Kosmas Marinakis, Ph.D. Before we begin Lecture 11 Oligooly art IIΙ microeconomics II first module Test Homework 4 available due next Monday Voice recordings in the lecture are rohibited Disincentive to kee notes Distributed to eole who do not come This is not a distance course Photos or videos are allowed 2013-18 Kosmas Marinakis, HSE MII Lecture 11 2 Cometition in rices Oligooly cometition (1883) So far we assumed that cometition was in quantities and were the choice variables What if firms comete by setting rices instead of quantities? There is a fundamental difference between rice and quantity cometition Any firm has an incentive to want to sell more than rivals Any firm has an incentive to want to sell for less than rivals Assumtions 1. Two firms roduce a homogeneous good 2. There is a ositive marginal cost of roduction 3. Firms choose rices simultaneously 2013-18 Kosmas Marinakis, HSE MII Lecture 11 3 2013-18 Kosmas Marinakis, HSE MII Lecture 11 4 The model equilibrium Two firms with marginal cost is Firms face market demand Who will consumers buy from? since good is homogeneous, consumers buy from cheaest seller Then, the demand for firm 1 will be /2 0 if firms charge same rice, consumers are indifferent who they buy from 2013-18 Kosmas Marinakis, HSE MII Lecture 11 5 What is the NE in the model? that is, the combination, from which no firm has an incentive to deviate alone If you charge your rival will resond with will undercut you and grab your entire market share this way If you charge you will be losing money If you charge your rival will follow suit neither you or your rival have any incentive to deviate is the NE! 2013-18 Kosmas Marinakis, HSE MII Lecture 11 6
The aradox Paradox resolutions In, the incentive to cut rice leads firms to roduce PC outut At NE both firms earn zero rofit,, 2 0 This solution is aradoxical do firms have market ower? The model demonstrates the imortance of the strategic variable rice versus outut What is the source of this aradox? a slight difference in rice changes the market shares dramatically This may not be the case under: Caacity constraints Reeated interaction 2013-18 Kosmas Marinakis, HSE MII Lecture 11 7 2013-18 Kosmas Marinakis, HSE MII Lecture 11 8 Differentiation model We will try to resolve the aradox by lifting the assumtion for roduct homogeneity Market shares are now determined not just by rices, but by differences in design, erformance, or durability of each firm s roduct In markets of differentiated goods it makes sense to comete using rice instead of quantity your customers will not desert you if you increase the rice more than your rival Firms face symmetric demand curves decreasing in but increasing in The effect of own-rice dominates the cross-rice effect Marginal cost for both firms is Firms choose rices simultaneously announced rice is binding for the firm cannot take it back 2013-18 Kosmas Marinakis, HSE MII Lecture 11 9 2013-18 Kosmas Marinakis, HSE MII Lecture 11 10 with differentiation reactions NE in rices Profit for the two firms is Π Π Each firm 1,2will maximize rofit as Π / 0, which yields the reaction curves for each firm 2 1 2 1 2 2 1. Both reaction functions are ositively sloed 2. They both intersect at the ositive quartile, where 2013-18 Kosmas Marinakis, HSE MII Lecture 11 11 2 2 1 equilibrium 2 1 2 2 2013-18 Kosmas Marinakis, HSE MII Lecture 11 12 1
Examle If firms collude Firms face symmetric demand curves 10 2 10 2 Marginal cost for both firms is 1 Reaction functions are calculate as 12 4 1, 3 2 Equilibrium will be 4, 6, Π Π 18 Firms collude setting one common rice The two firm demand curves 10 2 10 2 collase into one demand curve 20 2 10 0.5 With 1, maximization of rofit yields 4.5, 5.5 Π Π 20.25 Firms benefit if they collude 2013-18 Kosmas Marinakis, HSE MII Lecture 11 13 2013-18 Kosmas Marinakis, HSE MII Lecture 11 14 NE in rices Sequential 2 $5.5 Firm 1 s Reaction Curve Collusive Outcome What if firm 1 sets rice first and then firm 2 makes ricing decision? Firm 1 would be at a distinct disadvantage by moving first The firm that moves second has an oortunity to undercut slightly and cature a larger market share $4 Firm 2 s Reaction Curve Nash Equilibrium $4 $5.5 1 2013-18 Kosmas Marinakis, HSE MII Lecture 11 15 2013-18 Kosmas Marinakis, HSE MII Lecture 11 16 model criticisms The rosect of collusion When firms roduce a homogenous good, it is more natural (?) to comete by setting quantities rather than rices When firms set the same rices, what share of total sales will go to each one? it may not be equally divided Collusion imroves rofits for both firms Although collusion is illegal, why don t firms cooerate without exlicitly colluding? that is, set rofit maximizing collusion rice and hoe others follow Collusive rice is never on the otimal resonse curve, thus, collusion is never a NE 2013-18 Kosmas Marinakis, HSE MII Lecture 11 17 2013-18 Kosmas Marinakis, HSE MII Lecture 11 18
Waiting for the rival Collusion Cometition vs. collusion Collusion If you select the collusive rice and then wait for your rival to do the same Your rival most likely will not follow Because has a better resonse than following you Can do better by setting slightly lower rice and steal your market share NE is a non-cooerative equilibrium each firm maximizes rofit, given actions of cometitors In our revious examle with differentiated roducts: The equilibrium was at = 4 and the collusion outcome was at = 5.5 If both charge 4, they make a rofit of 18 each If both charge 5.5 they make a rofit of 20.25 each If you charge 5.5 but your rival charges 4 you make rofit 13.5 and your rival makes 22.5!! Charging 5.5 and waiting leaves you oen to your rival 2013-18 Kosmas Marinakis, HSE MII Lecture 11 19 2013-18 Kosmas Marinakis, HSE MII Lecture 11 20 Collusion sum u Collusion Intension of cometition Collusion will lead to greater joint rofits exlicit or imlicit collusion is ossible Once collusion is established, a strong motive to break it arises there is a significant incentive for cheating by undercutting Collusion is not a NE it is unsustainable In some oligooly markets, ricing behavior in time creates a redictable ricing attern imlicit collusion may occur In other oligooly markets, firms are aggressive and collusion is not easy This creates high tension A change in rice may give the wrong signal to rivals Firms become reluctant to lower the rice even when this is economically necessary In this case, rices tend to be relatively rigid 2013-18 Kosmas Marinakis, HSE MII Lecture 11 21 2013-18 Kosmas Marinakis, HSE MII Lecture 11 22 Price rigidity Demand under rice rigidity Firms have strong desire for stability Managers might be worried that a rice cut may send a signal of rice war to cometitors Even when cost or demand change, managers might be reluctant to adjust rices downwards They give u roer rofit maximization to avoid the risk of causing a rice war This is an one-way behavior, though increasing rice does not carry a risk of starting a rice war cometitors may or may not follow Each firm faces a demand curve kinked at the current revailing rice, The resonse of rivals to a rice change is asymmetric Above, demand is more elastic if the firm increases rice above, other firms may not follow Below, demand is less elastic if the firm decreases rice below, other firms will follow suit With a kinked demand curve, marginal revenue curve is discontinuous 2013-18 Kosmas Marinakis, HSE MII Lecture 11 23 2013-18 Kosmas Marinakis, HSE MII Lecture 11 24
The kinked demand curve Decrease in MC Prevailing rice D q MR 2013-18 Kosmas Marinakis, HSE MII Lecture 11 25 MR MC can change without resulting to rice change yet, MR = MC is still the equilibrium condition Change in MC must be significant to cause 0 is a descrition of rice rigidity does not really exlain oligoolistic ricing 2013-18 Kosmas Marinakis, HSE MII Lecture 11 26 D q The dominant firm model In many markets there is one firm who is the undisutable leader usually because of size or suerior skill The leader regularly signals the rice changes and other firms follow immediately this can be imlicit or exlicit collusion If the leader serves a significant ortion of the market, it may want to act as a dominant firm Set the rice that maximizes its own rofits Fringe firms become followers and serve the residual demand 2013-18 Kosmas Marinakis, HSE MII Lecture 11 27 * The dominant firm s demand curve is The is derived by, as usual Dominant firm maximizes rofit by ricing at Fringe firms rice at yields total quantity q F q D q T q 2013-18 Kosmas Marinakis, HSE MII Lecture 11 28 *Cartels *Cartel ricing A subset of roducers, who roduce the main mass of quantity for the market, exlicitly agree to collude in setting rices and outut The cartel, then, acts as a dominant firm and those who do not join become the fringe firms The fringe firms may benefit, too, from the choices of the cartel if demand is sufficiently inelastic and cartel is enforceable, rices may be well above cometitive levels Members of cartel must take into account the actions of non-members when making ricing decisions Examles of successful cartels OPEC, International Bauxite Association, Mercurio Euroeo Examles of unsuccessful cartels CIPEC (Coer), Tin, Coffee, Tea, Cocoa We will use the dominant firm model to analyze OPEC cartel 2013-18 Kosmas Marinakis, HSE MII Lecture 11 29 2013-18 Kosmas Marinakis, HSE MII Lecture 11 30
OPEC The OPEC back and now Organization of Petroleum Exorting Countries (OPEC) is an intergovernmental organization that was created in 1960 12 members Iraq, Kuwait, Iran, Saudi Arabia, Venezuela, Libya, UAE, Qatar, Algeria, Nigeria, Ecuador, Angola Its mission is to coordinate the olicies of the oil-roducing countries $ * C 1980 2008 2008 2017 $ Eq. without cartelization * C 2013-18 Kosmas Marinakis, HSE MII Lecture 11 31 q F q OPEC Q q OPEC 2013-18 Kosmas Marinakis, HSE MII Lecture 11 32 Q *Cartels 3 conditions for success 1. Robust organization Members should follow cartel s olicy without cheating This is hard because members have different costs, assessments of demand and objectives 2. Potential for market ower Elastic demands offer little room to raise rices If cartelization offers large otential gains, cartel members will have stronger motive to make it work 3. Control of suly The cartel must either control a substantial market share Or, the fringe suly must not be too elastic 2013-18 Kosmas Marinakis, HSE MII Lecture 11 33 Thank you! Kosmas Marinakis www.kmarinakis.org kmarinakis@hse.ru WARNING! This rintout is rovided as a courtesy, so that lecture time can be dedicated to note taking. These slides are not standalone material and should be used strictly as reference, side by side with notes taken in the lecture. Studying solely from the slides is not recommended and in some cases may mislead those who have not attended the relevant lecture. Less than 5% of tasks in tests and exams can be answered from the slides.