3rd Week

the practice of charging different prices for the same good

1st degree:

• charge each client a different price, i.e. personalized pricing

2nd degree:

• The seller does not know the client, but offers a menu to choose from

3rd:

• different prices for identifiable groups (for social reasons, younf and old pay less for public transport)

It is impossible/ forbidden or burdensome to

• Fake one’s identity

• have someone else buy and then trade

• 
  

• Pricing follows local purchasing power

EU Common Market Law:

• As an EU national or resident you can't be charged a higher price when buying products or services in the EU just because of your nationality or country of residence.

• Applies in particular if there is no difference in delivery cost – such as digital goods

• This does not apply to protected media, such as licensed music, film or books, because licensing agreements differ between countries

Economic theory benchmark: If the seller were to exactly know the preferences and willingness to pay of a client, he or she can

• design the product to fit the client (as much as costs permit)

• try to extract all the surplus!

Good or bad?

• If design is personalized, this increases social welfare – good

• If there is not enough competition, all these gains go to the firm – unfair?

• large amount of data (indivis. preferences, waltch willingness to pay) available

• immediate testing and feedback offers

• individ. com channel: client can not see other offers

Back in 2000, Amazon made a personalized pricing experiment

• customers noticed it -> protests -> Amazon reimbursed consumers, did not try again

In many markets, sellers cannot identify customer characteristics before these buy

• Design a menu of choices, then let consumers pick – by doing so, they reveal private information about preferences, wealth, and willingness to pay

• Increases profits if customers are very different from each other

• Choices differ in number of included items, quality, and price

  -> versioning, with a downgrading of the lower offers

• main challenge in designing a menu is to make sure that each type of consumer picks the option on the menu that is meant for him

• consumers with high valuation for the good have a strong incentive to “lie” and

  pick a cheaper option

-> instead of lowering the price of the high-quality options, the seller downgrades the quality of the cheaper options (Intel chip), due to efficency at the top they rather downgrade the xpensive option than upgrading it

Two-part tariff: fixed fee + constant price per unit

• Utility pricing (water, electricity), Cable TV + subscription, printers + ink

Three-part tariff: fixed fee + bundled units + constant price for additional units. Examples: Telecoms tariffs

Often goods are sold together, even if consumers have different tastes for them: bundling

• Software: office

• media: Magazines, CD’s

• Bertie Bott’s all flavour beans etc.

Bundling overcomes Customer heterogenity:

• Low prices attract many customers, but bring low margins

• High prices have high margins, but only high-value customers buy

• Bundling smoothens the distribution of values of the products on offer, so that a higher “average” price item can be charged

• makes sense when there are enough customers that only want to buy one single item and not the full bundle

• Under mixed bundling, the seller offers

  both the bundle and parts of it

The sum of parts is priced higher than the bundle

• Big Mac Combo Meal: 7.30

• Big Mac + fries + drink: 4.35 + 2.20 + 1

• Search costs are much lower online than offline!

Much quicker and cheaper for shops to change their prices online than offline, thus we should expect shops to:

• Change prices more often

• React faster to changes in prices by competitors

• Consumers only buy at the firms that charge the lowest price -> market settles quickly on one price

• Any firm that has higher cost is forced to leave the market

  -> Nice, implies an efficient market outcome

• Lots of price comparison sites sprung up, as online shopping took off in the early days of the internet, but price dispersion endured

some consumers search, others don’t

• charging a high price only captures a given share of uninformed consumers

• charging a low price foregoes the rents made from uninformed consumers

Amount of competing firms:

• Difference between the two lowest prices (the 'gap') averages 23 per cent when two firms list prices, and

  falls to 3.5 per cent in markets where 17 firms list prices

Obfuscation:

• practices that frustrate consumer search or make it less damaging to firms—resulting in much less price sensitivity on some other products

Minimum retail prices are floors on retail prices imposed by a good’s producer, Idea: prevent sellers from competing down the retail price

• A low retail price implies a low margin for the seller

• This feeds back into downward pressure on the wholesale price

• Thus reduces the producer’s returns

Per-se illegal in the US:

• Considered anti-competitive conduct that coordinates retail prices

Permitted in the EU, in some cases:

• Protection of certain sectors, such as culture – minimum prices for books

-> Prevent destructive price wars and therefore protects profit margins

Big-sellers like Amazon Marketplace have big advantages

• Market power through network effects allows them to charge high prices

• Control of access to the platform disciplines pricing behavior by smaller sellers

• Control of access may imply direct intervention in the pricing of other sellers (next

• As mentioned before, observed transaction data of sellers is a valuable source of information

How do smaller sellers react?

• Align their prices with the platform’s higher price (price umbrella)

• Desist from competing on price, due to fear of retaliation

Platforms prohibit associated sellers from offering their goods elsewhere at a lower price than on the platform -> Competition authorities have acted in many cases

Idea. Transparency can also benefit the sellers, as they can align on standardized higher prices

• Algorithms using machine learning adjust prices repeatedly in competition.

• Over time, they “learn” that charging higher prices together is more profitable than constant price wars.

• When one algorithm lowers prices, others “punish” it by lowering theirs temporarily — then all return to higher prices again.

• This mimics human collusion, but it happens automatically and without agreement.

Implications:

• Results in supracompetitive prices (above normal competition levels).

• Hard for regulators to stop — it’s anti-competitive, but not illegal, since there’s no explicit coordination or intent.

• Exactly what economists feared: algorithms can “collude” on their own.

✅ Advantages:

• Fast & data-driven: react instantly to new market information.

• Automated: can update prices continuously without human effort.

• Customisable: prices can adapt to each product, customer, or situation.

• Predictive power: algorithms learn patterns to forecast demand or competitor behavior.

❌ Disadvantages:

• Excessive personalization: can lead to unfair or discriminatory pricing (different customers see different prices).

• Price spirals: algorithms may rapidly push prices up or down uncontrollably (seen in financial markets or dynamic online stores).

• Unintentional collusion: independent algorithms may learn to coordinate prices — keeping them high without explicit human agreement.

• Monitored competitors’ prices and automatically matched them on Amazon.

• If a competitor’s lower price came from a seller also selling on Amazon, the algorithm forced that seller to lower their Amazon price too → no chance to attract customers on rival sites.

• Raised prices on Amazon and tested whether competitors followed.

  • If competitors matched the higher prices, Nessie kept them high.

  • If competitors didn’t follow, Nessie lowered prices again later.