ECON 2244 Midterm 1

[A1 HIGH] What are the two defining characteristics of a public good?

(1) Non-rivalry in consumption: one person's consumption does not reduce the availability of the good to others. (2) Non-excludability: the good cannot be confined to those who have paid for it. (Source: L6; L9; PQ1)

[A1 HIGH] 'The consumption of one individual does not reduce the availability of the good to others.' Which property is this?

Non-rivalry in consumption. (Source: L6)

[A1 HIGH] 'The good cannot be confined to those who have paid for it.' Which property is this?

Non-excludability. (Source: L6)

[A1 HIGH] T/F: Knowledge is a public good. Electronic devices are public goods as well.

FALSE. Knowledge is a public good, but electronic devices are not: they are rival (one person using a device stops another using that same unit) and excludable (the seller can refuse to sell to anyone who doesn't pay). (Source: L9 (answer B); definitions from L6)

[A1 HIGH] Is knowledge a public good? Explain using both properties.

Yes. Non-rival: information isn't used up when it's consumed (a theorem doesn't wear out; many people can use it at once). Non-excludable: once known, it can't be confined to those who paid (e.g., one book can be lent and read by many). (Source: PQ1; L6; R-LM ch.1–2)

[A1 HIGH] Why must the efficient (competitive) price of a public good such as knowledge be zero?

Demand side: all consumers can use the good at the same time. Supply side: the investment to produce it is made only once, so excluding a user frees up no resources. The marginal cost of one more user is about zero, so the efficient price is zero. (Source: L6)

[A1 HIGH] If the efficient price of knowledge is zero, what problem does that create, and what is the course's solution?

At a price of zero, the creator can't recover the (fixed) cost of producing the knowledge, so there is too little incentive to create it. Solution: intellectual property rights make the owner a (temporary) monopolist. (Source: L6; R-LM ch.1)

[A1 HIGH] What trade-off does intellectual property law strike (Lévêque & Ménière)?

Incentive vs. access, i.e. dynamic vs. static efficiency. Non-excludability calls for protection (so creators get paid); non-rivalry calls for zero protection (to avoid deadweight loss). A temporary, exclusive right is a compromise: users pay during protection, then the work enters the public domain. (Source: R-LM ch.1)

[A1 MEDIUM] Define static efficiency and dynamic efficiency.

Static: resources are allocated to maximize surplus (producer profit plus consumer gain) at a given time. Dynamic: production techniques and goods improve and renew over time through investment in R&D, design and creation. (Source: R-LM ch.1)

[A1 MEDIUM] What alternatives to IP rights exist for producing public goods, and what are the drawbacks of subsidies?

Subsidization/public funding (Samuelson's prescription, e.g., defense, street lighting, public research, NSF/NIH grants), prizes (e.g., Jenner's reward for the smallpox vaccine) and secrecy. Subsidy drawbacks: the taxes that fund them distort other markets, and the government lacks accurate cost/benefit information, so it over- or under-compensates. (Source: R-LM ch.1)

[A1 HIGH] What's wrong with this: 'Because knowledge is non-rival, granting patents is the efficient outcome.'

It reverses the logic. Non-rivalry means the efficient price is zero, so a patent's monopoly price creates deadweight loss. A patent is a second-best fix for the incentive problem that non-excludability causes, not an efficient outcome. (Source: L6; R-LM ch.1)

[A2 HIGH] What right does a patent give its owner?

The right to sue for infringement anyone who makes, uses, sells, offers to sell, imports or offers to import the invention into the country that issued the patent. (Source: L5)

[A2 HIGH] State the four requirements an invention must meet to be patentable (utility patents).

(1) Patentable subject matter, (2) utility/usefulness, (3) novelty, (4) non-obviousness. (Source: L5; L6; L9; PQ2)

[A2 HIGH] Explain each of the four patent requirements in one line.

Subject matter: a new machine, manufactured product, manufacturing process, composition of two or more substances, a method with concrete, tangible results, a business method, a selling technique, etc. Usefulness: offers some positive benefit to society. Novelty: not previously used or described in a single publication. Non-obviousness: differs from the prior art in ways that would not be obvious to someone with ordinary skill in the technology. (Source: L5; L6)

[A2 HIGH] 'Must differ from the prior art in ways that would not have been obvious to somebody with ordinary skill in the technology.' Which requirement?

Non-obviousness. (Source: L5; L6)

[A2 HIGH] 'Must not have been previously used or described in a single publication.' Which requirement?

Novelty. (Source: L5; L6)

[A2 HIGH] 'Must offer some positive benefit to society.' Which requirement?

Usefulness (utility). (Source: L5; L6)

[A2 HIGH] A firm applies to patent a device that is described in full in a 2019 trade-journal article. Which requirement fails?

Novelty: the invention was already described in a single publication. (Source: L5; L6)

[A2 HIGH] An applicant combines two well-known parts in the standard way any engineer in the field would. Which requirement is most at risk?

Non-obviousness: the difference from the prior art would be obvious to someone with ordinary skill. (Source: L5; L6)

[A2 HIGH] How long does a utility patent last?

20 years from the filing date. (Source: L6; PQ4)

[A2 HIGH] What's wrong with this: 'A patent lasts 20 years from the date it is granted.'

Patent duration runs 20 years from the FILING date, not the grant date. (Source: L6)

[A2 HIGH] What is patent breadth, and what are 'claims'?

Breadth: how different another product must be to avoid infringement. Infringement must be established with respect to one of the patent's claims. Claims are the minimal combinations of elements that qualify for protection. (Source: L6)

[A2 HIGH] What is the doctrine of equivalents?

It protects the patent against any product that does the same work in substantially the same way and accomplishes the same result, even if it falls outside the literal claims. Reading example: Ford's iron with under 0.02% magnesium was held equivalent to International Nickel's patented alloy (at least 0.04% magnesium). (Source: L6; R-LM ch.2)

[A2 HIGH] What is the role of disclosure in a patent? Give all four points.

Patented inventions must be disclosed. (1) All parties (rivals and courts) know what is protected. (2) Disclosure facilitates follow-on innovation. (3) Disclosure puts the innovation in the public domain after the patent expires. (4) It creates knowledge externalities (free access to patent databases). (Source: L9; PQ3; R-LM ch.2)

[A2 HIGH] PQ3 asks: 'Is disclosure a goal for granting copyrights?' What answer do your materials support?

No, not in the patent sense. Copyright protects the EXPRESSION of an original work and requires minimal creativity, not novelty. Protection is not granted in exchange for revealing technical know-how, and there is no prior-art search or disclosure of 'how it works' (software source code can stay secret). FLAG: the slides never answer this directly. This is inferred from L6 and R-LM ch.2/4, so check your class notes. (Source: PQ3; inferred from L6 + R-LM)

[A2 MEDIUM] Why might a firm patent rather than rely on trade secrecy (and vice versa)?

Patent: protects even against INDEPENDENT invention and reverse engineering, but requires disclosure and expires. Secrecy (e.g., Coca-Cola's formula): no disclosure and no time limit, but no protection if a rival invents the same thing independently or reverse-engineers it. (Source: R-LM ch.1; L6)

[A2 MEDIUM] Can an AI system be listed as the inventor on a patent?

No. Patent laws recognize only natural persons (individuals) as inventors, and a patent that lists the wrong inventor could be deemed unenforceable. Open question from the slides: who gets credit for an AI invention (the AI, the person behind it, no one)? (Source: L6)

[A2 LOW] Give the WWI examples of governments suspending patent rights in times of national need.

The US military pressured airplane companies into cross-licensing each other; radio patents were suspended; the US seized 5,000 German chemical patents (including aspirin) and licensed them to American manufacturers. (Source: L6)

[A2 LOW] Amazon's 1-Click patent: when was it issued and what did it protect?

Issued Sept. 28, 1999. It protects any e-commerce transaction executed with one click using stored customer credentials (a business-method patent). (Source: L6; R-LM ch.3)

[A3 HIGH] List the four ADVANTAGES of patents.

(1) Concentrates costs among users (e.g., Gameboy). (2) Information is disclosed publicly. (3) Decentralized: gathers diffused ideas with no need to negotiate. (4) Risk is borne by individuals. (Source: L6; L9; PQ5)

[A3 HIGH] List the DISADVANTAGES of patents (including the three 'one-size-fits-all' problems).

(1) Deadweight loss. (2) Duplicate costs (patent race). (3) One size fits all: the reward is not related to the cost of invention; it may be disproportionate to the inventive step; and the rights can't be modified to fit the market structure. (Source: L6; L9; PQ5)

[A3 HIGH] Why does a patent create deadweight loss?

The holder is a monopolist and prices above marginal cost, so consumers whose willingness to pay is above MC but below the price are excluded ('the marginally excluded consumer'). Their lost surplus is the DWL. (Source: L6; R-Ch2)

[A3 HIGH] What are 'duplicate costs' or a 'patent race'?

Because the patent is winner-take-all, many firms invest in the same research. Total R&D spending exceeds what is needed to produce the innovation, so resources are wasted (rent dissipation). (Source: L6; R-LM ch.2)

[A3 HIGH] Explain 'one size fits all' as a patent disadvantage.

Every patent gets the same term and rights whatever its R&D cost or value. Some classes of innovation end up over-rewarded and others under-rewarded relative to their cost (not based on economic reasoning), and the rights can't be adjusted to the market structure. (Source: L6)

[A3 MEDIUM] When is IP the right tool, and when is public sponsorship more appropriate?

IP is useful for innovations with a narrow clientele (e.g., video games): the users pay. For innovations with widely dispersed benefits (e.g., a satellite, weather-forecast sensing stations), public sponsorship is more appropriate. (Source: L6)

[A3 MEDIUM] Explain 'concentrates costs among users' and 'risk is borne by individuals'.

Those who actually buy the innovation (e.g., Gameboy buyers) fund it through the price they pay, not all taxpayers. The inventor, not the government, bears the risk that the R&D fails. (Source: L6 (interpretation of slide bullets))

[A4 HIGH] Patents vs. copyrights: what does each protect?

Patents protect knowledge focused on discovering useful information (novelty). Copyrights protect information goods (e.g., computer software) focused on creativity. Copyright protects EXPRESSION. (Source: L6)

[A4 HIGH] T/F: Information goods, e.g., computer software, are focused on creativity and are protected by patents.

FALSE. They are protected by COPYRIGHTS. (Outside-course context in R-LM ch.3: US offices increasingly grant software patents. On the exam, give the course answer.) (Source: L9 (answer B); L6)

[A4 HIGH] What exclusive rights does copyright give right holders?

The exclusive right to copy, reproduce, distribute or display their works. (Source: L6; L7-C)

[A4 HIGH] What is copyright's subject matter?

'Original works of authorship fixed in any tangible medium of expression': music, literature, graphics, movies, sound recordings, architecture. (Source: L6)

[A4 HIGH] What's wrong with this: 'Like patents, copyright requires the work to be novel.'

Copyright requires only MINIMAL CREATIVITY, not novelty. (Source: L6)

[A4 HIGH] What is copyright breadth? Name the doctrine.

The doctrine of 'comprehensive nonliteral similarity': changing the characters' names or reordering the pieces still infringes. Protection also extends to derivative works (sequels, translations). (Source: L6)

[A4 HIGH] How long does copyright last?

Protection starts when the work is published and continues until 70 years after the author's death. It is not renewable. (Source: L6; PQ4)

[A4 HIGH] T/F: Copyright protection starts when a work is published and continues until 50 years after the author's death.

FALSE: 70 years after the author's death. (R-LM context: the Berne minimum is 50, but in practice it's 70 in the US and Europe. The Sonny Bono Act extended the US term from 50 to 70.) (Source: L9 (answer B); L6; R-LM ch.4)

[A4 HIGH] Copyright ADVANTAGES compared with patents?

The same as patents, except: there is more competition (narrower breadth), and there are fair-use exceptions. (Source: L6)

[A4 HIGH] Copyright DISADVANTAGES compared with patents?

The same as patents, except: there is no problem with duplicate costs or with excluding more valuable ideas, since almost every piece (art, writing, etc.) gets copyright status. (Source: L6)

[A4 MEDIUM] Why doesn't copyright cause a 'race' with duplicate costs the way patents do?

A copyright fences off only an infinitesimal part of the space of creations, and independent creation doesn't infringe. Almost every work gets protected, so there is no winner-take-all prize to race for. (Source: L6; R-LM ch.1)

[A4 MEDIUM] What is fair use (US), and what logic do courts use to apply it?

Use of a copyrighted work without the author's permission for criticism, commentary, teaching or research. Courts apply Coase's transaction-cost logic: fair use applies where high transaction costs would otherwise block the use. Princeton Univ. Press v. Michigan Document Services: NOT fair use, because publishers had permissions departments (low transaction costs). (Source: L6; R-LM ch.4)

[A4 MEDIUM] Monkey selfies: who owns the copyright?

No one. In Dec. 2014 the US Copyright Office said works created by a non-human are not subject to US copyright, so the photos are in the public domain. (Source: L6)

[A4 MEDIUM] Can AI-generated works be copyrighted under US law? Give the course's examples.

US copyright protects 'the fruits of intellectual labor' 'founded in the creative powers of the mind', and the Office registers works 'created by a human being'. Open question: nobody, or the person who made the AI's operation possible? Examples: Google funding AI-written local news; The Next Rembrandt (2016); a Japanese AI novel reaching the 2nd round of a literary prize (2016); DeepMind's music generator. (Source: L6)

[A5 HIGH] What is a trademark?

A name, word, phrase, logo, symbol, image, color, etc. that distinguishes goods and services from those of other entities (® and TM). Purpose: consumer protection, i.e. it indicates the commercial source or origin of products or services. (Source: L7-C; PQ8)

[A5 HIGH] T/F: Trademarks are used to protect producers.

FALSE. Trademarks are for CONSUMER protection: they tell consumers the commercial source/origin of the product. (Source: L9 (answer B); L7-C)

[A5 HIGH] When does a trademark become generic, and what happens then?

When it becomes the generic description of a general class of product or service rather than an indicator of source. If a court rules it has become generic through common use, the registration may be ruled invalid. (Source: L7-C; PQ8)

[A5 HIGH] Give generic-trademark examples from lecture (with owner).

Aspirin (Bayer; generic in the US), Videotape (Ampex), Photoshop (Adobe, used for any digital photo editing), Jet Ski (Kawasaki, any personal watercraft), Heroin (Bayer, trademarked 1898), Thermos (declared generic in the US in '63), ZIP code (registered servicemark that expired), Yo-Yo (Papa's Toy Co.; declared generic in 1965). (Source: L7-C)

[A5 HIGH] What's wrong with this: 'Coke is generic, because people use it to mean any soft drink.'

Coke is STILL a trademark despite that common use. Generic status requires a court ruling that the mark has become the generic name for the class. (Source: L7-C)

[A5 MEDIUM] How did Chrysler try to keep 'Jeep' from becoming generic?

Trademark-awareness advertisements. (Source: L7-C)

[A5 MEDIUM] What happens if a trademark isn't used for a period of time? Who keeps the US database?

Failure to use it results in abandonment of the mark, and any party can re-register it. The USPTO maintains the database of registered US trademarks. (Source: L7-C)

[A5 LOW] Why does the Lévêque & Ménière book leave trademarks out?

Trademarks answer a different economic logic: they signal origin/quality to consumers rather than give incentives to innovate, so different analytical tools apply. (Source: R-LM intro)

[A6 HIGH] What are the two types of creative environments, and when is each reward set?

(1) Knowledge creation addresses a KNOWN need: the incentive can be set ex-ante. (2) The need wasn't identified until someone thought of the idea (the idea is scarce): the reward must be set ex-post. (Source: L7-P)

[A6 HIGH] Targeted prizes: definition (2 features).

Directed at well-known needs that sponsors know about, and posted ex-ante. (Source: L7-P; L9; PQ7)

[A6 HIGH] Blue-sky prizes: definition (2 features).

'The sky is the limit': offered for innovations that are not identified in advance, and tailored ex-post to the value of the innovation. (Source: L7-P; L9; PQ7)

[A6 HIGH] 'Offered for innovations not identified in advance; tailored ex-post to the innovation's value.' Which prize type?

Blue-sky prize. (Source: L7-P)

[A6 HIGH] Kaggle's $1.2m competition to improve the accuracy of Zillow's home-price predictions: targeted or blue-sky?

Targeted: a known need, with the prize posted ex-ante. (Source: L7-P)

[A6 HIGH] A scientist has a great idea that no posted prize covers. Which prize design addresses this gap?

A blue-sky prize (set ex-post according to the innovation's value). Targeted prizes miss ideas that no sponsor anticipated. (Source: L7-P)

[A6 HIGH] ADVANTAGES of (targeted) prizes?

(1) Avoids deadweight loss (if the technology is put in the public domain). (2) Achieves goals. (3) Encourages basic research. (Source: L7-P; L9; PQ6)

[A6 HIGH] DISADVANTAGES of prizes?

(1) Administrative costs. (2) Duplication of costs. (Source: L7-P; L9; PQ6)

[A6 HIGH] What's wrong with this: 'Prizes always eliminate deadweight loss.'

Only if the technology is put in the PUBLIC DOMAIN. If the winner also keeps exclusive rights, monopoly pricing (and DWL) can remain. (Source: L7-P)

[A6 HIGH] Compare patents and prizes on deadweight loss, duplication and administration.

Patents: DWL from monopoly pricing; duplicate costs (race); decentralized, with no administrator needed. Prizes: avoid DWL if the tech goes to the public domain; still have duplicate costs; have administrative costs. (Source: L6; L7-P)

[A6 MEDIUM] What three things does an organization provide when it sets up an ML competition?

(1) Some data, (2) a problem statement, (3) prize money. (Source: L7-P)

[A6 MEDIUM] Why can letting more teams enter a prize competition lower social welfare, even when each team expects a profit?

Each extra team adds its full participation cost but raises the probability that SOMEONE succeeds only a little. Much of a team's private return comes from taking the prize from others (shared prize), not from new value. In the chatbot example, welfare falls from $250 (1 team) to ≈ $166 (2) to ≈ −$139 (3). (Source: L7-P; PS3)

[A7 HIGH] T/F: If a monopolist can implement perfect price discrimination, she produces the efficient level of output.

TRUE. Each consumer pays their reservation price, so the firm sells to everyone with WTP ≥ MC. The marginal consumer has p = MC, there is no deadweight loss, and the firm captures all surplus. (Source: L9 (answer A); L3; R-Ch5; PQ9)

[A7 HIGH] The three properties of perfect (1st-degree) price discrimination (R-Ch5)?

(1) Profits = total net surplus (the firm takes all CS). (2) The marginal consumer's reservation price = MC, so output is efficient. (3) Product-choice and quality decisions are socially efficient, because the firm's objective coincides with society's. (Source: R-Ch5)

[A7 HIGH] T/F: If a monopolist uses 2nd-degree price discrimination (two-part tariffs) and serves all consumers, high-valuation consumers are left with positive surplus.

TRUE. The fixed fee is set to extract the LOW type's surplus (so low types still buy). High types get the same deal but value it more, so they keep positive surplus. (Source: L9 (answer A); PS2; R-Ch5)

[A7 HIGH] What is the motivation for price discrimination?

A monopolist's market power shows up as her ability to appropriate consumer surplus. Price discrimination extracts more of the UNEXPLOITED surplus (remaining CS plus the DWL triangle) to increase profits. (Source: L3; L4; R-Ch5)

[A7 HIGH] What three assumptions are needed for single-unit price discrimination?

(1) The firm has market power. (2) Consumers have different valuations of the good. (3) Resale is not possible (no arbitrage). (Source: L3; R-Ch5 (two necessary conditions: market power and preventing arbitrage))

[A7 HIGH] 1st, 2nd and 3rd-degree price discrimination: what does the firm know in each?

1st (perfect/personalized): knows each consumer's WTP. 2nd: knows consumers differ but can't identify them, so it offers a menu and consumers SELF-SELECT (two-part tariffs, versioning, quantity discounts). 3rd (market segmentation): knows differences ACROSS observable groups but not within a group. (Source: R-Ch5; L3; L4)

[A7 HIGH] 'The firm cannot identify types, so it offers everyone the same menu and lets consumers sort themselves.' Which degree?

Second-degree price discrimination (self-selection). (Source: R-Ch5; L4)

[A7 HIGH] 'Different prices for identifiable groups distinguished by observable characteristics (e.g., student discounts).' Which strategy?

Market segmentation (3rd-degree price discrimination). (Source: L4; L9)

[A7 HIGH] If the market is segmented, why does the monopolist charge a higher price to the group with LOWER price elasticity? (PQ11)

Profit maximization requires MR1 = MR2 = c. Since MR = p(1 − 1/ε), p1(1 − 1/ε1) = p2(1 − 1/ε2) = c. If ε1 > ε2 then p1 < p2. Intuition: demand in the inelastic market responds less to a higher price, so raising its price loses few sales, while raising the price in the elastic market loses many. (Source: L4; L9; R-Ch5; PQ11)

[A7 HIGH] Describe the graph for PQ11 (market segmentation).

Draw two side-by-side panels with a common horizontal MC = c. In each, draw that segment's demand and its MR (twice as steep for linear demand). Find q where MR_i = MC, then read p_i up on demand. The steeper/less elastic segment gets the higher price (R-Ch5 Fig. 5.5). Label p1, p2, q1, q2, MC, MR1, MR2, D1, D2. (Source: R-Ch5 Fig 5.5; PQ11)

[A7 HIGH] What's wrong with this: 'Under segmentation the monopolist charges the more elastic group more, because they buy more.'

Reversed. The MORE elastic group gets the LOWER price (ε1 > ε2 means p1 < p2). (Source: L4; L9)

[A7 HIGH] What are the challenges with perfect price discrimination?

(1) Preventing resale: works best for services or goods that are hard to transfer, or whose value is tied to the individual. (2) Knowing individual reservation prices: almost impossible, so firms look for characteristics correlated with WTP. (Source: L3)

[A7 MEDIUM] How do data and AI make personalized pricing more feasible?

Data on location, past purchases, demographics, browsing history, credit score and friends predict preferences. Analytics plus customer identification, tracking tools and ML allow real-time individual prices and promotions (e.g., electronic shelf labels for dynamic pricing). (Source: L3; L4)

[A7 MEDIUM] What is the 'ethical' discussion of personalized pricing in lecture?

Selling to consumers in the DWL region seems good: more efficient, and those consumers get the product at a lower price. But Safeway using it to cut into consumer surplus seems to harm consumers. The course view: capturing surplus, by anyone, is better than letting it go to waste. (Source: L3)

[A7 MEDIUM] Late-night vending machine: same products, higher prices on the lower shelf. What's going on?

Price discrimination by time and availability. The cheap items sell out during the day; late at night there are few alternatives and WTP is higher, so only the expensive shelf is still stocked. (Source: L3)

[A7 MEDIUM] Give real-world price-discrimination examples from lecture.

Airline tickets (advance purchase, Saturday-night stay); movie tickets (student/senior discounts); restaurant happy hour; cell phones and electricity (off-peak/on-peak); locals' discounts; the Savored app (restaurant discounts by day and time). (Source: L3)

[A7 MEDIUM] What are the two types of arbitrage that can undo price discrimination?

Transfer of commodity: the low-price buyer resells the good (e.g., gray-market Sony electronics). Transfer of demand: a buyer switches to the option meant for another group (e.g., high-end buyers switching to the basic model). (Source: R-Ch5)

[A7 MEDIUM] List firms' strategies for preventing arbitrage.

Warranties (voided outside the intended country); high transaction costs (coupons); contractual remedies (no-resale clauses on student discounts); vertical integration (Alcoa integrating into wire); adulteration (the idea of adding arsenic to industrial methyl methacrylate); legal restrictions. (Source: R-Ch5)

[A7 MEDIUM] Welfare effects of moving from uniform pricing to 3rd-degree price discrimination?

Profits rise. The inelastic group is worse off (its price rises), and the elastic group is better off (its price falls). If total output doesn't increase, total surplus FALLS (units are misallocated across groups). Exception: if the uniform price excluded the elastic group entirely, segmentation is a Pareto improvement. (Source: R-Ch5; PS1)

[A7 HIGH] PS1/Amtrak: total output was the SAME under segmentation and uniform pricing (1,750 VR units; 345 tickets). What does R-Ch5 say about total surplus then?

Total surplus must have DECREASED. Output is the same but no longer allocated efficiently: the last unit in the low-WTP market goes to someone who values it less than the marginal buyer in the other market. (Source: R-Ch5; PS1; L4)

[A7 MEDIUM] Amtrak: why does a uniform price leave money on the table, and why might discrimination still not pay?

The price-sensitive seniors 'drag down' the uniform price, so price discrimination (verified with senior IDs) captures extra surplus from business travelers. But the gain is only $750 ($8,685 vs. $7,935): implementation costs, such as business travelers cheating, could wipe it out. (Source: L4)

[A8 HIGH] Define a 'near monopoly', a dominant firm and fringe firms.

A near monopoly is a firm with a market share below 100% but large enough to dominate the industry (it has considerable market power). The dominant firm is a large, strategic firm. Fringe firms are small firms with little market power (price takers). (Source: L2; R-Ch4)

[A8 HIGH] Lecture examples of near monopolies?

PayPal (online payments); Intel (servers); Google (US web searches and online advertising). (Source: L2)

[A8 HIGH] T/F: A monopolist with a competitive fringe faces the full market demand.

FALSE. It acts like a monopolist on its RESIDUAL demand: Q_D(p) = Q_M(p) − Q_f(p) (market demand minus fringe supply). (Source: L9 (answer B); L2; R-Ch4)

[A8 HIGH] What are the assumptions of the dominant-firm model?

The dominant firm is large and strategic, has LOW cost C_D(Q), and sets price p. Fringe firms are price takers with higher cost; each produces where p = MC_f(Q_i) (supply curve = MC curve). Summed, this gives fringe supply Q_f(p), with dQ_f/dp > 0. (Source: L2)

[A8 HIGH] Write the Lerner index of the dominant firm and define every term.

L_D = (p* − MC_D)/p* = s_D / (ε_s^f·s_f + ε_M). Here s_D = Q_D/Q_M is the dominant firm's share, s_f = Q_f/Q_M is the fringe share, ε_s^f = (dQ_f/dp)(p/Q_f) is the fringe supply elasticity, and ε_M = −(dQ_M/dp)(p/Q_M) is the market demand elasticity. (Source: L2; L9; R-Ch4 eq. 4.7)

[A8 HIGH] What determines the dominant firm's market power? (PQ10) Give the direction of each effect.

From L_D = s_D/(ε_s^f s_f + ε_M): (1) more elastic market demand gives LESS power; (2) more elastic fringe supply gives LESS power; (3) a larger dominant-firm share s_D (smaller fringe share) gives MORE power; (4) a more efficient dominant firm (lower MC than the fringe) gives MORE power. FLAG: the arrows on the L2 'increases as' slide came through garbled, so check it against the formula. (Source: R-Ch4; L2; PQ10)

[A8 HIGH] Why does raising its price reduce the dominant firm's sales, for TWO reasons?

(1) A higher price makes it profitable for the price-taking fringe to expand output (dQ_f/dp > 0). (2) Market quantity demanded falls (dQ_M/dp < 0). So dQ_D/dp = dQ_M/dp − dQ_f/dp. (Source: R-Ch4; L2)

[A8 HIGH] What does the dominant-firm Lerner index reduce to if there is no fringe?

With s_f = 0, s_D = 1 and ε_s^f = 0: L = 1/ε_M, the standard monopoly Lerner condition. (Source: R-Ch4)

[A8 HIGH] What's wrong with this: 'A dominant firm with 80% share prices exactly like a monopolist facing market demand.'

It maximizes over RESIDUAL demand. The fringe's readiness to supply makes the dominant firm's demand more elastic, so its profit-maximizing price is LOWER than the pure monopoly price. (Source: R-Ch4; L2)