Session 13: Monopoly — Pricing Power, Welfare Cost, and the Logic of Discrimination

Part of the Microeconomics Knowledge System. This post covers Session 13 — single-price monopoly, the multi-plant firm, price discrimination, long-run equilibrium, and the policy debate around market power.

Session 12 established perfect competition as the benchmark — the case where no firm has pricing power. Session 13 moves to the opposite extreme: a single seller facing the entire market demand curve with no competition to discipline its pricing.

The key shift is this: in perfect competition, the firm takes price as given and chooses quantity. In monopoly, the firm chooses quantity and the market demand curve determines what price it can charge. These are not symmetric — they lead to fundamentally different revenue behaviour, output decisions, and welfare outcomes.

This session has five parts. Work through them in order — each one builds on the previous.

Part 1: Single-Price Monopoly — The Core Model

Why monopoly exists

A monopoly requires two conditions: a single seller, and barriers that prevent entry. Barriers take different forms:

• Legal barriers: patents, copyrights, exclusive licences (pharmaceuticals, broadcast licences)
• Cost barriers: high fixed costs creating natural monopoly (railways, water utilities)
• Resource barriers: control of a critical input (early De Beers and diamonds)
• Strategic barriers: ecosystem lock-in, switching costs, network effects

Without entry barriers, above-normal profit attracts competitors and the market moves toward competition. The barriers are what make monopoly persist.

Why MR < P — the central fact of monopoly

In perfect competition, the firm is so small that selling one more unit does not affect price. MR = P.

In monopoly, the firm faces the entire downward-sloping market demand curve. To sell one more unit, it must lower price — not just for that unit, but for all units already being sold (assuming a single uniform price). This price reduction effect means each additional unit generates less revenue than its price suggests.

Formally, for a monopolist:

MR = d(TR)/dQ = P + Q(dP/dQ)

Since dP/dQ < 0 (demand slopes down), the second term is negative. Therefore MR < P.

The gap between P and MR grows as quantity increases — the further down the demand curve the monopolist moves, the steeper the price reduction needed per additional unit.

Linear demand — the workhorse model

If demand is linear: P = A − BQ

Then:

• TR = AQ − BQ²
• AR = A − BQ (same as demand curve)
• MR = A − 2BQ (same intercept, twice the slope)

The key geometric result: MR hits the quantity axis at A/2B — exactly halfway between the origin and where demand hits the quantity axis at A/B.

On a graph: AR and MR share the same vertical intercept (at A), but MR falls twice as steeply. Every horizontal line from the price axis intersects MR at half the distance to the demand curve.

The elasticity connection

Using the definition of price elasticity ε = (dQ/dP)(P/Q):

MR = P(1 + 1/ε)

Since ε < 0 for a downward-sloping demand curve:

• When |ε| > 1 (elastic demand): MR > 0. Selling more increases total revenue.
• When |ε| = 1 (unit elastic): MR = 0. TR is at its maximum.
• When |ε| < 1 (inelastic demand): MR < 0. Selling more reduces total revenue.

Critical implication: A profit-maximising monopolist never voluntarily operates in the inelastic portion of demand. If it did, it could raise price, sell less, and increase revenue while reducing costs — strictly more profit. So the monopolist always operates where |ε| > 1.

Finding the optimum: MR = MC

The profit-maximising condition is the same as always — MR = MC. But the procedure in monopoly is a two-step process that students often mix up:

Step 1: Set MR = MC to find optimal quantity Q*

Step 2: Read the corresponding price P* from the demand curve (AR) at Q*

This is the critical distinction. The monopolist does not price at MC. It prices at what the demand curve says consumers will pay for Q* units — which is higher than MC.

The markup pricing rule

Rearranging MR = P(1 + 1/ε) and setting MR = MC:

P = MC / (1 + 1/ε)

Or equivalently:

(P − MC)/P = −1/ε

This is the Lerner Index of market power. It tells you the markup as a fraction of price:

• Perfect competition: ε → −∞, so Lerner Index → 0 (P = MC, no markup)
• Monopoly with inelastic demand (|ε| = 2): Lerner Index = 0.5 (markup is 50% of price)
• The less elastic the demand, the higher the markup the monopolist can sustain

The Lerner Index ranges from 0 (pure competition) to 1 (pure monopoly with perfectly inelastic demand). A monopoly with highly elastic demand has significant market power in structure but limited ability to exercise it — because customers will walk.

Graph logic — read this cold

Price
  |        MC      ATC
  |       /       /
P*|------/-------(profit rectangle top)
  |     /   *   /
C*|----(profit rectangle bottom)
  |   / MR=MC
  |  /
  | /          D = AR
  |/
  +----Q*---------→ Quantity
       |   MR

• MR = MC at Q* — this determines quantity
• P* is read from the demand curve (AR) at Q*
• C* = ATC at Q* — this determines unit cost
• Profit rectangle = (P* − C) × Q
• MC and MR both pass through Q, but P lies above both

The welfare cost: deadweight loss

A competitive market would produce where P = MC. A monopolist restricts output to Q* < Q_competitive, raising price to P* > MC.

The units between Q* and Q_competitive would have been worth more to buyers than they cost to produce — but they are not produced. This foregone value is deadweight loss (DWL) — the triangular area between the demand curve and MC, between Q* and Q_competitive.

DWL is not a transfer — it is genuine destruction of value. Consumer surplus shrinks by more than producer surplus grows. The difference is the triangle lost.

Why does this matter? It is the economic justification for antitrust regulation. The monopolist's private gain comes partly at consumers' expense (a transfer) and partly at society's expense (the DWL triangle). The DWL is the part that motivates policy.

Part 2: The Multi-Plant Monopolist

The problem

A monopolist with two plants (different locations, vintages, or technologies) must decide:

• How much to produce in total
• How to split that production across the two plants

Plants have different cost structures: C₁(Q₁) and C₂(Q₂). Consumers see only the total quantity and cannot distinguish which plant produced their unit.

The allocation rule

The profit function is:

π = P(Q₁ + Q₂) · (Q₁ + Q₂) − C₁(Q₁) − C₂(Q₂)

Taking partial derivatives and setting them to zero gives the optimality conditions:

MR = MC₁ and MR = MC₂

Which implies: MC₁ = MC₂

Intuition: If MC₁ > MC₂, the firm is spending more to produce the marginal unit at Plant 1 than at Plant 2. Shifting one unit from Plant 1 to Plant 2 produces the same output at lower cost — increasing profit. This reallocation continues until marginal costs equalise.

The firm then sets total output where MR equals this common MC level.

Why this matters in practice

This is not just a textbook result. The MC equalisation principle is the economic foundation for:

• Production scheduling across facilities with different shift costs
• Make-or-buy decisions (outsource to the lower-MC source until MCs equalise)
• Network optimisation (route traffic through lowest-cost nodes until node MCs equalise)
• Multi-site capacity planning (invest in capacity at the lower-MC location)

Any time a firm operates multiple production units with different cost structures, this logic applies.

Part 3: Price Discrimination — Capturing What You Leave Behind

The basic problem with uniform pricing

A monopolist charging a single price P* leaves money on the table. Some customers would willingly pay more than P* — that is consumer surplus the firm is giving away. And some customers value the product above MC but below P* — those customers are not served, creating deadweight loss.

Price discrimination addresses both. By charging different prices to different buyers based on willingness to pay, the firm can capture more consumer surplus and — often — expand output closer to the efficient level.

The essential condition: no arbitrage

Price discrimination only works if customers paying low prices cannot resell to customers facing high prices.

If resale is possible, price discrimination collapses. Buyers in the cheap segment arbitrage the gap, undercutting the firm in the expensive segment. The practical implication is that discrimination requires either:

• Physical barriers: geography, customs, shipping costs preventing cross-border resale
• Legal barriers: contracts, licences, warranties voided on resale
• Product design barriers: versions that are difficult to transfer (streaming subscriptions, non-transferable tickets)
• Time barriers: perishable access (last-minute pricing, concert seats)
• Identity barriers: student IDs, senior cards, verified accounts

The professor's emphasis: no-arbitrage is the key constraint. Understanding why a particular pricing scheme works or fails almost always comes down to whether arbitrage can be blocked.

First-degree price discrimination (perfect discrimination)

What it is: Charge each unit at exactly the maximum willingness to pay for that unit.

Result: The entire demand curve becomes the firm's marginal revenue curve. The firm expands output all the way to Q_competitive (where P = MC) because every unit generates its full value as revenue.

Consumer surplus: Zero. Every surplus dollar is extracted by the firm.

Welfare: Efficient — output equals the competitive level, DWL = 0. But all surplus goes to the producer.

Real-world approximation: This is theoretically clean but practically impossible in pure form. Approaches include:

• Auctions (each bidder reveals their maximum WTP through the bidding process)
• Personalised pricing (using data to estimate individual WTP — increasingly feasible with digital platforms)
• Negotiated prices (car dealerships, B2B contracts)

The professor used the example of auctioning collectibles sequentially — releasing one item at a time, each time extracting the highest WTP buyer from the pool.

Second-degree price discrimination (quantity-based screening)

What it is: Price varies by quantity purchased, not by customer identity. Customers self-select into tiers based on their usage.

Examples discussed:

• Bulk discounts (buy 10 units at Rs. 50 each, buy 100 units at Rs. 40 each)
• Tiered utility pricing (first 100 units at a base rate, next 100 at a higher rate)
• SaaS pricing tiers (free/starter/pro/enterprise)

The mechanism: High-WTP customers, who buy more, face different effective prices than low-WTP customers who buy less. The firm does not need to identify customer type — quantity choice reveals it.

No-arbitrage concern: Can the low-quantity buyer simply buy multiple small batches? Product design and minimum order requirements often prevent this.

Third-degree price discrimination (market segmentation)

What it is: The firm charges different prices to different identifiable market segments — geographies, demographics, use cases — based on differences in demand elasticity.

The model: One production facility, two markets (or more). The firm chooses Q₁ and Q₂ to maximise total profit:

π = R₁(Q₁) + R₂(Q₂) − C(Q₁ + Q₂)

Taking FOCs:

MR₁ = MC and MR₂ = MC

This implies: MR₁ = MR₂

The pricing rule: Since MR = P(1 + 1/ε), equalising MR across markets means:

• The market with more elastic demand (larger |ε|) gets a lower price
• The market with less elastic demand (smaller |ε|) gets a higher price

Intuition: In the elastic market, a price cut generates significant extra sales — it is worth lowering price to capture volume. In the inelastic market, buyers will not shift much quantity in response to price changes — the firm can charge more without losing much demand.

Real examples discussed

Textbooks (India vs US): Third-degree discrimination across geographies. Indian editions cost a fraction of US editions. Arbitrage is blocked by customs, import restrictions, and "not for sale outside India" prints. Demand in India is more elastic (lower incomes, more substitutes); in the US, students often have no choice. The price gap reflects the elasticity gap.

E-books as the counter-example: Digital goods make arbitrage nearly free. A user can share a file globally in seconds. This is why e-book pricing is harder to segment geographically than physical books — and why publishers use DRM, regional accounts, and licensing restrictions instead of pure price segmentation.

Airlines: Business travellers have inelastic demand (must travel on specific dates, employer pays). Leisure travellers have elastic demand (flexible, many alternatives). Airlines discriminate via advance purchase requirements, change fees, refundability, and seat class — all of which function as self-selection mechanisms revealing traveller type.

Harry Potter (hardcover → paperback → digital): Intertemporal discrimination. Enthusiasts with high WTP buy hardcovers at launch. As time passes, price falls and a second segment (lower WTP) enters via paperback. Then digital versions reach the most price-sensitive segment. The sequence extracts value from each segment in turn.

Inox popcorn: No outside food is a space restriction that creates a localised monopoly for cinema food. Once you are inside, substitutes are zero — your effective elasticity collapses. The cinema monetises this with popcorn margins that would be impossible outside.

Key insight from all these cases: Price discrimination is not exotic or rare. It is standard practice in airlines, hotels, software, pharmaceuticals, textbooks, entertainment, and professional services. The question to ask in any industry is: what prevents arbitrage here, and how does the firm use that to segment?

Welfare effects of price discrimination

Compared to single-price monopoly:

• Output is usually higher under discrimination (the firm serves segments it would otherwise exclude)
• DWL is usually lower (approaching zero under perfect discrimination)
• Consumer surplus shifts toward the producer (some segments pay more than they would under uniform pricing)

The welfare judgment is ambiguous: discrimination reduces DWL (efficiency gain) but redistributes surplus toward the producer (distributional concern). Whether you view discrimination as good or bad depends on which you weight more.

Part 4: Long-Run Monopoly Equilibrium

Why monopoly does not reach minimum ATC

In perfect competition, long-run forces drive firms to minimum ATC — entry pushes price down until only the most efficient scale survives.

In monopoly with sustained entry barriers, this discipline does not operate. The monopolist can persist at any point on the demand curve where it earns profit — including points where ATC is above its minimum.

The excess capacity result: A monopolist produces at Q* < Q_min_ATC. It deliberately restricts output to keep price high. The "spare capacity" it maintains relative to minimum efficient scale is the measure of the productive inefficiency created by market power.

The graph: At the long-run monopoly equilibrium, the price line intersects ATC somewhere on the downward-sloping portion of ATC — to the left of minimum ATC. The monopolist has not exhausted scale economies.

Why doesn't the monopolist just produce more and reduce unit costs? Because expanding output requires moving down the demand curve — price falls. The revenue loss from the price reduction outweighs the cost saving from lower ATC. The monopolist is maximising profit, not minimising unit cost. These are different objectives.

When entry barriers erode

If barriers weaken (patent expiry, regulatory change, new technology), entry becomes possible. Profits attract entrants. The monopolist faces a choice:

• Limit pricing: Set price below the monopoly optimum — low enough to make entry unprofitable. Sacrifice some short-run profit to protect the long-run position.
• Accommodate entry: Accept market share reduction as entry occurs.
• Invest in barriers: Lobby for regulatory protection, accelerate innovation, lock in customers before entry.

The Tata Nano case referenced in lecture is relevant here: the Singur-to-Sanand factory relocation delayed production by years, reportedly giving competitors time to develop alternatives and raising logistics costs that undermined the Southeast Asia strategy. Operational disruptions can erode time-based barriers that were the monopolist's key advantage.

Part 5: Two Views of Monopoly

The policy view

Most governments treat monopoly as a problem to be managed or corrected. Tools include:

• Antitrust / competition law: preventing monopolisation, blocking mergers that create undue concentration
• Price regulation: capping monopoly prices (utility regulation, price controls)
• Public ownership: nationalising natural monopolies (railways, water, electricity networks)
• Breaking up monopolies: forced divestiture where concentration is deemed excessive

The UK railway privatisation under Thatcher was referenced as an example where the outcome was mixed — privatising a natural monopoly does not automatically create competition; it can create a regulated private monopoly instead.

The business view

From a firm's perspective, some degree of monopoly power is a strategic goal. Being better than competitors means earning margins above competitive returns. The mechanisms:

• Intellectual property: patents, copyrights, trade secrets — legally protected monopoly for a defined period
• Brand and differentiation: reduces demand elasticity, increasing pricing power
• Ecosystem lock-in and switching costs: Apple, Salesforce — customers who have integrated your product face high exit costs
• Exclusive contracts and distribution rights: limiting rivals' access to customers or inputs
• Scale advantages: cost structures that make entry unprofitable at competitive prices

But business moats are typically bounded:

• Time-limited: car design gives a 2–4 year lead before imitation
• Space-limited: Inox popcorn pricing only works inside the cinema
• Regulation-limited: antitrust can challenge practices that create or maintain monopoly power

The professor's framing: governments try to prevent monopoly; firms try to create it. Much of strategy is the business-side of this tension — building pricing power while staying on the right side of competition law.

Concept Connections

• MR < P → this is the root of every monopoly result (output restriction, DWL, markup)
• Lerner Index = −1/ε → connects market power directly to elasticity
• MC₁ = MC₂ → multi-plant version of MR = MC; generalises to any multi-unit allocation
• No-arbitrage → enabling condition for all price discrimination; failing this collapses the scheme
• Excess capacity → monopoly's long-run inefficiency; reappears in monopolistic competition

What the Professor Is Likely to Test

1. Draw and explain monopoly equilibrium Demand/AR, MR (twice the slope for linear demand), MC, MR = MC gives Q, read P from demand. Identify the profit rectangle. State why P > MC.

Trap: drawing P* at the MR = MC intersection rather than reading it from the demand curve.

2. Lerner Index L = (P − MC)/P = −1/ε. Apply to calculate markup given elasticity, or infer elasticity given markup. State that L = 0 in perfect competition and rises with market power.

3. Two-plant monopoly allocation MC₁ = MC₂. Explain why: if unequal, reallocate to the cheaper plant. Then set total output where MR = common MC.

4. Three degrees of price discrimination — distinguish them 1st degree: charge each unit at WTP (consumer surplus = 0). 2nd degree: price varies with quantity. 3rd degree: price varies by identifiable segment (lower price in elastic market).

Common trap: confusing 1st and 3rd degree. Key distinction — 1st degree charges per unit, 3rd degree charges per segment at a uniform price within each segment.

5. No-arbitrage condition Why discrimination fails without it. Give an example of a discrimination scheme and identify what prevents arbitrage. E-books vs physical books is a clean contrast.

6. Why monopoly has excess capacity Expanding output would push price down by more than it saves in unit cost. Monopolist maximises profit, not productive efficiency. Result: Q* < Q_min_ATC.

Quick Reference

Previous: Session 12 — Perfect Competition

Next: Session 14 — Monopsony and Monopolistic Competition

Back to: Microeconomics Master Index

Source: Lecture slides by Prof. Kaushik Bhattacharya, IIM Lucknow (Session 13). Textbook: Pindyck & Rubinfeld, Microeconomics, Global Edition (2017), Ch. 10, Sections 11.1–11.2.