The Social Welfare of Price Discrimination

In our last video, we saw that price discrimination is good for the monopolist. It increases profits, but what about for society as a whole, does price discrimination increase social welfare? That's the topic of today's talk.

It's complicated, but here's a rule of thumb -- if price discrimination increases output then it's very likely to be beneficial, to increase social welfare. If output, however, does not increase then welfare probably is reduced. Let's give some intuition for when price discrimination increases welfare. Think about our previous example of the pharmaceutical company GSK setting a high drug price in Europe and a lower drug price in Africa. Suppose that GSK were forced to charge only one price. Do you think it would charge closer to the European price of $12.50 per pill or closer to the African price of 50 cents per pill?

What's more likely to happen if GSK is required to set only one price? If they can't price discriminate, GSK very likely will simply abandon the African market where they weren't making that much profit anyway and set a single world price pretty close to the European level. People sometimes think that if only everyone were allowed to import pharmaceuticals to the United States from Canada, Mexico, or Africa where they're cheaper, then we would all enjoy lower prices. Probably not. If smuggling or legal re-importation of pharmaceuticals were to become more common, then pharmaceutical companies would stop price discriminating and set higher prices for everyone.

Who would be made better off by the resulting single price? Well, Europeans are not better off because they're still paying a high price under the single price rule, but Africans are going to be worse off, because they will no longer have the option of buying important drugs at the lower prices. In this case, price discrimination is beneficial because it increases output. It gives some Africans the chance to buy at a lower price when they otherwise would not have had that chance under a no price discrimination rule. For industries with high fixed costs, price discrimination has another benefit -- the extra profits generated by price discrimination mean that it's more profitable for the company to engage in research and development to produce more new drugs for instance.

For example, the extra profits from selling in Africa mean that research and development is more profitable, and that benefits Europeans too. When it comes to new drugs, you might say that misery loves company. That is, the larger the market for a potential drug, the more research and development will be applied. Price discrimination similarly means airlines can offer more flights to more places at better times, and that also helps business people. Even though they're paying the higher prices, they have a better chance at being able to get there at a good time in the first place. When it comes to software, lower prices for the students also is going to help support software R&D. If the students wouldn't buy the software at all at the higher price, well then the price discrimination is a net benefit to pretty much everyone.

More generally, price discrimination can help spread the fixed costs of research and development over a larger population, and that means more innovation which is to virtually everyone's benefit. The ultimate form of price discrimination is when each person is charged his or her maximum willingness to pay. Economists call this “perfect price discrimination.” Under perfect price discrimination, consumers end up with zero consumer surplus. All of the gains from trade go to the monopolist, but the efficient quantity is produced. There's no deadweight loss.

Let's look at this with a diagram. Think of the demand curve as showing the maximum willingness to pay by different individuals to buy a single unit of this good. Here, for example, is Alex's willingness to pay. Here's Tyler's willingness to pay, Robin's, and on, all the way down to Brian's willingness to pay for the good. If the monopolist could charge each and every consumer his or her maximum willingness to pay, the monopolist would walk down the demand curve producing each unit such that the willingness to pay just exceeded the marginal cost.

In other words, the monopolist would produce every unit up until the efficient quantity of output, the same quantity as would be produced by a competitive industry. The difference being that in the competitive industry, the gains would go to the consumers. In the case of perfect price discrimination, all the gains go to the monopolist. This kind of price discrimination requires that the monopolist have a lot of information about each consumer.

Are there examples of this in practice? In fact there are some, and you may be very familiar with one of them. Universities are fabulous price discriminators. They're even better than the airlines, especially because few people realize what is actually going on. Universities give many students financial aid, which is another way of saying that they charge some of their students more than others. Financial aid is a way of doing well while doing good because it's a form of price discrimination. It increases profits for universities.

Moreover, to get the aid, students and their parents must give the university an incredible amount of financial information, including their tax forms, their W2's, information about their bank accounts, the home they own and so on. All of this information means the universities can create many, many different prices in a way that approaches perfect price discrimination. At Williams College for instance, half the students pay full fare, which is about $32,000 a year. The other half gets some form of financial aid, but the amount varies tremendously. Students whose parents have incomes of about $91,000 a year or higher, they pay an average in tuition of about $22,000 a year. While students from very poor families may pay as little as $1,600 a year. That's meaning that one price can be about 20 times higher than the other. That's a lot of price discrimination.

Price discrimination makes a lot of sense for universities because their marginal costs are low while their fixed costs are pretty high. If a professor is teaching Economics 101 anyway, then the marginal cost of putting an extra student in the classroom is pretty close to zero. Even a student who is paying a smaller amount in tuition is probably adding more to profits than to costs. That helps the university cover its fixed costs such as the salaries and the buildings necessary to support the operations of the university.

So again, price discrimination by the universities increases profits, but it also probably increases their output as well. More students attend university than otherwise would be the case. And again, price discrimination also helps to spread the fixed costs around a larger number of customers. For these reasons, price discrimination by universities probably increases social welfare.

That's it for the more obvious forms of price discrimination. In the next talk we'll be looking at some quite common pricing strategies, such as tying and bundling, which also can be understood as more subtle forms of price discrimination.