The highly intelligent and thoughtful Greg Goelzhauser has some very well-put comments on the law and economics of predatory pricing:
Greg Goelzhauser: Professor DeLong's point is one that has led to the virtual nullification of predatory pricing post Matsushita and Brooke. Relying on neoclassical microeconomics, the Chicago School led the charge against predatory pricing. Their suggestion was that successful predatory pricing would be most improbable and that therefore it bordered on irrational for firms to attempt to carry through such a strategy. Many of the arguments put forth by Chicago School theorists were advanced by John McGee in his seminal article examining the alleged predatory practices of Standard Oil: Predatory Price Cutting: The Standard Oil (N.J.) Case, 1 Journal of Law & Economics 137 (1958).
In that article, McGee suggested that there was little evidence pointing to predatory pricing by Standard Oil. He also suggested that such practices would be irrational in the sense that Standard Oil--being a much larger firm than its competitiors--would have to lower prices over a very large market share. Realizing that a predatory pricing scheme could not be enforced ad infinitum, smaller firms were not likely to exit the market. "Ah," you say, "but the smaller firms may run out of money. Then what?" The answer to that has been that so long as the smaller firms were efficient market participants, capital markets would provide funds to ensure continued operation.
But that's not all. Even if firms were to exit the market, the firm engaging in predatory pricing would eventually need to recoup its losses by charging monopoly rents. If this were to occur in a contestable market, new entrants would be lured into the market by the possibility of earning large profits. The result would be increased competition and a lowering of price back near marginal cost. Firms--being rational maximizers of their profits--would understand that recoupment of lost profits would be unlikely and would thus refrain from engaging in predatory pricing. If I've read him correctly, this is Professor DeLong's point. And, I should add, it is one that has had a profound impact on predatory pricing law--to the point that the modern day standard for convicting a firm for predatory pricing is so high that it is almost impossible to meet for an antitrust plaintiff.
Unfortunately, this is not the whole story. As astute readers will have noticed, the traditional account advanced by McGee et al. relies on perfect information. More recent economic accounts of predation taking into account strategic behavior and incomplete information have presented quite a challenge to the traditional account. Theories grounded in the literature on strategic behavior and signaling, for example, suggest that a dominant firm may be able to drive out smaller, less efficient, or more risk averse competitors from the market through predatory pricing. For example, firms possess incomplete information with respect to rivals' cost functions. Theories of cost signaling suggest that firms may be able to mislead competitors into believing that they will not be able to compete with the dominant firm over a long period of time.
Moreover, successful predation combined with incomplete information and risk aversion on behalf of potential competitors may keep the market clear for the dominant firm. If Wal-Mart successfully engages in predation only to later raise gas prices to $12 a gallon, competing firms may enter the market. But, of course, the recoupment period is not likely to include so drastic a rise in prices. More likely is a much smaller rise in prices. Given past successful predation, imperfect information, and risk aversion, potential competitors may choose to invest their capital in another industry. The point is simply that the traditional economic story with respect to predatory pricing--as advanced by Professor DeLong in his post--has been complicated to some extent by models of predation being advanced by theorists taking into account incomplete information and other market mechanisms that may play a role. None of this challenges Professor DeLong's bottome line: "Few industries have cost structures such that attempts at monopolization through 'predatory pricing' harm consumers."
While I think it good to be aware of some of the modern theory with respect to predatory pricing--thus this post--I am not sure what impact the models should have on predatory pricing law. The bottome line is that antitrust is designed to protect consumers and it is not obvious that these models suggest greater harm to consumers than envisioned by classical theory. Moreover, it is doubtful that our evolving understanding of the theoretical possibilities with respect to predatory pricing can overcome the high potential error costs should judges decide to back away from current predatory pricing jurisprudence and attempt more complex analyses of the operational forces and circumstances at issue in cases alleging predation by a dominant firm.
As I understand it, the thrust of the models is that predatory pricing becomes very possible once you have high fixed *sunk* costs that make hit-and-run entry against a monopolist trying to recoup its predatory-price expenditures a risky proposition, and not otherwise. Otherwise, firms can try predatory pricing strategies. And they may succeed at diminishing (static) competition. But they are unlikely to reduce consumer welfare: the gains to the early consumers while predatory pricing was taking place are likely to outweigh the losses to the later consumers suffering under the limit-pricing monopoly.
How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system for Intel-compatible microprocessors is left as an exercise to the reader.
Posted by DeLong at June 9, 2004 12:41 PM | TrackBack | | Other weblogs commenting on this post"If Wal-Mart successfully engages in predation....Given past successful predation, imperfect information, and risk aversion, potential competitors may choose to invest their capital in another industry."
In the case of Wal-Mart the opposite happened. Retailers and wholesalers in competition with Sam Walton realized it was either meet Wal-Mart's price/convenience mix or go out of business. If Wal-Mart thought it was going to be able to raise its prices after driving out competition it was sadly disappointed.
Posted by: Patrick R. Sullivan on June 9, 2004 01:00 PM"How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system for Intel-compatible microprocessors is left as an exercise to the reader."
Well, nothing like a rhetorical question. I have one for you:
How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system that is fully compatible with legacy windows apps is left as an excercise for the author.
Brad: "How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system for Intel-compatible microprocessors is left as an exercise to the reader"
Shame on you Brad, for being a bad Berkeley citiZen. You can get FreeBSD (or GNU/Linux) for free. The price in question is Zero dollars.
mac: "How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system that is fully compatible with legacy windows apps is left as an excercise for the author."
CodeWeavers and Lindows will sell you a full modern OS compatible with legacy Windows apps for about US$50 to US$100. I believe that most ISVs get WindowsXP for about the same price, and it's very hard to get them to switch over to non-MSoft versions of anything. I understand that is mostly because of kickbacks, MSoft's sales team, and predatory practices.
Posted by: Brian on June 9, 2004 01:44 PMI've seen predatory marketing (not just pricing).
I used to live in the DFW area, and American Airlines did this all the time.
If a serious competitor attempted to enter the market, AA would:
* Match prices (even if the competitor were more efficient).
* Bracket flights (if the competitor flew at 10:00am, they'd have flights at 9:45 and 10:15).
These would work, because they owned so many of the gates there, which was a barrier to entry.
Of course, they would also screw around with SABER to screw flight reservations too.
Lovely company.
Posted by: Matthew Saroff on June 9, 2004 01:50 PMI don't think you can measure consumer welfare by averaging out the gains and losses to different individuals. You can only measure the average consumer price or even perhaps the "value" but to reduce "welfare" to the notion of average prices is an abuse of the term.
Maybe that is what economists do nowadays, but then economics ends up being game theory and nothing more.
Posted by: i ain't telling on June 9, 2004 01:55 PMI'm from eastern Canada, where there is a pretty successful example of predatory pricing - Irving Oil. They became the dominant gas station company in the area by moving into each local area, lowering prices long enough to destroy smaller competitors, and then raising prices back to a fairly high level. Some extensive investigations were done about their tactics. Because gasoline outside of urban cores is a geographic oligopoly, there are never really that many competitors in the market. If one of them has much less access to capitol, it's easy to destroy them. By demonstrating your willingness to destroy them, this scares anyone else from trying to compete once you raise prices back up.
This is anything but a "perfect market" situation. And not just because of the limited competition. It strikes me that demonstrating willingness to "punish" competitors is just the sort of thing that can be very effective in game theoretic terms even if it is irrational in conventional static analysis.
The ability to isolate chunks of the market is very important. Temporarily lowering prices in a single town to drive a small competitor out of business is doable. It also leads to people getting the message, as Ian has suggested (one man with a revolver and six bullets can intimidate far more than six people).
If the monopolist has to lower prices over a large fraction of its market to deal with any single small competitor, then it has problems.
The message is probably the most imporant part of the scenario. If sources of capital don't want to fund competitors, or treat it as a high-risk investment, then a monopolist is in a comfortable position.
Posted by: Barry on June 9, 2004 02:36 PMFrom:
http://abcnews.go.com/sections/business/DailyNews/amoco010108.html
It lseems like BP thought it knew how to recover those predatory costs.
‘Shorting’ the Market
In a 1995 e-mail exchange described by the daily newspaper, BP managers Robert Aicher and Linda Adamany discussed “shorting the West Coast market” to achieve “West Coast price uplift scenarios.”
The e-mail describes shipping ANS to FE — Alaska North Slope crude to the Far East — to “leverage up” prices on the West Coast.
“When they say ‘leverage up,’ what does that mean?” asked McAfee. “It means, ‘We’re going to jack up the West Coast price by taking some of our production and selling it at a lower price elsewhere.”
McAfee and veteran engineer Steanson B. Parks, president of a consulting firm in Dallas, both concluded for the FTC that BP had tightened oil supplies to raise the price of millions of barrels of crude oil shipped to refineries in California and Washington state since 1996.
I've heard this situation similarly to Ian's post above.
Posted by: Marisius on June 9, 2004 02:59 PMIn answer to:
How high the price of Windows would have to go in order to induce a profit-seeking competitor to write a full modern operating system for Intel-compatible microprocessors is left as an exercise to the reader.
the answer is: high enoughj to induce large numbers of the firms and individuals using Windows applications to switch to alternative platforms, and hence ISVs to write for them. Given the investment in user training and the relatively small proportion of the cost of a pc represented by the OS, I think the answer is "very high indeed." And that is the level that would be required to induce a *profit-seeking* competitor to consider entering the field.
Of course, this just amounts to stating that the situation is dominated by network externalities, and everybody alreay knew that.
There are already modern non-Windows operating systems for Intel architecture processors; e.g., Linux and the late lamented Bee (now THAT was a modern OS). How much desktop market share do they have?
Posted by: Jonathan Goldberg on June 9, 2004 03:02 PMIt is not just how high the price must go. There must also be some guarantee it will be maintained at that price for a profit seeking competitor to invest the capital needed to produce a rival. If the day you launch your product Microsoft is able to reduce its price to undercut yours and keep it there until you go broke, you will not invest in producing a competing product. No matter what price Microsoft is currently asking.
Posted by: jam on June 9, 2004 03:14 PMBrad, I worked for a now-defunct software company whose product was an Intel-hardware personal computer operating system. We weren't Windows compatible, but (if I may be so immodest) we did a lot of things better than MS's OS did.
What happened? We were completely unable to achieve entry to the marketplace. We sued Microsoft for anticompetitive practices; the suit was settled out of court. The glaring omission in your accounts of monopoly behavior, to me, is the lack of attention paid to barriers to entry.
In the specific case of an operating system, you need at least two things: store shelf space, to pitch your product to folks who already own computers, and relationships with OEMs to sell systems with your OS already installed. This second requirement is far, far more important than the first.
Achieving the first means negotiating with companies like Best Buy to devote precious shelf space to YOUR product rather than, say, the latest Unreal Tournament release, or Tetris clone, or typing tutor, or what have you. There are bunches of software publishers trying to get shelf space for LOTS of titles; you're fighting against their combined weight. This is fairly achievable at a small scale -- getting shelf space at Fry's, say, or at local independent geek stores. Getting shelf space at a large-scale retailer like Best Buy is very difficult.
The second requirement proved even tougher. As far as we could tell, companies like Dell got squeezed by Microsoft: "don't bundle our competitors' products on any of your systems, or we'll triple the price we charge you per copy of Windows that you ship," or something along those lines. I don't recall exactly whether MS was ever found guilty of such anticompetitive practices. But the barrier to entry that this sort of behavior imposes is -huge-.
Furthermore, operating systems are *not* like gasoline: if Chevron gas is expensive, you can start buying Arco. But if Microsoft software is expensive, you can't "just" start buying something else instead. The setup cost can be very large -- say, buying a new Mac to replace your Intel-based PC -- and even if you're talking about moving to another software platform on the same hardware, you have to re-purchase all of your software or their equivalents for your newly-adopted system. How much did you spend on Office? Whoops, it was heavily discounted, since you bought your machine from Dell; you have to buy it anew. Same with Photoshop, etc. The costs of switching are substantial.
Why doesn't someone make a system that is *compatible* with Windows? Threat of lawsuit, mostly, which is itself a barrier to entry -even if the potential competitor is confident they'd win in court-. Microsoft has deep, deep pockets. Can your private startup survive three years in court against the best attorneys that money can buy?
Posted by: ctate on June 9, 2004 03:23 PMGreg seems to be talking about the Milgrom-Roberts AER 198?? paper in which a monopolist can have either low or high costs. Entry is assumed to only occur if the entrant believes that it will face a high-cost rival, post-entry. That is, with full information, high-cost incumbents always face entry while low-cost incumbents never do.
With imperfect information, pooling equilibria are bad for low-cost incumbents because they may face entry when they otherwise would not. Accordingly, under certain conditions, the low-cost incumbent lowers its price to make pooling too costly for high-cost incumbents, leading to the separating equilibrium in which no entry occurs.
That is, at the end of the day in the Milgrom-Roberts model, a low cost incumbent lowers its price and the effect is to deter entry (this sounds a bit like Wal-Mart). So this is not exactly what Bain had in mind (where incumbent's price just below entrants' MC, or "limit pricing," which is not subgame perfect), but it's similar.
Interestingly, in the full information outcome, a low-cost incumbent doesn't have to lower its price in order to deter entry. So the imperfect information case works out better for consumers: they pay the same price after entry is deterred, but pay less beforehand.
[/boring econ lecture]
AB
My understanding is that Standard Oil didnt just use predatory pricing ; it used ever lever it could.
Specifically, it took advantage of a structural feature of the oil industry - what comes out of the well-head is worthless, and difficult to get anywhere.
To be valuable, you need to refine it, and get it to the customer.
Therefore, Standard Oil controlled refineries, and it paid railroads not to carry competitive product, or allow alternate infrastructure to be built (try building a new refinery without using the railroads) and it used predatory pricing only as a last resort.
The net effect was that any potential competitor had to be prepared to fight Standard Oil on each of these levels ... and because Standard Oil had contracts with the railroads and so on, you essentially needed to build an entire new infrastructure.
It's little wonder that no-one wanted to contest Standard Oil's monopoly under these circumstances, and that they didnt use predatory pricing as much as their reputation may suggest.
Simply put, they didnt need to.
Posted by: Ian Whitchurch on June 9, 2004 04:27 PMFollowing on from Ian's point, I think that the debunking of the rationalit of predatory pricing above has three major weknesses in the case of Wal-Mart:
The assumption that Wal-Mart will behave rationally. It has so much money that the penalties for not doing so will be less severe than might otherwise be the case.
The assumption that predatory pricing will be expensive. It need only be done locally as long as competitors are small. That being the case the calculations of a small player are going to be even more conservative than implied above.
Cross subsidy. Wal-Mart makes money from other things too meaning that its ability to bear pain through predatory pricing is reduced. This obviously partly a natural advantage that ought to be reflected in the prices it charges but it must also affect any calculation of the viability of predatory pricing. It clearly afected Microsoft' browser business for example.
On the other hand there must be some value in the stability generated by having a natural winner. If money is confidence then monopolies ought to have some value.
Do the imperfect information calculations bear any resemblence to poker betting? This seems a highly suggestive comparison and one where the relative wealth of the contestants makes a big difference to rational behaviour.
Posted by: Jack on June 9, 2004 11:31 PMLegal/Political costs to competition.
It costs lots of money to remove the pollution from a gas station site to the point where you can build a gas station there again without legal risk from preexisting pollution. This is because gas stations are required to bury their tanks as a crash hazard abatement procedure.
If you were to change the law to allow hardened above ground gas tanks you could reduce the cost of setting up new gas stations to the point that predatory pricing by gas companies (or Walmart) would no longer make sense. It would just be too easy to haul new gas stations into town.
Doesn't first mover advantage factor into this as well? To enter a market dominated by a monopoly with high start up costs and be price competitive could, in many cases, involve losing money continuously until the start-ups cost strucuture made the pricing profitable.
Posted by: Lorenzo on June 10, 2004 07:03 AM"The assumption that predatory pricing will be expensive. It need only be done locally as long as competitors are small. "
Small competitors aren't going to be able to challenge Wal-Mart's ordinary prices, so predatory pricing isn't needed.
Posted by: Patrick R. Sullivan on June 10, 2004 01:21 PMI'm with ctate on this. I ALSO worked for a company that wrote a complete operating system and it WAS Intel-compatible. (BTW When I was the sales guy and we DID get Best-Buy.) As ctate says, the key to sucess is getting OEMs to put your OS on their machines. We had deals with OEMs - they were going to install our OS in a "dual boot" configuration so you could choose which OS to use. ctate writes, "As far as we could tell, companies like Dell got squeezed by Microsoft: "don't bundle our competitors' products on any of your systems, or we'll triple the price we charge you per copy of Windows that you ship," or something along those lines."
Well, that's almost exactly what happened to us, and we aren't guessing - we were told. Big Texas computer manufacturer was told that it would cost them about $50 million in increased Windows license fees each year if they put our OS on their computers. Well, you can't go to your shareholders and explain that you are taking the honorable position but it is costing them $50 million. So that was that.
The same thing was true of competitors to MS Office - which brings in even MORE money for MS. (They have backed off a bit on that, but file format issues keep any competitors from gaining any ground.)
When the administration changed and the anti-trust case was dropped - even though the government had won - that was that, and my company went out of business. I went on to work for a company with a new Office competitor, which for obvious reasons couldn't get funding.
At least in California, Walmart would need to lower its gas prices a lot to grab business away from competitors. How do you explain two gas stations perhaps 50 feet apart with price differentials as high as 15 cents? It’s not just branded versus generic; I’ve seen big differentials between adjacent Shell and Chevron stations. I buy gas at Costco, which is always at least 5 cents cheaper than the (unbranded) station around the corner. Costco is at least 10 to 20 cents cheaper than any of the banded stations near it. I’ve even asked people while they pump their gas: why don’t they go across the street and save 10 cents a gallon? I get various answers, like “this gas is better.” Or “they don’t take my credit card across the street.” You even hear “I never look at the price, I just pull up and pump!” The last statement was from a PhD (Stanford graduate) economist who does energy modeling!
Posted by: A. Zarkov on June 13, 2004 03:20 PMStarbucks uses a preditory model that relies not on pricing, but on principles noted in A. Zarkov's post. Consumers will often just walk into whatever coffee shop is on their side of the street or convenient for whatever reason. Starbucks stores can afford to lose money at one store, knowing that its competition will be losing equally as much and eventually it will have all of that area's coffee shop market when that independent shop goes out of business.
This strategy does not have to be 100% successful to win. The gains from successful ventures are enough to make up for the unsuccessful ones.
So, we've identified numerous predatory strategies, including predatory pricing. What do we do about companies who employ them?
Posted by: Paul Weaver on June 25, 2004 02:41 PMWe really enjoyed your site. We were wondering when you were going to update it again?
free business grants
Posted by: free business grants on July 2, 2004 11:14 AMhttp://sex-with-animals-free-animal-pic.lesbianrmnipples.com/sex-with-animals-free-animal-pic-s-beastility.html sex with animals free animal pic s beastility html http://stories-bestiality-dogfucking-girls-and-animals.lesbianrmnipples.com/stories-bestiality-dogfucking-girls-and-animals-sex.html stories bestiality dogfucking girls and animals sex html http://animal-xxx-xxx-animal-html.lesbianrmnipples.com/animal-xxx/xxx-animal.html animal xxx xxx animal html http://dogsex-movies-index-html.lesbianrmnipples.com/dogsex-movies/index.html dogsex movies index html http://gay-bestiality-sadboy-s-beastiality-sex.lesbianrmnipples.com/gay-bestiality-sadboy-s-beastiality-sex-sea-horse.html gay bestiality sadboy s beastiality sex sea horse html http://animal-xxx-xxx-animal-html.lesbianrmnipples.com/animal-xxx/xxx-animal.html animal xxx xxx animal html http://on-html.lesbianrmnipples.com/on.html on html http://illegal-porn-russian-animal-html.lesbianrmnipples.com/illegal-porn-russian-animal.html illegal porn russian animal html http://girl-index-html.lesbianrmnipples.com/girl/index.html girl index html http://beastuality-pic-index-html.lesbianrmnipples.com/beastuality-pic/index.html beastuality pic index html http://dogsex-free-index-html.lesbianrmnipples.com/dogsex-free/index.html dogsex free index html http://free-animal-sex-html.lesbianrmnipples.com/free-animal-sex.html free animal sex html http://stories-bestiality-beastality-mpeg-sex-animal.lesbianrmnipples.com/stories-bestiality-beastality-mpeg-sex-animal-porno-pictures.html stories bestiality beastality mpeg sex animal porno pictures html http://mpegs-html.lesbianrmnipples.com/mpegs.html mpegs html http://horse-fuck-videos-horse-fuck-videos.lesbianrmnipples.com/horse-fuck-videos/horse-fuck-videos.html horse fuck videos horse fuck videos html http://free-farm-sex-html.lesbianrmnipples.com/free-farm-sex.html free farm sex html
Posted by: Free Farmsex Pics on August 12, 2004 04:15 PM