Asymmetrical Information

Clearly price discrimination does not require monopoly power. The examples you provided are not monopolies. Companies that price discriminate, if they do it right, make higher profits than they would have made, had they not discriminated.

Monopolies may or may not price discriminate. Their main source of excess profits comes from the fact that they can control output, forcing those who want the product the most to pay more than would be required were there competition for the customers who are not willing to pay top dollar. This competition drives down the price for all consumers, including those who would have been willing to pay more.

Price discrimination seems to occur when companies face high fixed costs and low marginal costs. In this case they need to get high prices from some of their customers to pay off their fixed costs while not throwing away money by disallowing sales to customers with lower demand who could never the less cover marginal costs. They also have to make sure there is not a large arbitrage market from the low to the high demanders (i.e. reimporting drugs from Canada). Obviously doing this is tricky which is why many of these companies are not especially profitable.

The examples you provided are not monopolies.

Tabarrok does not say they are. He only makes reference to monopoly power. When economists refer to monopoly power, they refer to the size and concentration of the firms in the market wherein some firms have enough power to affect price.

Also note it's important to properly define what one means by "market" before one can consider at all the concentration and resulting power of firms in that market. The geographic aspect is probably something you're ignoring. There may be thousands of movie theaters across the US, but there are likely only a few where you live. There's a great more to be said, but pick up an IO text sometime if you're curious.

I did not mean to imply in the last post that market definition and concentration therein should be one's only considerations. It's typically necessary to consider concentration, but it's by no means sufficient. While doing research as an undergrad on the Alcoa antitrust case, I recall one author saying that the behavior within a market is largely dependent upon its players. This is very true.

And Alcoa is a perfect example. Its monopoly control of inputs in no way caused the expected lack of technological development, progress toward efficiency, abusive pricing schemes, etc. In short, while it controlled 90% of what the government considered to be the market, it didn't behave very much like a monopolist.

Maybe this is because aluminum ingot doesn't even constitute a market. There are plenty of reasonable substitutes in other products. We would need to do additional research to compare, say, the cross price elasticities of demand between firms in a broader spectrum. Or maybe Alcoa did have a monopoly. Maybe the individuals controlling the company simply had other goals in mind besides restricting output for immediate gains.

Government intervention is most often argued for on the basis of a supposed market failure. But there are many reasons that one should question whether or not a market has actually failed. For example, Famous Fables of Economics: Myths of Market Failures makes such arguments. It's been some time since I read that, and I can't say that I would suggest it to anyone, but it shows nonetheless that credible arguments against intervention exist. Ok, I really know almost nothing about IO. I'll shut up now. My point is just that the specifics of market power is never a simple consideration.

Can't see why price discrimination needs monopoly power (cf. the health insurance market).

For another example, look at, say, an eyewash stand in your supermarket. Look at the list of active ingredients on the back. You'll find that, say, the Hayfever version will be more expensive than the say, the computer user version, which will be more expensive than the plain bog-standard version, even though they have the same composition.

Price discrimination requires nothing more than the ability (and incentive) to discriminate. The ability stems from the ability to restrict resale, which is completely unrelated to monopoly power. Thus, drug makers will charge more in the US than in Canada so long as they can stop drugs from leaking back to the US from Canada. Airlines discriminate by time and then require a photo ID to stop resale. Taking an IO text at random from my shelf (Carlton and Perloff) they cite three requirements for price discrimination (p. 435): (1) market power, i.e. the ability to charge about short run marginal cost, (which says nothing about profitability in capital intensive industries) (2) ability to infer willingness to pay for various groups and (3) limit of resales

I think you've confused the market power requirement with monopoly power. Airlines charge above marginal cost, but they don't necessarily make enough money to pay back the cost of the planes.

Sorry... that "about" should have been "above."

In the normal model, price descrimination requires monopoly power to prevent newcomers from entering and offering the high price paying consumers a lower price. I'd guess that that doesn't happen in the airline industry because wouldn't generate enough revneue to successfully run the airline.

I think we've learned that price discrimination can exist in a very competitive market that is characterized by low marginal and high fixed costs (to keep new entrants from bidding away the high price payers), and a highly perishable product (to keep low price paying people from reselling the good).

I'm not aware of the health insurance industry engaging in price discrimination. They charge different people different prices because of perceived cost differences, which is an entirely different issue. --sw

You'll find that, say, the Hayfever version will be more expensive than the say, the computer user version, which will be more expensive than the plain bog-standard version, even though they have the same composition.

The goods are not necessarily homogeneous. Major brands have a great deal invested in their names. Therefore there is a higher incentive for quality, and it's reasonable to think that quality may actually be higher. The same is true of a product like Advil. The generic is theoretically of the same composition, but perhaps Advil ensures higher quality. Or perhaps Advil's marketing leads people's beliefs about the success of the product to actually improve its function (a placebo effect).

Good comments Jonathan, and thank you. It's been a while since I've given IO much thought. Regardless of whether or not it's termed monopoly power or market power, the point is that there exist firms which are able to exhibit price setting behavior, for whatever reason (possibly as arbitrary as those in the previous paragraph).

Scott, your point applies readily to the MS discussion one post down. There are reasonably high fixed costs (I suppose) to developing a browser, but it can be distributed for virtually nothing. Market entry is easy, so MS maintains its market share by leaving price at 0.

Gee, haven't thought about this stuff in quite a while. As Scott says, you wouldn't expect to see price discrimination in an open market because your competitors would offer lower prices to the people you want to charge extra to.

It seems to me that there's a time frame argument here. Airlines charge high prices for people who suddenly need to fly to Cleveland right now; since most flights would be booked in advance, there are monopoly aspects to the "market" when it's defined as "flights from Dulles to Cleveland leaving in the next 6 hours".

Do movie theaters price discriminate? The only examples I can think of are very broad: matinee tickets are cheaper, and senior/kiddie/student discounts. No discounts on the popcorn, tho, which is where they make all their money...

I guess that universities price discriminate in that students receive different levels of financial aid? Hard to measure their profits, tho, since they're spent on staff and new buildings. (And how do you count alumni donations?)

Frankly, I think that price discrimination is usually a losing strategy. In most cases, you incur a fairly high administrative cost in maintaining the pricing structure (though in some cases - such as the airline industry - you can maybe offload some of that cost to your retailers - aka the travel agents - if you can trust them to play ball that is). Worse, you piss off your premium customers when they discover the pricing scheme, and that increases the risk you will lose them to a competitor who offers them not just a lower price, but what is percieved as a more fair price.

Those two things are maybe why people think some sort of monopoly is generally required to maintain price discrimination. If you have a monopoly, you can force your retailer to absorb the cost overhead and abide by the pricing scheme, and you can prevent your high-value customers from defecting. Airlines have something of a monopoly, since they've been successful in the past at getting the government to hamstring competitors (they seem to be less sucessfull these days, which is a good thing).

PS I wouldn't call matinee movies price discrimination - more like peak-load pricing. You pay more to see a movie at 8pm at night when everyone else wants to see it, and less at 3pm when the theater is relatively empty and they have "excess capacity" that would otherwise go to waste.

As an economist, I'd like to offer a few observations and clarifications.

1. Monopoly power is just another term for market power. There is no distinction bewteen the two.

2. To have market power, a seller does not need to be a monopoly--any seller who is not a price taker has at least a little market power.

3. Market power is necessary to engage in price discrimination. A price taker could not do so.

4. What IO theory actually predicts is that price discrimination leads to increased profits, other things being equal. Alex Tabarrak appears to have forgotten that the "ceteris paribus" assumption is implicitly part of any prediction made by economic theory.

Therefore, when you look at profits across industries, you can't simply compare profits for price discriminators vs. non-discriminators, because there are many other factors that cause differences in profits.

To give just one example: IO theory also predicts that non-price competition will be more profitable in the long term than price competition, so we would expect that, other things being equal, industries such as airlines that engage in intense price competition will be less profitable.

I wrote:

"You'll find that, say, the Hayfever version will be more expensive than the say, the computer user version, which will be more expensive than the plain bog-standard version, even though they have the same composition."

To which was responded:
"The goods are not necessarily homogeneous. Major brands have a great deal invested in their names. Therefore there is a higher incentive for quality, and it's reasonable to think that quality may actually be higher. "

It may be reasonable, until you look at the back of the packet, which most people don't do - so you have price discrimination on the basis of asymmetrical information. (Believe me, I spend 40 minutes transfixed by the eye-drops section after learning about price discrimination). The marketer is segmenting the market and taking advantage of hayfever sufferers' greater propensity to pay than other purchasers of eyedrops.

hi all,
out of curiosity why shouldn't we model this in an old-fashioned way? after all, there is a species of economic beastie called "monopolistic competition" into which category, this sucker falls. they have pricing power, but they comptete away profitability with other companies who produce similar products with different sounding names. i can draw a piccy if needed...

Comments are Closed.