Price discrimination exists when sales of identical goods or services are transacted at different prices from the same provider. In a theoretical market with perfect information, no transaction costs or prohibition on secondary exchange (or re-selling) to prevent arbitrage, price discrimination can only be a feature of monopoly and oligopoly markets. Otherwise, the moment the seller tries to sell the same good at different prices, the buyer at the lower price can arbitrage by selling to the consumer buying at the higher price but with a tiny discount. However, market frictions in oligopolies such as the airlines and even in fully competitive retail or industrial markets allow for a limited degree of differential pricing to different consumers. Price discrimination also occurs when it costs more to supply one customer than it does another, and yet the supplier charges both the same price.
Although the term "discrimination" has negative (e.g. racist, sexist) connotations in common usage, the meaning of the word "discrimination" (from discriminatio, "a distinction") is neutral. "Price discrimination" is a technical term meaning only differentiation in price by customer, and is not intended as an accusation of illegal or unethical behavior.
The effects of price discrimination on social efficiency are unclear; typically such behavior leads to lower prices for some consumers and higher prices for others. Output can be expanded when price discrimination is very efficient, but output can also decline when discrimination is more effective at extracting surplus from high-valued users than expanding sales to low valued users. Even if output remains constant, price discrimination can reduce efficiency by misallocating output among consumers.
Price discrimination requires market segmentation and some means to discourage discount customers from becoming resellers and, by extension, competitors. This usually entails using one or more means of preventing any resale, keeping the different price groups separate, making price comparisons difficult, or restricting pricing information. The boundary set up by the marketer to keep segments separate are referred to as a rate fence. Price discrimination is thus very common in services, where resale is not possible; an example is student discounts at museums.
Price discrimination can also be seen where the requirement that goods be identical is relaxed. For example, so-called "premium products" (including relatively simple products, such as capuccino compared to regular coffee) have a price differential that is not explained by the cost of production. Some economists have argued that this is a form of price discrimination exercised by providing a means for consumers to reveal their willingness to pay.
An alternative way to understand First Degree Price Discrimination is as follows: This type of price discrimination is primarily theoretical because it requires the seller of a good or service to know the absolute maximum price that every consumer is willing to pay. As above, it is true that consumers have different price elasticities, but the seller is not concerned with such. The seller is concerned with the maximum willingness to pay (or reservation price) of each customer. By knowing the reservation price, the seller is able to absorb the entire market surplus, thus taking all consumer surplus from the consumer and transforming it into revenues. From a social welfare perspective, first degree price discrimination is not undesirable. That is, the market is still entirely efficient and there is no deadweight loss to society. However, it is the complete opposite of a perfectly competitive market. In a perfectly competitive market, the consumers receive the bulk of surplus. In a market with first degree price discrimination, the seller(s) capture all surplus. Efficiency is unchanged but the wealth is transferred. This type of market does not much exist in reality, hence it is primarily theoretical. Examples of where this might be observed are in markets where consumers bid for tenders, though still, in this case, the practice of collusive tendering undermines efficiency.
Additionally to second degree price discrimination, sellers are not able to differentiate between different types of consumers. Thus, the suppliers will provide incentives for the consumers to differentiate themselves according to preference. As above, quantity "discounts", or non-linear pricing, is a means by which suppliers use consumer preference to distinguish classes of consumers. This allows the supplier to set different prices to the different groups and capture a larger portion of the total market surplus.
Additionally to third degree price discrimination, the supplier(s) of a market where this type of discrimination is exhibited are capable of differentiating between consumer classes. Examples of this differentiation are student or senior "discounts". For example, a student or a senior consumer will have a different willingness to pay than an average consumer, where the reservation price is presumably lower because of budget constraints. Thus, the supplier sets a lower price for that consumer because the student or senior has a more elastic price elasticity of demand (see the discussion of price elasticity of demand as it applies to revenues from the first degree price discrimination, above). The supplier is once again capable of capturing more market surplus than would be possible without price discrimination.
Note that it is not always advantageous to the company to price discriminate even if it is possible, especially for second and third degree discrimination. In some circumstances, the demands of different classes of consumers will encourage suppliers to simply ignore one/some class(es) and target entirely to the other(s). Whether it is profitable to price discriminate is determined by the specifics of a particular market.
The first/second/third degree taxonomy of price discrimination is due to Pigou (Economics of Welfare, 4th edition, 1932). See, e.g., modern taxonomy of price discrimination However, these categories are not mutually exclusive or exhaustive. Ivan Png (Managerial Economics, 2nd edition, 2002) suggests an alternative taxonomy:
The hierarchy -- complete/direct/indirect -- is in decreasing order of
Complete price discrimination is most profitable, and requires the seller to have the most information about buyers. Indirect segmentation is least profitable, and requires the seller to have the least information about buyers.
The purpose of price discrimination is generally to capture the market's consumer surplus. This surplus arises because, in a market with a single clearing price, some customers (the very low price elasticity segment) would have been prepared to pay more than the single market price. Price discrimination transfers some of this surplus from the consumer to the producer/marketer. Strictly, a consumer surplus need not exist, for example where price discrimination is necessary merely to pay the costs of production. An example is a high-speed internet connection shared by two consumers in a single building; if one is willing to pay less than half the cost, and the other willing to make up the rest but not to pay the entire cost, then price discrimination is necessary for the purchase to take place.
It can be proved mathematically that a firm facing a downward sloping demand curve that is convex to the origin will always obtain higher revenues under price discrimination than under a single price strategy. This can also be shown diagramatically.
In the top diagram, a single price (P) is available to all customers. The amount of revenue is represented by area P, A,Q, O. The consumer surplus is the area above line segment P, A but below the demand curve (D).
With price discrimination, (the bottom diagram), the demand curve is divided into two segments (D1 and D2). A higher price (P1) is charged to the low elasticity segment, and a lower price (P2) is charged to the high elasticity segment. The total revenue from the first segment is equal to the area P1,B, Q1,O. The total revenue from the second segment is equal to the area E, C,Q2,Q1. The sum of these areas will always be greater than the area without discrimination assuming the demand curve resembles a rectangular hyperbola with unitary elasticity. The more prices that are introduced, the greater the sum of the revenue areas, and the more of the consumer surplus is captured by the producer.
Note that the above requires both first and second degree price discrimination: the right segment corresponds partly to different people than the left segment, partly to the same people, willing to buy more if the product is cheaper.
It is very useful for the price discriminator to determine the optimum prices in each market segment. This is done in the next diagram where each segment is considered as a separate market with its own demand curve. As usual, the profit maximizing output (Qt) is determined by the intersection of the marginal cost curve (MC) with the marginal revenue curve for the total market (MRt).
The firm decides what amount of the total output to sell in each market by looking at the intersection of marginal cost with marginal revenue (profit maximisation). This output is then divided between the two markets, at the equilibrium marginal revenue level. Therefore, the optimum outputs are Qa and Qb. From the demand curve in each market we can determine the profit maximizing prices of Pa and Pb.
It is also important to note that the marginal revenue in both markets at the optimal output levels must be equal, otherwise the firm could profit from transferring output over to whichever market is offering higher marginal revenue.
Notice however that in this example "the seat" is not really always the same product. That is, the business person who purchases the $300 ticket may be willing to do so in return for a seat on a high-demand morning flight, for full refundability if the ticket is not used, and for the ability to upgrade to first class if space is available for a nominal fee. On the same flight are price-sensitive passengers who are not willing to pay $300, but who are willing to fly on a lower-demand flight (say one leaving an hour earlier), or via a connection city (not a non-stop flight), and who are willing to forego refundability.
On the other hand, an airline may also apply differential pricing to "the same seat" over time, e.g. by discounting the price for an early or late booking (without changing any other fare condition). This could present an arbitrage opportunity in the absence of any restriction on reselling. However, passenger name changes are typically prevented or financially penalised by contract.
Since airlines often fly multi-leg flights, and since no-show rates vary by segment, competition for the seat has to take in the spatial dynamics of the product. Someone trying to fly A-B is competing with people trying to fly A-C through city B on the same aircraft. This is one reason airlines use yield management technology to determine how many seats to allot for A-B passengers, B-C passengers, and A-B-C passengers, at their varying fares and with varying demands and no-show rates.
With the rise of the Internet and the growth of low fare airlines, airfare pricing transparency has become far more pronounced. Passengers discovered it is quite easy to compare fares across different flights or different airlines. This helped put pressure on airlines to lower fares. Meanwhile, in the recession following the September 11, 2001, attacks on the U.S., business travelers and corporate buyers made it clear to airlines that they were not going to be buying air travel at rates high enough to subsidize lower fares for non-business travelers. This prediction has come true, as vast numbers of business travelers are buying airfares only in economy class for business travel.
There are sometimes group discounts on rail tickets and passes. This may be in view of the alternative of going by car together.
This effect can lead to (seemingly) perverse incentives for the producer. If, for example, potential business class customers will pay a large price differential only if economy class seats are uncomfortable while economy class customers are more sensitive to price than comfort, airlines may have substantial incentives to purposely make economy seating uncomfortable. In the example of coffee, a restaurant may gain more economic profit by making poor quality regular coffee--more profit is gained from up-selling to premium customers than is lost from customers who refuse to purchase inexpensive but poor quality coffee. In such cases, the net social utility should also account for the "lost" utility to consumers of the regular product, although determining the magnitude of this foregone utility may not be feasible.
Middle- and lower-income students may be offerred discounts in the form of tuition waivers, scholarships, work-study programs that pay partly in free course hours, and government guaranteed loans.
Another example is textbooks. Publishers such as Prentice Hall and Pearson have low cost editions of textbooks for countries such as India. The textbooks are often printed on cheaper paper, are paperbacks and priced at 15-20% of the dollar price. This pricing has largely eliminated the practice of photo copying these books.
Governments can also use tax policy to increase prices in order to limit consumption and increase tax revenue, such as automobile prices, which incur a 100% tax in many countries with low automobile population.
Even online sales for non material goods, which do not have to be shipped, may change according to the geographic location of the buyer. A song in Apple's itunes costs 79 pence (1.49 USD) for Britons but only 99 US-cents for Americans. Britons pay 49% more than Americans for the same song. These differences may arise because of changes in exchange rates that occur much more frequently than changes in prices, or they may arise because the license-holders (in this case, record companies) are enforcing their existing pricing policy on new licensees or intermediaries.
"Universal" pricing is the opposite of price discrimination — one price is offered for the good or service. This is usually preferred by consumers over tiered pricing. For example, the European Union is currently making efforts to set a single-price protocol for automobile sales.