practice problems for final exam_5_5

Part 1: True/False/Uncertain: Credit depends entirely upon your BRIEF explanation (1 paragraph is usually plenty). (7 points each; 42 points total) 1.1 As we look across coal mines and coal-fired power plants, the theory of specific investment suggests that common ownership (vertical integration) is least important for plants located at the mouth of a coal mine because the efficiencies of co-location ensure high mutual profits regardless of ownership form. True, False, or Uncertain? Explain. False. Mine-mouth power plants have a high degree of specific investment (in Joskow’s paper on contracting in this sector, he argues site, capital, and supply specificity are all highest for these plants. These investments therefore are extremely susceptible to potential “hold-up” problems ex post. Williamson’s theory of investment specificity therefore predicts that vertical integration (or very long-term contracts that mimic common ownership) is most efficient for these types of vertical relationships. 1.2 Search cost models predict more price dispersion across local gasoline service stations in the price of regular gasoline than in the price of a flat tire repair because a tank of gasoline costs more. True, False, or Uncertain? Explain. Likely to be false (credit may be given for alternative well-explained and well-argued answers). Search costs and search benefits are likely to be higher for gasoline than for tire repairs. Gasoline prices are posted in large signs visible from the road, so it is easy to learn what price is charged by each station (low search cost). Consumers generally have considerable flexibility when they purchase gasoline, so they can explore prices at a number of possible stations ((low search cost). Since gasoline is purchased frequently, the benefits of finding a consistently low-price station are realized over multiple purchases (high search benefits, assuming station’s relative prices are similar over time). In contrast, the price of a flat tire repair requires calling or stopping at a station and asking about the price (high search cost), consumers who have a flat tire may need that repair done quickly and in a close-by station (high search cost), and aren’t likely to need this service often (reducing the benefit of search). Applying the intuition of Sorensen’s search cost model of pharmaceutical prices, these factors suggest less price dispersion (relative to mean price) for gasoline than for tire repairs. [Maybe a potential 14.33 project?] 1.3 The recent decline in housing prices across the U.S. is likely to produce significant decreases in the income of the average realtor. True, False, or Uncertain? Explain. False (with caveat). This question asks you to draw on the discussion of the realtor market. If commission rates as a percent of house values are approximately fixed (less true now than in the period analyzed by Hsieh and Moretti, but perhaps still more true than assuming they are independent of house values), lower house prices reduce realtor income per transaction. With a fixed number of realtors, this would reduce income/realtor (perhaps true in the very short-run). But since this is a relatively free entry/exit market, we would expect exit of realtors to the point where expected realtor income will once again equal opportunity wages.

1.4 Firms that expect to compete in prices (Bertrand) have a greater incentive to differentiate their products through advertising than do firms that expect to compete in quantities (Cournot), all else equal. True, False, or Uncertain? Explain. True. Since differentiation softens price competition, increasing profitability, it is likely to be more important in pricing games, where the equilibrium with no differentiation is P=MC. 1.5 In a capacity-constrained pricing game with proportional (random) rationing like CSG market D, you do better when your rival’s price is far below your price than when your rival’s price is just slightly below your price, all else equal. True, False, or Uncertain? Explain. True, if your rival’s capacity is low relative to market demand, and product differentiation is low (as it is in market D). Basically, your residual demand is much greater when your capacity-constrained rival charges a very low price than when that rival just epsilon-undercuts you, increasing your potential profits. Refer to the model of Bertrand competition with homogeneous goods and capacity constraints. Intuition: Since very low prices generate a lot of consumer demand for that product, proportional rationing means that more of the consumers buying from your rival are low willingness- to-pay consumers who wouldn’t buy the product at the higher price. This will leave more of the high value customers in the market after your rival runs out of output, ready to purchase from you (these generate your residual demand curve). If your rival just epsilon-undercuts your price, all the consumers they sell to would have purchased from you. At this price, there also is less excess demand relative to your rival’s fixed capacity. These factors substantially reduce your residual demand, reducing your potential profits. Note that this effect is not as important when product differentiation is high, since a rationed consumer with a high valuation for your rival’s product isn’t likely to have a high valuation for your product. 1.6 Congress recently extended the length of copyright protection by about 20 years. To maximize social welfare (measured by the sum of producer and consumer surplus), the extension should have applied only to work created after the extension, not to existing copyrighted works. True, False, or Uncertain? Explain. True. Copyrights, by creating a legal monopoly, increase the price above marginal cost, inducing deadweight loss (the gain to producers is less than the lost surplus of consumers). But if works were sold at marginal cost in a competitive marketplace, their creators might not be willing to invest their time in creating those works in the first place, since they wouldn’t be able to appropriate the surplus created by the work. We’re willing to sustain the deadweight loss (static inefficiency) to provide incentives for creation of new works (dynamic efficiency). An extension for existing work increases producer profits over the extension period, but those profits are more than offset by a reduction in consumer surplus, if profits and consumer surplus are equally weighted in social welfare. Once works have been created under the old copyright regime, there is no further incentive effect, so the extension increases deadweight loss without any offsetting benefit. 1.7 DuPont’s decision to add substantially to its titanium dioxide capacity in the 1970s was strategic because the scale of the expansion was determined by expected future market growth, not current conditions. True, False, or Uncertain? Explain. False. We call it strategic because the level of DuPont’s investment was determined not only by the direct benefits of the investment on DuPont profits, but also by its intended effect on reducing

• Tying hardware and cement sales • Preventing import and export of cement by controlling ports • Acquiring competition Part 2: Model-based problems 2.1 You’ve just won the 100K competition with a new product idea for DVD “walking tours” of the Boston area, and are ready to start production. You estimate summer weekly demand at: [D1] Q k (P K ) = 250 – 10P K You can manufacture to order, and estimate your production costs to be C M,K (Q K ) = 4Q K If you distribute this product yourself by setting up a pushcart network, your weekly distribution costs are an additional C D, K = 550 + Q k a. (7 points) If you produce and distribute the product yourself (vertical integration), what retail price, P K , should you set, and what quantity do you expect to sell? What are your expected weekly profits in this case? Price will be 15, and quantity 100. Expected weekly profits (don’t forget the fixed cost!) are 450. There are two retailers who could distribute the product for you through their network of stores. Each retailer has distribution costs for a new product of: C R = Q R,K and each charges a non-negotiable weekly “shelf” fee of $100 to cover the fixed cost of stocking a new product in its stores. If only one retailer carries your product, it will be a monopoly distributor. If both retailers carry your product, they will compete in retail prices. Consumers do not perceive any differentiation between the retailers, and will buy the product from the store with the lowest retail price (randomizing across stores if the retail price is the same), according to the market demand you estimated above.

b. (7 points) If you choose to distribute your DVD through a single retailer, what price would the retailer set as a function of your wholesale price P W ? Given this, what is your profit- maximizing P W ? What will be the retail price P K , sales Q K , and your profits? What are retailer profits? Note: you could have solved this using an optimal two-part tariff (P W = 4, Fixed Fee = 1000 - 100 shelf fee paid to retailer= 900), or maximum RPM (P W = 14, maximum P K = 15-- if the courts would recognize maximum RPM as legal in this circumstance, or quantity forcing (P W = 14, minimum Q K = 100) to “force” the vertically- integrated price and quantity outcome at higher profit (900 for you, 100 or 0 for retailer, depending on how you interpreted the shelf fee payment), due to the lower distribution cost. Full credit was given for this if you recognized the opportunity and explained how to achieve it. If you assumed the only vertical contract was a linear wholesale price P W , the solution involved P W = 14, P K = 20, Q K = 50, and profits of 400 for you and 350 (or 250 if the shelf fee just covers cost) for the monopoly retailer. This outcome reflects the cost of double marginalization. c. (7 points) Suppose you pay both retailers the $100 shelf fee to carry your product. What price would the retailers set, as a function of your wholesale price P W ? [Think before writing!] Given this, what is your profit-maximizing P W ? What are the retail price P K , sales Q K , and your profits? What are retailer profits? Which distribution method (a, b, or c) should you choose? This set-up implies undifferentiated Bertrand competition among retailers, so they will set P K equal to their marginal cost (P W + 1). Solution is P W = 14, P K = 15, Q K = 100. Your profit is 800. Choice of structure will depend on whether you found the efficient vertical restraint in (b)--in which case (b) is the best outcome. If you used the linear P W contract in (b), then (c) is the best method (difference is that you pay two $100 shelf fees in c to obtain the no double marginalization outcome. But this is less than your $550 self-distribution cost). Another entrepreneur (L) is thinking of entering this market. Her production and distribution costs are the same as yours. If her DVD L is sold in the same retail outlets as yours, consumers will compare the products and determine them to be perfect substitutes. In this case, they will choose which to purchase solely on the basis of lowest retail price (randomizing between products K and L if the retail prices are identical). Market level demand when both DVDs are sold in the same outlets at a price of P* is [D2] Q(P*) = Q K + Q L = 250 - 10P* d. (7 points) What would be the equilibrium price, sales, and profits of you and your rival if both DVDs are sold through the same retail networks? [Hint: Think about the nature of competition in this market before writing.] Assuming both retail networks sell both DVDs, and compete as in an undifferentiated Bertrand model, we would expect P i = MC i . This implies P WK = P WL = , P K = P L = 5. Note that if you pay $100 shelf fees to each retailer, you lose $200 under this outcome. 2.2 (20 points) bigPHARMA has a patent monopoly on a new hepatitis vaccine. Demand from the US is estimated at: P US = 100 - Q US Demand in sub-Saharan Africa is estimated at

a. (8 points) If there is no way to prevent arbitrage (trans-shipment) so that bigPHARMA must set a single worldwide price, how would you advise the firm to determine the profit-maximizing price to charge for the vaccine? What price should it set, what are its sales of the vaccine in each market, and what is its total profit? The key here is to evaluate two options: set one price, sell to all v. set one price, sell to high value only. Note that with CRS, production will take place in just one facility in either case, so you incur only one fixed cost. In this example, profits are highest by targeting just the US market. b. (8 points) The US requires drugs and vaccines sold in the US to be produced in FDA- certified facilities. Once a facility is certified, all vaccines produced in that facility are certified. Vaccines sold in Africa do not require certification. Certification imposes an additional fixed cost of 100 per certified facility, but shifts demand out in each market according to: P US,C = 110 - Q US,C P A,C = 55 - Q A,C How, if at all, does the certification requirement change bigPHARMA’s optimal production and pricing decisions, and what effect does this have on sales and profits? Certification has two benefits--increases demand for both groups, and now provides the possibility of price discrimination w/o arbitrage--sell the low price uncertified good to Africa, it can’t be shipped to the US since it doesn’t meet FDA standards, and sell the high price certified good to US. If you check the single price (all certified), sell to both option, the single price sell to US only option, and the two price price- discrimination option, profits turn out to be highest under the latter. This requires two fixed costs, since you need the certification/noncertification distinction in output to enforce no-arbitrage. Profits go up (both demand expansion in US and increase from sales to Africa). c. (4 points) The World Health Organization is encouraging developing countries to ensure their medications meet the same quality standards as those imposed in developed economies such as the US and Western Europe. Would it be beneficial to African patients to insist that any vaccine sold in Africa meet the FDA certification standard? No-- if the company can’t price discriminate through the certification/non- certification, it won’t sell to Africa at all. You might have suggested that without certification, the quality might be lower--but we can assume that effect is embedded in the demand (the value of certification is pretty modest--see the shift out by 5 in the African demand curve--which we could interpret to mean that the quality effect w/o certification doesn’t seem huge. This doesn’t seem to be an asymmetric info market where consumers worry that w/o certification, quality is really low).

Production Production Price Price

2.3 Price Discrimination. Suppose that Sanofi expects to be an unregulated monopolist when it introduces its new Zika vaccine. It has identified two types of potential markets: developed countries where the virus has taken hold, such as the US, and emerging market countries such as Brazil (B). To keep it simple, suppose these are the only two markets. Sanofi estimates its marginal cost to be $40 per dose, and the demand in each market to be: Demand in US ࠵? ࠵?࠵? =9−࠵? ࠵?࠵? /40 Demand in Brazil ࠵? ࠵? =6−࠵? ࠵? /40 a. (5 points) If Sanofi can sell the vaccine at different prices in each market, and FDA regulations prevent cross-shipment of vaccines between the two markets, what market(s) will Sanofi serve, what price(s) will it set, and what are its profits? Sanofi will set the monopoly price to both types of markets. So it will sell in both markets, setting P US = $200, and P B = $140. b. (5 points) If Sanofi cannot prevent cross-shipment of vaccines, and so must charge the same price in each market, what market(s) will Sanofi serve, what price(s) will it set, and what are its profits? Sanofi solves max P (P-40)(Q US (P)+ Q B (P)), so it sets P = $170. It earns $1105, which is more than what it would earn selling only in the US. c. (4 points) Suppose there is a second vaccine producer, Me2Drugs, also with marginal costs per dose of $40. Me2Drugs plans to enter only the Brazilian market. If Sanofi and Me2Drugs both serve the Brazil market, compete in prices (undifferentiated Bertrand), and Sanofi can block cross-shipment of vaccines from Brazil to the US, what would be the equilibrium price and quantity in Brazil? What are Sanofi’s profits in each market in this case? Due to undifferentiated price competition, P B = $40. Since Sanofi can block cross-shipment, it charges the monopoly price in the US, P US = $200. d. (4 points). Suppose Sanofi cannot prevent cross-shipment of Sanofi-produced vaccines from Brazil to the US, but can get an injunction for patent infringement against Me2Drugs to prevent its products from entering the US. So, while Me2Drugs vaccines can’t be exported to the US, Sanofi can’t stop an exporter from ordering Sanofi vaccines to be delivered in Brazil and exporting them to the US. In this case, how would Me2Drug’s entry affect Sanofi’s choice of which market(s) to serve? What would be equilibrium prices in each market, and each manufacturer’s profits in this case? If Sanofi continued to sell to Brazil, it would have to sell at the same price in both markets. But in Brazil, the only price at which Me2Drugs doesn’t have a profitable deviation to fully corner the market is P = $40. Clearly Sanofi is better off by selling only to the US.