Saturday, December 08, 2007

(Anti) Competition Chicken Noodle

Suppose you are a chicken noodle addict and chicken noodle is, well, chicken noodle. They are similar product.

Case #1: There are two chicken noodle sellers in your area, Rizal and AP. Each shares half of total sales, or market share.

OK, it seems like a fair competition.

Case #2: Still in your area, but now there are five sellers, --Rizal, AP, Aco, Ujang, and Sjamsu. Rizal's market share is 96 percent, each of remaining seller shares 1 percent of total market sales.

Are you gonna sue me for anti-competition behavior based on such market share indication?

Recall: what matters is how many chicken noodle sellers that an addict like me you now has (five), not the number of sellers in a whole market, nor its market share.

Case #3: Now, Rizal, AP, Aco, Ujang, and Sjamsu decide to set a chicken noodle cartel.

Does the competition vanish? No, they just shift it from the streets in your neighborhood to a table at cafe salemba where they usually meet up.

Case #4: Rizal is spying your house. He wants to steal your BMW. 

It's politics. Economics has no answer. 

18 comments:

  1. I don't understand. On case #3, market entrance would be difficult because of the cartel right? That is why they have to be curtailed.

    ReplyDelete
  2. Is it still about tamasek??. Anyway, I will try to respond the cases.

    1. So, you see fairness just from the market share? What is fairness? Fairness for who ? What about the price?

    Think about the symmetrics equlibria under collusion and under bertrand. They at the end share the same market share, yet the equilibrium price is totally different

    Both charging 1000 USD and then share the market share, while at the same tume charging 100 USD and then share the market share. Which one do u think is fair? Got my point?

    2. NO. as long as you indeed compete..

    3. It is partial collusion. As far as I know, almost all the collusion is a partial collusion. (which not all firms join the collusion). U can name it, eg : elevator and escalator cartel that just punished in EU owns 90 % of the market share, Vutamin cartel, beer cartel in EU, etc.

    U imagine, there is competion between Rizal, AP, Aco, Ujang, Sjamsu and Manager. (6 players)

    and competition between (Rizal, AP, Aco, Ujang, Sjamsu) as one player with manager as another player.

    At the end "real" market structure" is different, isn't it? One is pure oligopoly, while the others is "duopoly"

    so what is the different between these two condition? The price and quantity equiliria.

    Analytical solution indeed show this results.

    Does it matter? Of course. Consumer loss is bigger in partial collusion rather than that of in competition, Hence sosial welfare loss.

    and it does matter for many fellow economist and policy makers.

    After all, we mainly concern about the social welfare, don't we?

    4. Does the question have anything to do with the market structure and collusion?

    It is economics. It is strategic behavior of the firms. It is very economics. Otherwise, u will not see any articles on collusion in many major jounal in economics. U name it, JPE, QJE, Rand JE, RES, R E&Stat, AER, EER, etc.

    Just my two cents

    ReplyDelete
  3. Mova, did I say that Manager, if she really wanted to, can not sell the chicken noodle in my neighborhood without being the cartel member?

    And what do you think the guys do in the cafe salemba? Not only for good coffee, but also to fight for the best price and quantity setting for each of them. Unlike in the streets, it is now more coordinated, but the nature of competition remains.

    Or, if you insist that such talk is a collusive act to deter competitor's entrance, can I say that meatball soup vendors are actually also their competitors?

    Ado,
    #1. Yes, sir, I got it. Me just being satirical. And before putting yourself into technicalities troubles of equilibria, Bertrand model, etc, make sure that artificial market barriers to entry exist in the first place (e.g lobby or bribe to security guard to forbid Manager to operate in my neighborhood) as the (perhaps, the only) base for anti competitive behavior allegation.

    #3. Didn't you just say in #2 that the number of players (either the barristas and one Manager or six players) doesn't say anything about the anti-competition? You won't sue anyone as long as there is a competition?

    #4. Of course it has nothing to do with collusion in economics sense. And that is why economics has no answer for that particular #4. (Hint: replace BMW with something like national integrity or security information)

    ReplyDelete
  4. Case #1. It would be fine for me as a consumer as long as you are not discussing the business with AP. I hope the antitrust authority will be spying both of your communication :-)

    Case #2. No, I am not. I guess your chicken noodle is very yummy.

    Case #3. Naturally there is no incentive for you to set cartel with the other 4 players. You don't need them to set higher price, because you have very large market share! 96%! The case is not relevant.

    Case 4#. It is criminal. :-))

    dendi
    http://gocleancorporate.wordpress.com

    ReplyDelete
  5. Dendi, even if I and AP do conspire against our customers (and you), in my opinion, as long as we do not bribe the security guards for not allowing other sellers to operate in the neighborhood or issuing discriminative licenses, I think you are still perfectly fine, no?

    ReplyDelete
  6. Of course you are not allowed to set the price together with AP as well.

    However, we as costumers could threaten you if you set the price too high. We are going to stop buying chicken noodle and consuming more sausage... :-)

    ReplyDelete
  7. Rizal :

    Sorry for all the jargon :).

    for 3 : Zal, let me ask you one question. Do you think number of firms will affect (eq)price??

    [Note : bertrand paradox came from a highly stylized model. Try to expand this stylized model, ex using capacity constraint or many class of variation, you will get a different result with the prediction by bertrand. Moreover, In real world, I think, none is as simple as that original bertrand model. :))]

    Dendy, I think there is always an incentive to collude :). Charging monopoly price is certainly much better than charging competitive price. It is individual rationality, higher profit is always better :)

    Comment rizal on dendy:

    I agree with you that entry blockage is indeed a problem for competition.

    Yet, collude with part of the firms, (thus form imperfect collusion) is also a problem. (becasue I think (in exception on highly stylized bertrand original model) number of firms indeed matter for price and quantity.

    for, case 4 : in crap movie "minority report", an intension to do crime is a crime it self. :)). But I think, in most law, the intention to steal, is not yet a crime. :) no?

    Anyway, Why noodle?. Do you miss them so much ? :-)).J/K.

    Just my two cents.

    ReplyDelete
  8. To the original post,

    I think the most important aspect that influence whether certain firms configuration is doing anti-competitive behavior or not, is their choice of motive: profit maximization _or_ "being the best of all".

    Profit maximization is the way to go, yay! As long as the market has more money to throw or willing to accept less item/service, competition is stupid, IMHO.

    Sometimes in a business we're in, collusion happen with no explicit mentioning what so ever. It feels like heaven :-). In an already cut-throat competitive market, a chance to start a new collusion usually happen when there is a new radical change of business environment.

    Anyway, that's probably what happen to Telkomsel and Indosat. Maybe they think, yeah let's kind of fight but don't make it too frontal, and for whatever reasons don't disturb the milking in Short Messaging Service. :-)

    And yeah, "competition" shift from neighborhood to negotiation table between colluding parties but the result of such negotiation won't improve customer's welfare.

    ReplyDelete
  9. Ado,
    Imagine payoff matrix 2x2 with the row and column collude and not collude. The case given by Rizal that he has very large market share (96%) and the other 4 players only 4%. You could also imagine payoff for Rizal's profits when they collude and don't collude. I guess Rizal's profits are not too much different between those cases. The small player can not punish him severely because small players have very small market share, and of course their capacity. On the other hand, Rizal can punish the other players severely.

    In other words, Rizal has huge bargaining power, but the others is very tiny one. I think the relevant model is price leadership not collusive behavior.

    Moreover, I think the most potential problem on the case given by Rizal is not collusive behavior, but entry restriction.

    ReplyDelete
  10. Dendy ....

    entry blokage is another thing. Based on the information on rizal case, there is no information what so ever that could suggest us that there is entry deterence.

    Entry is bad,not only for case 3, but also for all case 1-3. (even they compete or symmetrics "50%market share" collusion, blocking entry is still possible and does matter)

    About the matrix, and big market share. If you have different market share in competitive, you could also have the different marke share during collusion (you could arrange that in the collusive agreement), no?

    epsilon increase in price is indeed matter. At tleast in theoretically, we could see that rizal will be better off. How could then he decide not to collude? Individual rationality is not binding..:). You could ask anyone, if they get chance to increase profit by epsilon, will they do that??? :). Most likely they will.

    Thus, in your two by two matrix, collude is indeed a dominant strategy (because it gives more pay offs, doesn't it?). I think, it could also be a perfect equilibria.

    About the punishment, as far as I know, during the punishment period, they will again back to competitive price and quantity. They can just compete each other, thus get the competitive price and quantity. Does the market share and bargaining power matters? Yes, it is indeed matter in determining equilibrium competitive price and quantity.

    yet, we also know that under this condition, collusive profit willbe higher than that of competitive.

    it would be price leadership. Yet, you certainly know that price leadership could happen due to market structure and collusion. Hence, price leadership is not necessarily an outcome of market structure. It could also a form of collusion.

    I think there is some article about this during eighties
    e.g D'Aspremont,Jacquemin and JJ Gabszewicz(Southern economic journal,83), and Rotemberg and Saloner, (Jounal of IO, 1990),etc

    Just my two cents.

    ReplyDelete
  11. Ado,
    In an industry, the number of firms does not affect the equilibrium price. The number of chicken noodle produced does --along with the demand, of course.

    The number of chicken noodle produced depend on the market structure (price taker or price searcher).

    Whether you can limit the number of chicken noodle, hence price (using any kind of techniques you mentioned) depends on whether you can deter new entrants.

    Entry deterrence can be natural (economies of scale or geographical factor), but also artificial (collusion).

    Collusion among sellers can be effective only with the help of external authorities (security guards in my housing complex, or the state).

    And this is, in my opinion, the only basis for anti competition allegation.

    Dendi
    Inter chicken noodle sellers collusion is fine for me, even if they rig the price, as long as there is no artificial entry deterrence.

    Cartel is always prone for breakup for many reasons: high incentive to cheat, availability of substitution (including meatball soup or mutton tongseng).

    Price leadership is impossible. First, monopolists is always subject to supply and demand. Rizal and the gang can not just set the price over coffee and jazz without taking into account market supply and demand.

    Second, chicken noodle price also depends on any other goods price (meatball, tongseng, satay, etc). Rizal et al have no power to meatball sellers.

    Amitz Sekali
    The only chicken sellers motive (and any kind of firms) is maximizing profit. "Being best of all" is profit maximization, in the end is about making profit --which is perfectly okay.

    Competition is always there, even when you are the single chicken noodle seller. You are subject to consumer demand (including their ability to shift to other products) and supply (factor price, etc). You compete not only with chicken noodle sellers, but also with other goods (meatballs, tongseng, and even, if you think about it, iPhone)

    If you can still squeeze the customer money, without reducing their demand, you are not yet at your optimum point. And if I am the shareholder, I am gonna fire you :-)).

    But when you are already there, where your marginal cost (additional cost of producing another bowl of noodle) equal to your marginal revenue (additional revenue for that bowl of noodle), you are price sensitive --you can't sell more without reducing your price.

    The logic of competition and profit making will always be the same whether you fight in the streets market or at cafe salemba table.

    All
    I think, based on limited information I gave you, we can not decide whether there is an anti competition practice in chicken noodle business.

    The alleged collusion, price rigging, etc can only happen if there is artificial market barrier to entry. You should prove this at the very first of your allegation.

    Anything beside that is shaky, and rather than overpaying those lawyers, the chicken noodle sellers can always hire me to defend their case :-)).

    ReplyDelete
  12. Rizal,
    Yes, it is a contestable market... market with zero sunk cost and elasticity of demand is perfectly elastic

    ReplyDelete
  13. Rizal :

    Price do not affected by number of firms??

    By the line of questions, I suspect you will say so..:).

    I think your statement is only valid under standard (highly stylized) bertrand model (no capacity constraint, etc). Otehrwise, it most likely won't happen. Furthermore, under cournot model, we could see that price is affected by number of firm. However, It is possible that i miss lots of literature that support ur argument. Hence, can u give me the literature that support ur argument?

    But, can u tell me your reference of saying so?

    "The number of chicken noodle produced depend on the market structure (price taker or price searcher)"

    not only that, but also consumer. Search cost, taste, etc.

    "Whether you can limit the number of chicken noodle, hence price (using any kind of techniques you mentioned) depends on whether you can deter new entrants"

    There supposed to be a correlation. I don't know the degree of correlation. Yet, you argue this causality. Any reference for this causal relationship??


    "Collusion among sellers can be effective only with the help of external authorities (security guards in my housing complex, or the state)"

    Any reference??

    "And this is, in my opinion, the only basis for anti competition allegation"

    Any empirical or theoretical justification?

    Thanks for the reference anyway.:)

    Just my two cents

    ReplyDelete
  14. Dendi
    Here is the catch-22, will you punish the firms that take a big risk in disbursing high sunk cost in the highly elastic demand market? Those who made it would, naturally, become a monopolist at least for some time. Those who did not would, well, sink.

    Ado
    Reference? That's hard :-). All of my opinions are based on my scattered readings on some micro-econ books. On the very market assertions, I recall some passages from an elementary book by Alchian and Allen, Exchange and Production: Competition, Coordination, and Control.

    I am thinking on why we differ on the number of firms issue. One possible explanation is the assumption in the model that firms make identical choice. Correct me if I am wrong here.

    On correlation issue. It's not the correlation, but rather a feature of oligopoly/monopoly market --in which there is no free entry to the market, no?

    On the role of authority, it's my inference from the idea that cartel is basically very prone to breakup.

    ReplyDelete
  15. No, I won't. If an industry is characterized by high sunk cost but very elastic demand, monopolist would never have large market power. Whenever monopolist increases price, he/she will be punished by costumers. The costumers will quickly stop buying and find another closely substituted products.

    ReplyDelete
  16. Dendy,

    Apakah dalam sebuah struktur pasar monopoly masih terhadap closed substitusi?

    ReplyDelete
  17. Fariz,
    No..., of course in monopoly market structure the product has no close substitutions. I am just responding the case given by Rizal.

    ReplyDelete
  18. Meatball is a different good (but i will need a survey to confirm this). oh no, leave the survey to the economist, hehehe

    ReplyDelete