, fixed prices and equal weights. This is an absolute line; income equal to the line is not counted as poor. A relative line would instead move with a distribution statistic.

Poverty headcount and depth

A headcount counts people below a poverty line; a poverty gap measures how far their resources fall short. Transfers to the poorest can reduce depth without changing the headcount.

Hypothetical equal-size, one-person households; all income is in fixed-price CU per month. Shading is the shortfall. A person exactly at the line has zero gap and is not counted as poor.

Income and wealth inequality

Inequality describes the distribution of resources across people or households. Income is a flow over time; wealth is net assets at a date, and their distributions need not match.

Hypothetical households, ordered A–D in both panels. Income is a monthly flow; net wealth is assets minus liabilities at one date. Scales differ across panels, and negative wealth is retained.

Lorenz curve and Gini coefficient

A Lorenz curve plots cumulative resource shares against the population ranked from poorest to richest. For nonnegative resources, the Gini summarizes the area between this curve and equal shares.

Hypothetical nonnegative income distribution. The polyline, shaded inequality area and Gini use the same sorted incomes. With zero total income, income shares and this Gini calculation would be undefined.

Redistribution through taxes and transfers

Taxes, transfers and public services change the distribution of resources. Evaluating redistribution requires attention to eligibility, take-up, financing and behavioral responses as well as distributional aims.

Hypothetical one-person households with full take-up stipulated. Gray bars show market income; violet bars show disposable income. Taxes finance all transfers and leave 10 CU for other spending or saving; behavior is held fixed.

Benefit withdrawal and work incentives

Means-tested benefits can fall as earnings rise, reducing the gain from additional paid work. High effective marginal withdrawal rates are one possible poverty-trap mechanism, not a universal account of poverty.

Hypothetical weekly schedule: benefit = max(0, 40 − 0.5 × max(earnings − 20, 0)); resources = 0.8 × earnings + benefit. The phase-out is gradual and continuous, with no benefit cliff.

Institutions and economic incentives

Institutions are rules and enforcement arrangements that shape cooperation, exchange and power. Their effects depend on implementation and context, including informal norms as well as formal law.

A hypothetical contractual mechanism, not a cross-country causal estimate. Rules and enforcement arrangements are institutions; courts, firms and agencies are organizations operating within them.

Property rights and contracts

Property rights specify permitted uses and claims over resources; contracts arrange exchanges and obligations. Enforceability, transaction costs and incomplete terms affect which agreements are feasible.

Hypothetical land transaction. Ownership is a bundle of qualified rights, not unlimited use. Contracting parties arrange payment and conveyance; enforcement institutions resolve claims within the applicable legal setting.

Work and wages

Labor supply and demand, employer power, wage setting, discrimination and migration.

Labor demand and marginal revenue product

An employer's labor demand is derived from what workers add to revenue. In a basic profit-maximizing model, hiring compares marginal revenue product with the marginal cost of labor.

Hypothetical discrete hiring schedule, fixed capital and competitive output market. MRP = marginal product × price here; with product-market power, use marginal revenue instead. The third worker adds less revenue than their wage.

Labor supply and the work-leisure choice

Labor supply reflects participation and hours choices subject to wages, nonlabor resources and constraints. A wage rise changes both the reward for work and the ability to afford leisure.

Hypothetical individual budget: consumption = 20 + wage × (16 − leisure). A and B are alternative feasible responses for different preferences, not simultaneous choices by one person. Income and substitution effects determine the hours response.

Wage differentials

Pay differs with skills, job conditions, bargaining power and unequal opportunities. A wage premium can compensate for unpleasant work, but observed gaps rarely isolate one mechanism.

Hypothetical wage offers and injury risks per worker-week. A compensating differential is one possible mechanism; these offers alone do not identify a causal risk premium or explain broader group pay gaps.

Monopsony and employer wage-setting power

An employer has monopsony power when attracting additional workers requires improving its offer. With an upward-sloping labor supply, marginal hiring cost can exceed the wage, reducing employment in a simple model.

Hypothetical single-wage employer: w = 4 + 0.5L; total wage bill = wL, so MLE = 4 + L. Hire at MRP = MLE, then read pay from supply. The competitive benchmark would be L = 12 and w = 10.

Minimum wages in different labor markets

A wage floor changes pay and hiring incentives. Competitive and monopsonistic models can predict different employment responses, so model assumptions and empirical identification matter.

Hypothetical matched models: S = 4 + 0.5L, D = 16 − 0.5L; monopsony MLE = 4 + L without a floor. The floor makes marginal hiring cost flat at 9 up to 10 willing workers. At floor 11, both models hire 10. These conditional examples do not predict every real labor market.

Unions and collective bargaining

Collective bargaining lets workers negotiate pay and conditions together. Outcomes depend on bargaining coverage, employer power, outside options and the legal setting.

Hypothetical voluntary bargaining for a fixed job in a setting permitting collective negotiation. Outside options bound the 30 CU surplus. The visual varies the division of surplus while holding employment and productivity fixed.

Employment discrimination

Discrimination changes access, treatment or pay because of group membership rather than relevant job characteristics. Raw earnings gaps mix multiple mechanisms and cannot by themselves identify discrimination.

Both examples are hypothetical. Each dot in the experiment represents 10 applications. Randomizing the cue with otherwise equivalent applications can identify a callback response to that cue; it does not identify all causes of the raw wage gap or lifetime earnings differences.

Migration and labor markets

Migration changes where people live and work, affecting labor supply, demand, skills and household transfers. Effects differ across groups, locations and time as firms and workers adjust.

Conceptual channels, not signed effect estimates. Labor demand may adjust through consumption and investment; migrants, incumbent workers, owners and origin households can experience different effects over different horizons.

Market failures and public policy

External effects, collective resources, information problems and evaluation of policy mechanisms.

Market failure

Market failure occurs when decentralized market outcomes leave feasible gains in economic efficiency unrealized. External effects, public goods, market power and information problems are distinct mechanisms.

Conceptual examples of unrealized efficiency gains. The candidate responses require their own evidence about implementation and costs; unfairness or an unpopular price alone does not establish market failure.

Externalities

An externality is an uncompensated effect of an action on others outside the transaction. Negative effects can cause overproduction and positive effects underproduction relative to a social optimum.

Hypothetical linear models. Negative: MB = 12 − Q, MPC = 2 + Q, MSC = 4 + Q. Positive: MC = 2 + Q, MPB = 10 − Q, MSB = 12 − Q. The 2 CU marginal external effect is added exactly once.

Corrective taxes and subsidies

A corrective price makes decision makers face external costs or benefits. At an efficient target, a Pigouvian tax matches marginal external damage rather than total damage.

Read first
Externalities
Hypothetical one unit of emissions per output unit and constant marginal damage of 2 CU per emissions unit. The output tax is therefore 2 CU per unit. With a different emissions intensity, convert between emissions and output tax bases; total damage is not the tax rate.

Cap and trade

A cap limits aggregate covered emissions while tradable permits let firms shift abatement toward lower-cost options. The permit price emerges from scarcity and trading rules.

Read first
Externalities
Hypothetical interior cost minimum with total abatement 10. Five permits are initially given free to each firm; A sells 2.5 permits to B. Permit trades redistribute payments without changing the cap. The assumed MAC curves determine the common price, not the free allocation.

Coasean bargaining

With clear rights and sufficiently low bargaining costs, affected parties may negotiate away external inefficiency. Initial rights influence payments, while transaction costs and information can block agreement.

Hypothetical transferable monetary benefits, clear rights, negligible bargaining costs, enforceability, and no binding wealth constraints or wealth effects. Both rights assignments support cleanup here, while payments differ. Transaction costs can prevent the bargain.

Public goods

A public good is nonrival and difficult to exclude others from using. These properties can weaken private payment incentives, regardless of whether provision is public or private.

Examples are conditional on the stated setting. Public financing does not by itself make a good nonrival or nonexcludable; congestion, access rules and technology can change the classification.

Free riding

A free rider benefits from a collectively provided good without contributing. Individually sensible nonpayment can leave provision below the level the group values.

Read first
Public goods
Hypothetical additive public-good benefits. A’s payoff is 15 times the number of contributions minus A’s own cost, and similarly for B. Arrows show the individual incentive to withhold, despite a larger combined payoff from both contributing.

Common-pool resources

Common-pool resources are costly to exclude users from but are depleted or congested by use. Open access can encourage overuse; enforceable community rules can sometimes sustain cooperation.

Read first
Public goods
Hypothetical one-period stock accounting in fish units: 100 + 20 − 30 = 90 versus 100 + 20 − 20 = 100. Renewal is stipulated for this period, not a constant biological law. Common property with enforceable rules differs from unrestricted open access.

Collective action and cooperation

Collective action coordinates contributions toward a shared objective despite individual incentives to withhold effort. Monitoring, repeated interaction and legitimate rules can change those incentives.

Read first
Free riding
Hypothetical risk-neutral private payoffs in CU, holding all other payoffs fixed. Fines are transfers, and enforcement uses resources. Legitimacy, mistaken detection, appeal and monitoring costs matter; sanctions do not always improve cooperation.

Asymmetric information

Information is asymmetric when one party knows something relevant that another does not. Hidden quality or actions can change contracts, prices and which exchanges occur.

Hypothetical transaction. The information gap exists before contracting; a price alone need not reveal quality. Inspection is a possible verification mechanism, with its own cost and accuracy limits.

Adverse selection

Adverse selection arises when hidden characteristics affect who enters a transaction. A pooled price may attract higher-cost participants and discourage lower-cost ones.

Hypothetical claim size 100 CU and hidden pre-contract risk. Each dot represents five people. High-risk willingness to pay is stipulated as 60, so that group remains at premium 30 or 50. No administrative costs; participation rules are explicit rather than inferred from expected claims alone.

Moral hazard

Moral hazard occurs when imperfectly observed actions change after a contract alters incentives. Insurance may reduce precautions even without fraud or bad intent.

Hypothetical risk-neutral choice after contracting, with effort hidden and premiums fixed. Expected private cost includes precaution cost plus uninsured expected loss. Full coverage removes this example’s monetary precaution incentive; a large deductible restores it while providing less risk protection.

Signaling and screening

An informed party can signal hidden qualities; a less-informed party can screen by offering choices or tests. Useful signals must be more credible than costless claims.

Hypothetical credible, enforceable warranty and risk-neutral sellers. High quality earns 15 − 5 = 10 from the warranty; low quality loses 15 − 25 = −10. The initiating party distinguishes the two mechanisms; a costless quality claim would not separate these types.

Principal-agent problem

A principal delegates to an agent whose incentives or information differ. Monitoring and performance contracts can help, but imperfect measures may reward the wrong behavior.

Hypothetical delegated maintenance contract. Repair counts are an observable proxy, not the full objective. The unintended response is a possible incentive effect; monitoring quality or changing the contract may help at a cost.

Social cost-benefit analysis

Cost-benefit analysis compares a policy's incremental social benefits and opportunity costs against a counterfactual. Valuation, distribution, uncertainty and timing affect the comparison.

Hypothetical one-period project against a no-project baseline, all values in the same price-year CU. Health value includes the counted health channels once. Tax collection and distortion costs are assumed zero here; distribution and omitted effects require separate evaluation.

Government failure and rent seeking

Policy can fail through weak information, incentives or implementation. Rent seeking expends resources to obtain preferential treatment rather than create value, including when benefits are concentrated and costs dispersed.

Read first
Market failure
Hypothetical fixed-output example. Beneficiary gains are concentrated and payer costs dispersed; lobbying uses real labor and materials. This isolates one rent-seeking cost and does not establish that all lobbying, subsidies, or public action reduce welfare.

Economic evidence

Measurement, causal reasoning, empirical designs, uncertainty and transfer of findings.

Distributions, means and medians

A distribution describes how outcomes vary across observations. The mean uses every value, while the median locates the middle; identical averages can conceal very different outcomes.

Hypothetical daily incomes with equal weights. A: sum 120 / 5 = 24. B: sum 180 / 5 = 36. The outlier remains visible on the same scale, and both medians are the third sorted observation.

Sampling and measurement

A sample describes a target population only through its selection and measurement process. Nonresponse, missing observations and inconsistent definitions can bias results even with many observations.

Hypothetical counts, not outcome values. Group A is violet and B blue throughout. Equal initial selection does not prevent differential nonresponse and invalid measurement from changing composition; missing values are not coded as zeros.

Correlation and causation

Correlation describes association; causation asks how an intervention changes outcomes. Common causes, selection and reverse causality can produce association without the proposed causal effect.

Hypothetical observations in arbitrary X and Y units. Arrows are competing causal hypotheses, not effects identified by the plotted association. Timing, design and substantive assumptions must distinguish them.

Counterfactual outcomes

A causal effect compares an outcome under an intervention with the outcome that would occur without it. Both outcomes cannot generally be observed for the same unit at the same time.

Simulated CU per day, not observed causal data. The dashed path represents the unobserved potential outcome for this same person and date. Another person’s income is not automatically a valid substitute.

Causal identification

Identification states what data and assumptions allow a causal effect to be learned. More precise associations cannot repair a design whose comparison group fails to represent the counterfactual.

Conceptual adjustment design for applicants. Identification additionally requires consistent treatment and outcome definitions and appropriate weighting for the target effect. The arrows encode assumptions; sample size cannot establish that no relevant confounder was omitted.

Randomized experiments

Random assignment creates comparable treatment groups in expectation. Differences in outcomes can estimate assignment effects, while attrition, noncompliance and spillovers need separate consideration.

Hypothetical allocation. Randomization creates baseline comparability in expectation, not identical realized groups or a representative population. No effect is calculated from these incomplete outcomes; selective attrition, noncompliance and spillovers require separate checks.

Natural experiments

A natural experiment uses an external event or rule that creates plausibly exogenous exposure. Its causal value depends on the assignment argument, not simply on a striking before-after change.

Hypothetical policy-generated variation, not researcher assignment. The eligibility event alone is not a causal argument: sorting, coincident rules and other pathways must be investigated. Nearby comparison supports a local claim rather than a population-wide one.

Regression and control variables

Regression summarizes how an outcome varies with measured predictors. Controls can address some confounding, but omitted variables, measurement error and inappropriate controls can undermine causal interpretations.

Hypothetical data: lower-experience pairs (10,10), (12,12), (14,14); higher-experience pairs (8,18), (10,20), (12,22). The pooled fitted slope is −4/11; within-group slopes are +1. This adjustment is causal only under the assumed confounding structure; controlling for mediators or selection variables can instead introduce bias.

Difference-in-differences

Difference-in-differences compares outcome changes in treated and comparison groups. A causal interpretation requires parallel untreated trends and absence of confounding differential shocks.

Hypothetical group means. Causal interpretation requires parallel untreated trends, no differential coincident shock, and suitable stable composition. Similar pretreatment trends are a diagnostic, not proof of the missing post-treatment counterfactual.

Regression discontinuity

Regression discontinuity compares observations near an eligibility cutoff. With smooth potential outcomes and no precise manipulation, an outcome jump can identify a local treatment effect.

Hypothetical fitted lines and observations. A local causal interpretation requires smooth untreated potential outcomes, no precise score manipulation, and no other policy jump at 50. The cutoff jump is distinct from the average gap across all eligible and ineligible people.

Sampling uncertainty and confidence intervals

Estimates vary across samples. A confidence interval comes from a procedure with a stated long-run coverage rate under its assumptions; it does not resolve design bias.

Hypothetical intervals in outcome units, illustrating a nominal 95% procedure; eight intervals need not achieve exactly 95% realized coverage. The 95% describes long-run coverage under the model, not a 95% probability that the fixed mean lies in a particular realized interval. Sampling intervals do not remove design bias.

Hypothesis tests and p-values

A p-value measures how extreme the observed result is under a specified null model. It is neither the probability that the null is true nor a measure of effect size.

Hypothetical z-test with a continuous standard-normal null. Both tails are shaded. The p-value is an extremeness probability under the null model; it is not the probability the null is true, an effect size, or evidence that a study design identifies causation.

Effect size and practical significance

An effect's substantive importance depends on its magnitude, units and context. A precisely estimated tiny effect and an imprecisely estimated large effect pose different policy questions.

Hypothetical estimates on the same outcome scale. A is precise but below the stipulated practical threshold; B is compatible with zero and with important benefits. The threshold is a context-specific value judgment, not a universal significance rule.

External validity and policy transfer

External validity concerns whether evidence applies beyond the studied people, setting or intervention. Different institutions, implementation or economy-wide responses can change an effect when a policy scales.

Conceptual scale comparison; the dots denote scope, not a measured population. Random assignment supports an internal comparison in the pilot but does not make its volunteers representative or hold implementation and equilibrium responses constant at scale.

Behavior and choice

Limits to attention, reference dependence, intertemporal choice and choice architecture.

Bounded rationality and limited attention

People decide with limited time, attention and processing capacity. Shortcuts and simplified choice rules can be useful while producing systematic departures from fully informed optimization.

Hypothetical fixed preferences and scores. The explicit attention constraint is inspecting the first two listed options; all choices remain available. A different preference or a cost of searching could change whether stopping early is desirable.

Loss aversion and reference dependence

Reference-dependent preferences evaluate outcomes relative to a benchmark. Loss aversion gives losses greater weight than comparable gains in some settings, distinct from ordinary aversion to risk.

Hypothetical curve: v(x) = √x for gains and −2√(−x) for losses. The factor 2 is an illustration, not a universal estimate. Reference dependence and the loss/gain asymmetry differ from ordinary risk aversion over final wealth.

Present bias and time inconsistency

Present bias places extra weight on immediate costs and rewards. Plans made for the future may reverse when an option becomes immediate, even without new information.

Hypothetical quasi-hyperbolic model: immediate value is the reward; delayed value is βδᵗ times the reward, with β = 0.7 and δ = 0.99 per day. Rewards, risk and one-day gap stay fixed. With β = 1, exponential discounting prefers 110 in both comparisons.

Framing effects

Descriptions and reference points can influence choices even when economic consequences are equivalent. Framing effects are empirical patterns whose size and direction depend on the setting.

Hypothetical equivalent descriptions; no response proportions are asserted. A framing effect would be a change in choices between these descriptions despite identical final-outcome distributions. Its presence and size require empirical evidence.

Defaults and nudges

Choice architecture can change behavior by altering defaults, salience or friction while preserving options. Effectiveness and welfare depend on whose goals are served and how easily choices can be changed.

Static schematic forms with identical options and the same effort to change the selection. Defaults can alter salience or friction; any uptake and welfare effect depends on preferences, comprehension, implementation and whose goals are served.