This Session Implicit Price Discrimination (continued) 1. Implicit - PDF document

Road Map for Prices and Markets: Pricing under Different Market Structures Monopoly Pricing • Pricing by a monopolist (e.g., Microsoft) • Some pricing fallacies • Not all gains from trade realized or extracted What we did so far: Pricing with Market Power Price Discrimination • More exotic pricing strategies • Explicit market segmentation • Implicit market segmentation What we’re starting today: The polar opposite Competitive Markets (“Commodity Markets”) • Pricing under competition (commodity markets) • Short run and long run decisions • Strategies to survive in a competitive market This Session Implicit Price Discrimination (continued) 1. Implicit Price Discrimination 2. Airline pricing 3. Bundling Next Session Perfect Competition

Bundling • Selling several goods in one bundle ! Hardware and software ! Software suites ! Sports/Concert tickets ! Auto accessories Exercise 6.6: Screening via Bundling Pricing of the polo shirt and the dinner party at the Lebanese week. Highly segmented, with only three types of customers: Valuation Type of Customer Polo Shirt Dinner A $50 $5 B $40 $40 C $5 $50 A customer may buy either the polo shirt or the dinner.

– Benchmark: Perfect Price Discrimination – – No Bundling – – Pure Bundling – – Mixed Bundling – The Genius of Dell??

What is this? Buy One Get One Free (BOGO) " Shoppers don't understand why retailers offer this kind of promotion when it's no better for customers and no more profitable for stores than a half-price sale…. - Washington Post consumer columnist Margaret Webb Pressler Hmmm…it is more profitable than a half-price sale !! Pizza: Valuation for 1 st : $15.01; Valuation for 2 nd : $5.01; MC = $2 BOGO at $20 Profit = 20 – 4 = 16 2 Pizzas for $10 each Profit = 10 – 2 = 8 This Session Implicit Price Discrimination 1. Implicit Price Discrimination 2. Airline pricing 3. Bundling Next Session Perfect Competition

Bottom Line Marketing guru says: � the right product for the right customer at the right price, � Economist sage says: � segment, differentiate, then segment some more!! � 1. Find products where you can effectively prevent high-end consumers from buying the low-end product 2. If you can � t, implicitly segment the market – design a product line through bundling, differentiation, versioning, inter-temporal pricing, damaging that achieves price discrimination Do not ignore the composition of your market! Prices and Markets Session 9 Perfect Competition Prof. Amine Ouazad

Alusaf � s $2 billion question 1994: Alusaf wants to build a primary aluminum smelter " Capacity: 466 ktpy " Located at Richard � s bay, on the east coast of South Africa " Major bet: Capital expenditure close to 60% of Gencor � s net worth!! Feasibility study for this � Hillside � smelter was performed two years earlier " But aluminum prices plummeted to around $1,110 per ton ….as a result of the flow of Russian and other former eastern block countries Given that aluminum prices are at historic lows would you recommend dropping the Alusaf project? This Session: Competitive Supply and Market Price 1. Alusaf’s $2 billion question 2. Perfect Competition 3. Primary Aluminum Market’s Supply 4. Demand Shocks Next Session Short-Run Costs and Prices

Perfect Competition • Homogeneous product that is traded in a well-functioning transparent marketplace bringing together lots of buyers and sellers easily product sold at common “market price” determined by market-clearing condition: ⇒ quantity supplied = quantity demanded • Large number of small firms — no single producer can materially affect market price (at least when industry was on flat portion of supply curve) ⇒ firms act as price takers • Fragmented demand — no single buyer has significant “bargaining power” to affect the price • Industries with these characteristics are close to the conditions of Perfect Competition. Examples include: • mining, such as copper • primary metal fabrication, such as aluminum • agricultural commodities • certain types of commodity semi-conductors Modeling Supply Decisions in Competitive Markets Industry Firm $ $ s (P) mc (q) P = MR d (P) Q q q* Marginal conditions for optimal output (MR = MC): mc ( q ) = P → Supply curve s ( P ) is simply (inverse of) the marginal cost curve.

An Exercise • There are 10 firms on a perfectly competitive market. Firms have the same cost structure. • c(Q) = 100 + 10Q + Q 2 • mc(Q) = • ac(Q) = • Construct the supply curve of each firm, and the supply curve of the market. • A firm’s supply curve s i (P) = • The market’s supply curve s(P) = • Shut down production if P < 80 70 60 50 mc(Q) 40 ac(Q) avc(Q) 30 20 10 0 0 5 10 15 20 25 30 35

The Aggregate Supply Curve $ 15 14 s A (P) 13 s B (P) 12 11 s(P) = s A (P)+ s B (P) 10 9 8 P * 7 6 5 4 3 2 d ( P ) 1 0 0 2 4 6 8 10 12 14 16 18 20 Output Q Q * Elasticity of Market Supply – Measuring responsiveness of supply to changes in price – s % change in Q E = s % change in P dQ P Point Elasticity E s = dP Q

This Session: Competitive Supply and Market Price 1. Alusaf’s $2 billion question 2. Perfect Competition 3. Primary Aluminum Market’s Supply 4. Demand Shocks Next Session Short-Run Costs and Prices Alusaf’s Hillside Project Seems like a simple business… " Competitive Market – identical product; no tech. differences; price- takers " No room for fancy marketing strategies; nimble operational improvements " Profitability boils down to a simple equation price – average cost That cannot be that hard? " Need to understand: • What determines prices? • Which costs are relevant to Alusaf � s decision? • What drives industry dynamics? " For this, we need an economic model of what drives supply decisions and prices

Back to Alusaf: Cost (per ton) Analysis Smelter Sorocaba A Grand Baie Zaporozhye Country Brazil Canada Ukraine Company Other Alcan CIS Capacity (tpy) 122 180 100 Electricity usage (kWh/t) 15769 14215 17454 Electricity price ($/kWh) 0.005 0.005 0.008 Total electricity cost: 85.61 67.48 136.45 Alumina usage (t/t Al) 1.94 1.94 1.94 Alumina price ($/t Alumina) 111.83 179.39 146.58 Total alumina cost: 216.51 347.30 283.78 Other raw materials 156.19 95.66 58.28 Plant power and fuel 6.51 10.00 4.04 Consumables 79.28 42.16 67.31 Maintenance 30.13 52.93 32.57 Labor 150.40 149.85 17.27 Freight 43.43 39.09 48.86 General and administrative 57.66 66.27 72.16 Which costs are variable and which are fixed? Group 6, Section 6

Group 9, Section 7 MC is vertical once capacity is reached Supply of an Incumbent Firm: Constant MC $ Min. AC AC Assuming variable cost includes LR shutdown all but labor, maintenance, plant power, and G&A Min. AVC AVC = MC SR shutdown Max. capacity Short-run: P > MC = AVC then produce at capacity shut down in the short run if P<AVC Long-run: P > MC = AC then produce at capacity shut down in the long run if P<AC

The World’s Supply Curve of Primary Aluminum • Arrange firms by marginal cost in an ascending manner • Table below shows the lowest 4 firms. Firm Capacity Cumulative MC/ volume AVC Sorocaba 122 122 581 Grand Baie 180 302 591 Zaporozhye 100 402 594 Arvida 2 147 549 607

Section 7, Group 9 Primary Aluminum Short Run Supply Curve 2500. 2500.00 00 Market Demand Curve 2000. 2000.00 00 r ton) 1500. 1500.00 00 rice (per Pri 1000.00 1000. 00 500. 500.00 00 0. 0.00 00 122 938.5 1800.5 3125.5 4082.2 5344.2 6537.5 7112.5 8646.5 9547.5 10233.5 11202.8 12065.8 12951.8 13625.3 14222 15375 16206 16830.5 17473 18069 18428 19129 19706 20027 20296.7 20884.7 Ou Output ('000 tons)

How does the model measure up to the data? In 1993, at a price of $1,110 " Actual production = 19,781 " Supply model predicts = 19,412 (cumulative capacity of plants whose P = 1,110 > MC) " Wow!! Anything else?? " Idle capacity of 950 thousand tpy in some European plants; low cost and should be operating at $1,110 " Subtract 950 from 19,412 to get Adjusted Supply = 18,462 Demand = 18,500 Irrational Capacity: Incentives don � t matter! Is this the whole story? It can � t be: Case says that inventories are accumulating fast, → supply > demand Explanation: � Irrational Capacity � which stays up and running regardless of pricing " State-owned capacity that would not have operated under normal market incentives, but operates because decisions are driven by non-profit considerations " Add all state-owned capacity with average variable cost exceeding $1,110

How does the model measure up to the data? Actual demand 18,500 At a price of $1,110 the model predicts 19,412 Very close to actual production 19,781 Deduct 950 to adjust for idled European capacity 18,462 Add back 1,120 irrational capacity 19,582 (difference builds up as inventories) Question: � Given that Aluminum prices are at historical lows, should Alusaf drop the project? � Answer depends on a more fundamental question: � What is the relationship between current prices and long-run prices? � This Session: Competitive Supply and Market Price 1. Perfect Competition 2. Alusaf’s $2 billion question 3. Primary Aluminum Market’s Supply 4. Demand Shocks and Price Effects Are we missing something? Next Session Short-Run Costs and Prices