Industrial Organization - riassunti, Sintesi del corso di Economia Politica

Scarica Industrial Organization - riassunti e più Sintesi del corso in PDF di Economia Politica solo su Docsity! INDUSTRIAL ORGANIZATION Industrial organization studies how markets work and operate, in particular: - How firm obtain and exploit market power (= the ability to charge a price above cost of production) - The focus is on strategic interaction The notion of market power is strictly connected to fims’ strategies. C1: BASIC MICROECONOMICS Demand & Consumer Surplus • The demand function represents the individual WTP for different amount of a given product: o wtp decreases with the amount consumed o the demand decreases with the price 𝜕𝑞(𝑝) 𝜕𝑝 < 0 • The consumer surplus (CS) is the area below the demand curve (D), above the price of good; • The demand elasticity (h) is the percent variation of quantity demanded divided by the percent variation in price; o The demand elasticity is equal to the reciprocal of mark-up = − 𝜕𝑞 𝜕𝑝 𝑝 𝑞 Costs The cost function consists in the total cost of inputs the firm need to pay to (efficiently) produce output q. 1. Fixed cost (FC): the cost that does not depend on q 2. Variable cost (VC): the cost that change with q 3. Total cost (TC): FC+VC 4. Average cost (AC): TC/q 5. Marginal cost (MC): the cost of one additional unite of output MC is the appropriate cost concepts to decide how much to produce (MC < p) AC is the appropriate cost concept to decide whether produce or not (p>AC) Opportunity Cost of using a resource = foregone benefit from not using the resource in its best alternative use; Sunk Cost = the cost of a specific asset with no alternative use; Economies of Scale and Economies of Scope • Economies of Scale = when the average cost of product declines with the output produced outup = AC • Economies of Scope = when the cost of producing outputs q1 and q2 together is lower than of producing them separately C(q!, q") < C(q!, 0) + C(0, q") Monopoly • Def: there is a monopoly when only one firm produces in a well-defined market • Monopoly and Perfect Competition are the opposite • Monopoly price is always Pm > MC The monopolist maximizes its profit when MR = MC 𝑅 = 𝑃(𝑄)𝑄 , 𝑀𝑅 = 𝑃 + #$ #% 𝑄 The social cost of the monopolist (DWL): = the socially efficient quantity is where MC=P=PPC (perfect competition) but the monopolist sets a price PM > PPC , and this entails a deadweight loss (DWP). Perfect Competition There are five assumptions: 1. Atomicity = there are many suppliers 2. Product homogeneity = all firms sell the same product 3. Perfect information = all agents (firms and consumers) know the price set by all firms 4. Equal access = all firms have access to all production technologies 5. Free entry = every firm may freely entry or exit the market in the long run Firm in perfect competition maximize its profit when P = MC: - If a firm sets a higher price, no one buys from it (homogeneity) - If a firm undercuts, it servers all the market unless the other firms make the same - The unique equilibrium in perfect competition is P=MC. There is only one equilibrium, and it is also efficient. In perfect competition there are two implications: 1. Firms are price taker 2. Firms do not earn extra profits in the long run o In the short run (fixed number of firms), firms may have positive profits (if AC<MC) o in the long run other firms enter the market (free entry) up to the number where profit p = 0 NON LINEAR PRICE (NLP) – TWO PART TARIFF STRATEGIES In many industries, consumers pay a price f to access the service plus a price P for each unit consumed: (Total Price) T = f + qP (Price per unit) T q = f q + P This is a form of price discrimination based on quantity: the more you buy, the less you pay each unit. We are assuming that all consumers have the same demand (homogeneous consumers). In this case we have a the follow profit function: 𝜋(𝑃, 𝑓) = 𝜋(𝑃) + 𝐶𝑆(𝑃) Two part tariff strategies o The firm reduces P as much as possible to stimulate the demand 𝑷 = 𝒄 • The firm sets the fixed fee f as high as possible, at the maximum level compatible with the consumers partecipation: 𝒇 = 𝑪𝑺(𝑷) • Fims maximize its profit became: 𝝅 = 𝑨 + 𝑩+ 𝑪 NOTE: with non linear tariff, a profit maximizing firm is socially efficient. Hetereogeneus consumers and multiple tariffs When we have heterogeneous consumers, the rule does not induce self-selection. The optimal menu of tariffs must satisfy two constraints: 1. Incentive compatibility = each individual must prefer the plan (tariff) specifically designed for him 2. Participation constraint = each individual must be induce to consume The equilibrium with heterogeneous consumers is: • Type 1 à 𝑓! = 𝐶𝑆!(𝑃!) , 𝑃! > 𝑐 • Type 2 à 𝑓! < 𝑓" < 𝐶𝑆"(𝑃") , 𝑃" = 𝑐 The firm offer the two plans to everybody (no direct discrimination), then consumers choose their preferred plan. 2. Firm may design their prices to inducing buyer to self-select There are some situations where the seller knows that the potential consumers is divided into two groups, but it cannot identify which group each consumer belongs to. The seller/firms design their price to inducing buyer to self-select among different offerings. Now we’ll study two different methods to induce self-selection: Versioning and Bundling. VERSIONING Versioning is one of the strategies that a firm can do, a type of price discrimination where the seller does not know how the preferences are distributed. Version induce self-selection by adjusting the design of the product and its price to influence the sorting. The general rule to induce self-selection: • Set the price of the low-type version equal to the low-type customers wtp • Set the price of the high-type version such that high-type customers are indifferent between the two version Effect of versioning: 1. It allows to charge a higher price to high type customers (+) 2. Cannibalization (-) 3. It allows the firm to expand sales (+) BUNDLING Bundling is a special form of versioning in which two or more products are offered as a package. There are two type of strategies in bundling: o Pure bundling = the seller sells only the bundler, no separate products o Mixed bundling = choice of bundler and separates Bundling can be used as a mean to reduce dispersion across individuals’ preferences. By bundling the seller is able to eliminate heterogeneity in consumers’ wtp. C4: Oligopoly Competition Def: a market is said to be an oligopoly when there are few firms operating; if there are only two firms, we are talking about duopoly. o In PC and Monopoly = firms do not care about competitors; o In Oligopoly = firms know that their strategies/ decisions do affect rivals’ strategies (strategic interdependence) Two model that analyse strategic interdependence: • Bertrand = firm set the price and let the market set the quantity • Cournot = firm set the quantity and let the market set the price Bertrand Assumptions: 1) Product homogeneity 2) Perfect information 3) A single firm may eventually serve the whole market 4) Firms compete choosing The Nash Equilibrium (NE) for this game is when both firms do not have unilateral incentive to deviate from their price strategy, which is: 𝑷𝒂 = 𝑷𝒃 = 𝑴𝑪 The different with Perfect Competition is strategic interactions: firm A’s optimal function is function of the price charged by firm B. • Reaction function = is a function which gives the best price for each firm given any price set by the rival 𝑷𝒂∗ (𝑷𝒃) • Nash Equilibrium is the reaction function interaction: 𝑷𝒂∗ (𝑷𝒃) = 𝑷𝒃∗ (𝑷𝒂) • Price equals marginal cost 𝑃 = 𝑀𝐶 • Both firms get zero profit 𝜋 = 0 Relaxing assumption 3 Firm have capacity constraint (k) and cannot always cover the whole market. Let’s suppose that both firms have capacity constraint k and that at a price P=Pa=Pb with P > MC both firms sell at capacity k. In this case no one have the incentive to undercut since it cannot sell more then its capacity k. There is an equilibrium characterized by P > MC and with positive profit for both firms; Þ When total industry capacity is low in relation to market demand, then the equilibrium price are greater than MC: 𝑃 > 𝑀𝐶 Cournot Firms simultaneously choose the quantities and let the market sets the price. For each pair of choices (q1,q2), equilibrium prices will be: 𝑃! = 𝑃" = 𝑃(𝑞! + 𝑞") How to find the equilibrium 𝑃(𝑄) = 𝑎 − 𝑏(𝑄) => 𝑃(𝑄) = 𝑎 − 𝑏(𝑞! + 𝑞") 𝐶(𝑄) = 𝑐𝑞 • Each firm sets the quantity to maximize its profit: max𝜋 = 𝑞![𝑃(𝑞! + 𝑞") + 𝑐(𝑞!)] 𝜕𝜋 𝜕𝑞! = 0 • Solve for q1 and then do the same for q2 and we’ll obtain the two reaction function: 𝒒𝟏∗ (𝒒𝟐) = 𝒂 − 𝒄 𝟐𝒃 − 𝒒𝟐 𝟐 𝒒𝟐∗ (𝒒𝟏) = 𝒂 − 𝒄 𝟐𝒃 − 𝒒𝟏 𝟐 • The equilibrium is obtained by the reaction function intersection: ` 𝒒𝟏∗ (𝒒𝟐) = 𝒂 − 𝒄 𝟐𝒃 − 𝒒𝟐 𝟐 𝒒𝟐∗ (𝒒𝟏) = 𝒂 − 𝒄 𝟐𝒃 − 𝒒𝟏 𝟐 => 𝒒𝟏∗ = 𝒒𝟐∗ = 𝒂 − 𝒄 𝟑𝒃 𝝅 = 𝟏 𝒃 c 𝒂 − 𝒄 𝟑 d 𝟐 > 𝟎 Stackelberg (a Cournot oligopoly with leader and follower) When one firm decides the quantity to produce before the other, we’ll have: - Firm 1 is the leader = it sets q1 before firm 2 - Firm 2 is the follower = it observes q1 than decides its output q2 Leader acts rationally: it is able to anticipate firm’s 2 decision on output because it knows RF(2) when it sets its output. 𝑹𝑭(𝟐): 𝒒𝟐∗(𝒒𝟏) = 𝒂 − 𝒄 𝟐𝒃 − 𝒒𝟏 𝟐 6) Cross ownerships-cross shareholdings - Increase the possibility of rivals knowing each other - A’s management can be informed on rivals’ deviation from the collusive agreement 7) Buyers’ concentration The greater the degree of buyer’s concentration, the stronger their market power, the less likely sellers are able to collude 8) Multimarket contract when firms meet the same rivals in multiple markets. It may facilitate collusion to the extent that it reduces firms asymmetry and/or it increases the frequency of contracts between them. Market Transparency 1) Demand fluctuation are observable A firm receiving unexpectedly low demand understands whether it is the result of a secret price cut of a member of the cartel (hence it may retaliate) or it is the signal of an effective reduction in the market demand. 2) Observability of rivals’ strategies As above, it increases cartels’ stability. There are two ways: price transparency or exchange of information. C6: Vertical Relations Very often firms do not sell only to final consumers, but also to other firms. With vertical relations we mean a relation between two firms in sequence along the value chain. In a vertical relation we can individuate: • Upstream = is the firm selling its output to another firm • Downstream = is the firm buying the input Relationship between firms ¹ Relationship between a firm and its final consumer o A manufacturer who sells through a retailer does not control many dimensions of consumer’s demand o Retailers compete with each other, consumers do not o Retailers are relatively few compared to the number of final consumers, so they have more market power than consumers Double Marginalization and two-part tariff Consider an Upstream firm, Producer (P) and a downstream firm, Retailer (R); Both retailer and producer are monopolist. - P sells to R at unitary price w (wholesale price) à w =marginal cost for retailer - R sells the product to final consumers at price p à p = marginal cost for seller - Let c the cost of production of the unit for the manufacturer - Market demand D(p) = 1-p Game solution § P know that R will offer a quantity for which his MR=MC 𝑅+ = 𝑝(𝑞)𝑞 q𝑀𝑅+ = 𝑝(𝑞) + 𝑝,(𝑞)𝑞 𝑀𝐶+ = 𝑤 𝒘(𝒒) = 𝒑(𝒒) + 𝒑,(𝒒)𝒒 à Producer demand (w(q)) is equal to MR of the retailer § Also the producer will act like monopolist: 𝑅$ = 𝑤(𝑞)𝑞 = 𝑝(𝑞) + 𝑝,(𝑞)𝑞 𝑀𝑅- = 𝑤,(𝑞)𝑞 + 𝑤(𝑞) 𝑀𝐶- = 𝑐 𝒘,(𝒒)𝒒 + 𝒘(𝒒) = 𝒄 Þ Double marginalization implies a loss of profits: the sum of R and P’s profits is lower than vertical integrated profit Þ There is nothing worse than a chain of monopolists: 𝜋+ + 𝜋- < 𝜋. VERTICAL CONSTRAINT = method that the Producer can adopt to reduces the inefficiencies od double marginalization. The producer can increase his profits, trying to control the final price. Producer and Retailer may sign a franchising contract, based on the so called “franchise fee” (two-part tariff). Franchising Arrangement where one party (franchiser) grants another party (franchisee) the right to use its trademark or its trade-name as well as certain business systems and processes, to produce market its good or service according to certain specifications. The franchisee usually pays a flat franchise fee plus a per unit price (non linea price): 𝑇 = 𝑓 + 𝑤𝑞 𝑤 = 𝑐 𝑓 = 𝜋. o Is eliminated double marginalization to induce the increase of quantity offered by the retailer o Retailer acts like a monopolist and obtains gross profit equal to 𝜋. o Franchisee fee equal to 𝜋., extract the profit produced by the Retailer ® Upstream gets gross profit equal to zero, but it uses f to extract the gross profit of the downstream firm Competition between retailers The producer sells his products to more than one retailers. There is competition between downstreams. In case of perfect competition (or Bertrand) the retailers offer the same price, equal to marginal cost. o The producer acts like a integrated monopolist and sets a price 𝑤 = 𝑝.: v 𝑝+ = 𝑤 𝑤 = 𝑝. => 𝑝+ = 𝑝. o The solution is identical to the case with two-part tariff above: - Producer gets maximum profit (as vertical integration) - Retailer gets zero profit Investment (horizontal) externalities In many cases sales effort is very important and needs specific investments by retailers. But competition among retailers may retain them from investing (horizontal externalities). So often no one is willing to make them. There is a trade-off between competition and free riding. The producer can reduce this phenomenon through: 1) RPM (resale price maintenance) = practice whereby the producer imposes a minimum price to all retailers 2) Exclusive territories = producer gives each retailers the exclusive right to sell its product in a given area Merger and acquisition A radical strategy to solve the problem with Inter-firm vertical relations is mergers and acquisition. Advantages ü Avoid double marginalization or free riding ü Full control of the demand in the downstream market ü Reduces transaction costs ü Reduces competition in the market power Disadvantages û Direct cost of merger û Increase business complexity for management û More difficult to monitor the dealers’ efforts (sforzi dei commercianti) û Antitrust concerns C8: Mergers and acquisitions A radical strategy to eliminate a competitor => buy it. Effects of a merger on PRICE o It increases price (because the market becomes less competitive and more concentrated) o It allows companies to reduce costs : with economies of scale merger may imply increase production efficiency due to various types of synergies Effects of a merger on the OTHER FIRMS in the market o It reduces the competitive of the market (reduces the number of firms ¯n) o It increases the strength/power of competitors because the merged companies are much stronger than individually Merger wave Mergers seem to happen in rushes (in fretta) or “waves” rather than gradually over time. We can individuate both exogenous both endogenous events that ignite (accendono) a merge. • Exogenous event: - Market liberalization - Development of a new production technology - Radical change in market functioning • Endogenous event: Often when firms merge, it becomes convenient for the others to merge as well Public Policy toward mergers A merger needs to be approved by antitrust authority. Merger members must comply with antitrust condition. The antitrust needs to evaluate the effects of a merger: - Unilateral : merger Þ ­concentration Þ ­ P - Unilateral : merger Þ ¯ cost of synergies Þ ¯ P - Collusive effect : fewer active firm Þ collusion is more likely to occur (The last effect is strictly affected by the distribution of market share.) Killer acquisition Incumbent firms may acquire innovative targets solely to discontinue the target’s innovation projects and pre-empt future competition. o The products of the acquired firm are uncharged: the acquisition is motivated by R&D activities o These are anti-competitive behaviour that are not regulated by antitrust C9: Market Power & Market Concentration Market power: is define as the ability to impose a price above marginal cost (P>MC). In perfect competition firms do not have market power; but in the real word (with capacity constraint, differential products,.. ) each firm has a certain degree of market power. • Lerner Index (𝑳𝒊) : market power is usually measured with Lerner Index 𝐿0 = 𝑃0 −𝑀𝐶0 𝑃0 • Market power at the industry level 𝐿 =y𝑠0𝐿0 =y𝑠0 𝑃0 −𝑀𝐶0 𝑃000 Þ In a symmetric oligopoly (if the market shares are equal): 𝑠0 = 1 𝑛 𝐿 = 𝐿0 Market concentration E two most used measures: • Concentration index (𝑪𝒎): the sum of the market shares of the largest m firms 𝐶. =y𝑠0 . 03! • Herfindahl Hirshman index (HHI): the sum of the square of the market shares of all the active firms 𝐻𝐻𝐼 =y𝑠0" 4 03! ð When the firms are much asymmetric, the HHI could be inaccurate! HHI in Cournot oligopoly It is possible to demonstrate that in a Cournot Oligopoly, there is a strictly relationship between Market power and Market Concentration (HHI). Maximizing the profit we will obtain that: 𝐿0 = 𝑠0 𝜀 => 𝑳 =y𝒔𝒊𝑳𝒊 = 𝒊 y 𝒔𝒊𝟐 𝜺 𝒊 = 𝑯𝑯𝑰 𝜺 o The Lerner Index (Li) of the single firm is directly proportional to its market share ↑ 𝒔𝒊 = ↑ 𝑳𝒊 o The market power at the industry level (L) is directly proportional to the market concentration (HHI) ↑ 𝑯𝑯𝑰 = ↑ 𝑳 C10: Product Differentiation Firms often sell products that differ in their quality or in other characteristics. We can individuate two types of differentiation: 1. Vertical product differentiation => when products differ in quantity - Consumers agree on what is better - Different consumers have different valuation for higher quality 2. Horizontal product differentiation => when products differ in other dimension rather than quality Why firm differentiate their products? When firms produce homogeneous products, competition drives profits down because with perfect competition P=MC and profits are equal to zero. ð Firms need to differentiate their products to soften competition and to increase their market power (P > MC) Hotelling Model Assumption: - Each firm is located at the extreme of a unit length segment - Consumers are uniformly distributed along the segment - Transportation costs equal to t - The market is fully covered The utility of the consumers located in X, who buys at price P: o If he buys from Firm 1 à 𝑈 − 𝑡𝑥 − 𝑝! o If he buys from Firm 2 à 𝑈 − 𝑡(1 − 𝑥) − 𝑝" The indifferent consumer (X*): 𝑈 − 𝑡𝑥∗ − 𝑝! = 𝑈 − 𝑡(1 − 𝑥∗) − 𝑝" § Those to the left of X* will buy from Firm 1 𝐷!(𝑝!, 𝑝") = 1 2 + 𝑝" − 𝑝! 2𝑡 § Those to the right of X* will buy from Firm 2 𝐷"(𝑝!, 𝑝") = 1 2 + 𝑝! − 𝑝" 2𝑡 Maximizing profits we can obtain the two reaction function: 𝑅𝐹(1) = 𝑝!(𝑝") = 𝑡 + 𝑐 + 𝑝" 2 𝑅𝐹(2) = 𝑝"(𝑝!) = 𝑡 + 𝑐 + 𝑝! 2 Then we can individuate the Nash Equilibrium: ` 𝑝!(𝑝") = 𝑡 + 𝑐 + 𝑝" 2 𝑝"(𝑝!) = 𝑡 + 𝑐 + 𝑝! 2 => 𝒑 = 𝒕 + 𝒄 , 𝝅 = 𝒕 𝟐 ð Abbiamo dimostrato che anche la sola differenziazione di prodotto è sufficiente per superare il paradosso di Bertrand e permettere alle imprese di ottenere profitti positivi (P>c , p > 0) C12: R&D There are two types of innovation: • Process innovation = aimed to creating more efficient production technologies • Product innovation = aimed to developing new products Reasons of investment in R&D Schumpeterian view o Large incumbents have more incentives to invest in R&D o Incumbents are better positioned to innovate than smaller firm o Smaller firms often lack incumbent’s: - Knowledge of market - Experience - Access to funding - Distribution network Arrow view o Smaller firms/ entrants have a greater incentive to innovate in competitive markets - The competitive firm gets zero profits before innovation, but it serves the whole market if it innovates - Monopolistic firm makes money even before innovation These two views are not necessarily alterative, they can coexist: • Big firm have widely possibility of innovation • Small firm have strong incentives to innovate ð Incumbents have more incentives than entrants to invest in GRADUAL INNOVATION ð If there is UNCERTAINITY with respect to the threat of entry and/or there are DRASTIC INNOVATION, outsiders may have more incentive than incumbent to invest in R&D Gradual innovation = innovation that does not displace the existing product (if E enters the market, M is still active) Drastic innovation = with this innovation the existing product becomes obsolete (if E enters the market, M is forced to exit the market) Public Policy Government interventions to encourage investments in R&D: § Direct intervention - Public universities and research centers - Subsidies to companies that invest in R&D § Indirect intervention - Patents - Policies in support of R&D agreements Patent A set of exclusive right granted to an inventor for limited period in exchange for the public disclosure of the innovation. Patent’s characteristics: - Patent length : determined by the law - Patent breadth: It defines the technological territory protected by the patent, the area in the technological space within which competitors can not offer rival innovations without infringing the patent. Determined case by case the patent office. ð Narrow patent (small breadth) : does not provide strong protection to the innovator (an imitation may not infringe the patent) ð Broad patent (high breadth): it effectively protects from competitors imitations the patented Innovation = Patent should be last long but allow for further innovation! Effect of reducing Breadth o Small reduction of monopolist’s profit (loss A but gains C) o Strong increase of social welfare (increase B+C) Þ This is a firs-order welfare effect Effect of reducing Length o The monopolist’s loss from shorter protection is proportional to what society has to gain Þ Negligible effect Patent Races Races between firms to reach innovation first. Firms compete head-to-head in order to obtain the innovation. C13: Network Goods A product exhibits network effects if the utility of the users increases with: • The number of other users of that product • The number of products compatible with ð A product exhibiting network effects is called a network good The model The typical utility function of consumer i for a network good j: 𝑼𝒊𝒋 = 𝒂𝒊 + 𝒇𝒊(𝒏𝒋𝒆) 𝒂𝒊 = stand-alone benefit of the product 𝒇𝒊#𝒏𝒋𝒆% = 𝒗𝒏𝒋𝒆 𝒏𝒋𝒆 = is the expected number of users for product j The users must attribute a utility to the good depending on size of the network; there are two approaches: 1. Fulfilled expectation - consumers based their purchasing decision on their expectations about future network size - their expectations turn out to be correct => 𝒏𝒋𝒆 = 𝒏∗ 2. Myopic expectation Consumers based their purchasing decision on the actual network size Model-Consumers are Heterogenous: Mass one of consumers (q) is uniformly distributed on [0,1] 𝑼(𝜽) = 𝒂 + 𝜽𝒗𝒏𝒆 𝑝 = 𝑎 + 𝑣𝑛 − 𝑣𝑛" Affinché si inneschino le esternalità di rete è necessario il raggiungimento della massa critica di utenti adottanti, oltre la quale per il singolo utente il non appartenere al network diventa uno svantaggio rilevante. Una volta raggiunta la massa critica, la posizione di dominio del mercato è virtuosamente assicurata. Strategies to exploit critical mass: - Consumer first money later - Price below cost and early stages - Freemium - Subsidize early and influential users The Equilibrium 𝑛!(𝑝) is unstable: this dimension of the network represent the critical mass of consumers. o If it is reached, network growth exponentially o If it is not reached, the network empties Add the cross-network externalities: 𝒚𝟏(𝒑𝟏, 𝒑𝟐) = 𝟏 − 𝒑𝟏 − 𝜽𝟐𝟏(𝟏 − 𝒑𝟐) 𝟏 − 𝜽𝟏𝟐𝜽𝟐𝟏 , 𝒚𝟐(𝒑𝟏, 𝒑𝟐) = 𝟏 − 𝒑𝟐 − 𝜽𝟏𝟐(𝟏 − 𝒑𝟏) 𝟏 − 𝜽𝟏𝟐𝜽𝟐𝟏 o Standard negative relationship price/quantity: ∂y! ∂p! < 0 o Cross market effect: by reducing p2, y2 increases and this induces and increase in y1: ∂y! ∂p" < 0 Then if we maximize we can find the optimal prices. Typically the strategy leads to very asymmetric price: § The side paying a low price is said “loss leader side” => users who produce greater indirect network effect (make the other group earn more) § The side paying a high price is said “profit making side” => users that earn the most from the presence of the other category of users Non price strategies o Number of sides in a platform Pro of more sides: ü Larger indirect network effects and larger scale ü Diversified sources of revenue Cons of more sides: û Complexity and conflicts of interests between sides û Difficult in price setting o Design of the platform The platform can choose various functionalities and features to influence access to the platform a and interaction among users. o Governance rules Rulers that users must respect; the platform can regulate users’ behaviours on platform C15: Reputation in online market Online marketplaces differ from online retailers because of some reasons: - Facilitate trade between anonymous buyers and sellers - Do not directly control key variables (price, quality,..) - Host heterogeneous sellers Reputation Game To alleviate opportunism and asymmetric information we need to link past behaviour to future outcomes: - If the seller is loyal, a reputation for loyalty is built (= buyers will buy from him) - If the seller is dishonest, a reputation for dishonesty is built (= buyers will not buy from him) Ratings and Reviews o Reviews are useful especially for experience goods o Reviews can be important to mitigate or remove asymmetric information problems Cabral and Hortcasu It study the effect of reputation and negative feedback in sales: - Early negative review reduces sales growth and attract more negative reviews - Sellers are more likely to exit the market if they it have a negative reputation Vana and Lambrecht It study the effect of reviews on consumers: - Review matter most when there is limited information available on the product page - Individual reviews with high star-rating are particularly effective when the average rating of the product is relatively low Cabral and xu It study reputation and incentives to engage in price war during the pandemic. Sellers with better reputation are less likely to engage in price war. Biased Reviews Reviews can: o Be noiosy = when review are supposed to capture a quality dimension but reviewers react by commenting based on their tasted o Tend to over-estimated positive experience o Be face Moreover: - Unsatisfied buyers tend not to leave negative comments - Buyers can be afraid of retaliation Characteristics of reputation game: - One shot game - Buyer can trust seller or not trust him - Seller can be loyal or unfair Rationality and trust/loyal: o A rational seller should abuse trust o A rational buyer should not trust He et al Study on fake reviews on social o A wide array of products purchase fake review – including product with many reviews and high average ratings. o The fake reviews are allow because in the short run, platforms may benefit from allowing fake review if these increase revenue by generating sales or allowing for higher prices. Moreover fake reviews also be an effective way for seller to solve the “sold-start” problem and establish a good reputation. Solution to falsification Marketplaces measure quality and supplement feedback with governance (certification). There are both: - External certification - Internal data-driven certification Effects of certification: o Increased efforts of non-certificated sellers for products quality o Increased price for certificated seller