Please enable JavaScript.
Coggle requires JavaScript to display documents.
Digital Economics - Coggle Diagram
Digital Economics
Platform Business Model and Pricing
Characteristics of the Platform Ecosystem
Definitions
There is no consensus about the definition of multi-sided market. Nonetheless, the practical identification is consistent with the ideoa of
two or more groups of agents who need each other in some way and who rely on platforms to intermediate transactions between them
- Ganuza, J.J. and G. Llober 2018
There are many definitions of multi-sided platforms. ANd many of them use terms platform and market interchangebly. Nonetheless, almost all of them emphasize the
role of technology enabler ( the platform) to mediate between the transactions of two or more sides"
- Rysman, M 2009
A platform is an entity that brings together economic agents and actively
manages network effects
between them
Belleflamme, P. and M. Peitz 2021
Characteristics
Online marketplaces - facilitate transactions
Important in two-sided markets
Firstcomer - strong network effect -> barriers to entry
Network effect
Definition
The utility of a user is increasing with the number of other users of that product or with the number of products compatible with
Cross-network effects
In two sided markets, where users' benefits from participation depend on the extent of user participation on the other side of the market
Platform becomes more attractive the more it is used
Exponential growth rate
Facebook
Youtube
Instagram
Netflix
Types
Direct
Indirect
Connection with traditional firms
Most firms are still product or service based firms
These can work with existing platforms to become their value creation partner
Example - Restaurants X Uber Eats
Traditional firms can transform into platforms
Example - Apple - First closed ecosystem->jailbreak->App Store
Enablers and important concepts
Artifical intelligence
The area of computer science that emphasizes the creation of intelligent machines that work and react like humans
Machine Learning
The metho of making computers learn and think as humans do.
Big Data
Large volume of data collected from various sources, which contains a greater variety and increasing volume of information from millions of users
These are all data driven technologies
Big companies constantly collect and analyze user interaction data , run ML models to predict demand taste, behaviour
Optimized recommendations
Big example - Netflix
Was an online DVD rental store
Two major change
Turned into an online streaming service
Started to produce its own exclusive content
Netflixonomics
New market
It offers what you want to watch
Immediate feedback -> analysis, taste clusters, recommendations
Major content providers (many language, exclusive content, niche and blockbuster prodcuts)
Tailor made content
Very successful
Not easy to replicate
New Competitiors
Global player in the content market
Blurred frontiers between markets
Netflix competes with
Major TV channels
Amazon Prime Video
Disney Plus
Apple TV Plus
20% of the world's downstream bandwidth
Network externalities
Definition
Describes how the demand for a product is dependent on the demand of others buying that product
Positive network externality -> network effect
If there is network effect ->
network good
Types
Direct network effects
Consumers value the number of other users
Communication markets - benefits for consumers come from the ability to communicate with other users
Example - Waze
App offering real time traffic information
Quality of app increases with the number of users
Users
Active - Reports incidents
Passive - Share information about speed and location
To provide reliable information it need
mass of users
How do they make money?
Advertising (in-app location-based advertising services under a pay-per click (PPC) pricing model
Ad formats
Branded pins
Pins on Waze maps are like store signs and show up whether the user is driving
Search ads
Ads displayed in top of the search results when a user searches for anything from within the app
Zero-Speed Takeover
Digital billboard ads that pop up in a case a driver stops for at least 3 seconds
Training self driving cars
Based on passive users data
Waymo
Waze and Waymo are both part of Alphabet
Improving traffic predictions and route planning
Example - Fortnite
Huge success - larger community - free to play
In - game purchases - skins, emotes, weapons
Limited availability -> rushing -> classic behavioural psychology trick
Epic understood the social psychology behind its customers - natural inclination to be different wile also joining in
Travis Scotts' virtual performance - marketing campaign and reinforced network effects
Indirect network effects
Consumers value the number of compatible products
Sytem markets - Benefits for consumers come from the combination of complementary products or functionailites
Example - Google Play
More than 3.8 million apps available
Customers value the number of apps not users
App developers value the number of users
The number of applications is likely to depend (indirectly) in the number of Android users
Disintermediation
More likely to occur when
Risk - Low
Danger that a deal will go badly
Skill - High
Level of expertise required to perform tha task
Rake - High
Platform charge for facilitating a deal
Interaction frequency - High
How often partners must commuicate
Urgency - Low
The need to solve a problem now
Project modularity - High
Degree to which a task can be divided into smaller independent components
How to stop users from leaving platforms
Carrots - Supplier
Producer tools
Seller insurance
Frequent supply rewards
Royalty fee becomes subscription fee
Project management
Carrots - Buyers
Producer reviews
Frequent buyer rewards
Returns, money back guarantee
Safety
Bidding services
Sticks - Suppliers
Secret shopper
Bounties
Public shaming
Sticks - Buyer
Contact hiding
Secret seller
Bounties
Business models of digital goods
Types
Wholesale model
Upstream firm (manufacturer) offers the product to a downstream firms
The retailers sets the market (retail) price to consumers
Agency model
A retailer ( platform) is considered to be an agent of the seller (manufacturer) and has apparently) no involment in price- setting decisions
The retailer charges comission fee (rate) when the transaction is completed
The comission fee is usually a percantage of the market price (revenue sharing)
Examples
Apple takes 30% off
Amazon - around 30 but can exceed
Booking.com 15% on average
Apple example case study
Entering E-book market
Amazon used wholesale pricing arrangements
Apple launched Ibookstore with agency model (revenue sharing 70-30%)
Most-favored-nation MFN clause - Ibookstore would always sell the e-book at the cheapest price
This raised the prices in the long term since apple and the publishers colluded
Pricing and two-sided markets
Characteristics
Price structure is non-neutral as regards to the network effect
Platfroms aggregate demand and profits depend on the price structure, that is, on how the price is distributed on each side
Most cases the price structure is assymetric
Price discrimination is often adopted
Platforms with varying role in pricing
Sets price - Uber
Recommends price - Airbnb
No role in price - Task Rabbit
Price discrimination
Definition
The same prdocut can be sold to two different buyers at different prices
It is possible if
Identification
No arbitrage
Types
Personalized pricing - 1st degree
Individualized price for each unit purchased by each buyer -> full surplus extraction
Personalized prices are a seller's dream - perfectly kowing the willingness to pay of each consumer
Today in online markets it is almost possible
Scandals and practices of personalized pricing
Amazon - not deleting cookies results in higher prices
Orbitz - higher prices for Mac users
Home Depot - different price based on zip code
Uber - battery level
Group pricing - 3rd degree
Segmantation based on indicators related to consumers' preferences - > different preices per group
Examples
-50% for students
10% for 60s plus
Discount for locals
Different categories of products (elctronics, clothes - AMazon, ebay)
It can be Pareto improving while traditional on-sided makets price discrimination can be welfare enhancing but never Pareto optimal
Consumers are better of under price discrimination even if the arket price increases, as they value that the platform attracts more developers
Menu pricing - 2nd degree
No observable indicators -> use of self selecting devices (target specific package for each class of buyers)
Examples
Spotify - different packages
Apple - menu of Ipads
Uniform price - No information
Dynamic pricing
Definition
Pricing strategy in which business set flexible prices for products or service based on current market demands
Price can change multiple times a day - comes from tourism sector flights
Example
Coca Cola Vending machine - ifferent price based on temperature - big fail
Amazon have huge successes
Syncronization of e-commerce, omnichannel, and brick and mortar stores
Criticism - few minutes of browsing - 10 % increase
Freemium and versioning
Freemium
Combination of free and premium
Users are charged a premuim for access to advanced fundamental features of a product or service are provided chargefunctionalities
Examples - Spotify, Dropbox, LinkedIN, Tinder, Zoom
Versioning
Creation of multiple product releases which maintain a common general function while continually enhancing, upgrading, or customizing each iteration
Example - WIndows 10 and Windows 11
Pricing and Non-pricing Strategies in the Sharing economy
Sharing/Collaborative/Peer-to-peer economy
Connecting the "I need" with the "You have" resulting in efficiency and based on trust
Goal: Cooperation amongst a broad internet community
Botsman: An economic model based on sharing underutilized assets from spaces to skills to stuff for monetary or non-monetary benefits
Based on enterprise platfroms - regular people do business w neighbours tthan relying on big companies
It was made possible by the ease of sharing data brought by the internet
Local vs Global networks - Airbnb vs Uber
Growth of sharing economy
Rapidly expanding in the following sectors
Car sharing
Accomodation
Staffing
Music and video streaming
Peer to peer finance
Start-ups based on the sharing idea
Uber
Blablacar
Airbnb
Zipcar
On-demand economy instead of access economy
Crowdfunding
Platform gives opportunity to collect funds for projects or other goals
Bigger network -> quicker finance
Two classes
Reward and donation based CFPs
Investment based CFPs
4 types
Debt - Lending Club
Reward - Kickstarter
Equity-based - Angellist, CircleUp
Donation - Causes, Crowdrise
Review system
It requires volunteer data
More information allows consumers to make more informed choices
Especially useful with high information asymmetry
Advantage - less search cost
Disadvantage - Raise price, collect info, fake reviews
Essential - generate trust in the platform - reputation to not breakdown the market
Discourages bad behaviours from buyers
Bad reviews have bigger impact than good review on consumer purchasing decisions
Uber and Blabla
Uber
Generally
American multinational ride-hailing company
Offers peer-to-peer ridesharing, ride service hailing, food delivery and a micromobility system
Provides a technology platform -easy and reliable access to a ride
Uber pricing
Dynamic pricing algorithm
Adjust rates based on number of variables
Distance of you route
Traffic
Current rider to driver demand
Super pricing strategy
Very high demand - prices may increase to help ensure those who need a ride can get one
Adjusting the price attracts more drivers to an area so everyone an get a ride
Blabla
In general
French online marketplace for carpooling
Connects drivers travling form one city to another with travelers headed the same way
It does not serve customers looking to travel a few miles within a city
First it was an online platform for hitchhiking
Then he intoduced reciprocal rating system and making advanced payment compulsory and changed the name
Now
Leading car sharing platform in Europe
Non professional drivers and passengers meet and divide the costs
100 million members worldwide
User profile: name, age , profile pic
Bilateral review system
Prices are based in suggestion calculated by BlaBlaCar - cannot exceed a maximum set by BlaBla
Blabla could avoid many regulatory issues since drivers share the costs, they do not make profit - now they might sneak into Uber territory
Discrimination
Minorities as drivers
Sell fewer seats
Generate less revenue
Less popular listings
Blabla fight this with its rating system
Passengers tend to travel more frequently wiith drivers of the same ethnicity, and female pannsegers exxhibit a preferece for female drivers
Airbnb
In general
Founded in 2008
Online global platform where people share private spaces ti be rented for short term use
Regulatory issues
Differing regulation in basically every city
Barcelona study - Airbnb increased the rent prices and transaction prices in the neighbourhood areas
Cultural issues
UK - episde of racism
Australia - hosts raping and murdering their guests
US - black sounding nam eharder time booking
Possible solutions - now you cannot see the full name
Competition Policy Implications of Digital Platforms
Generally
Dominant platforms often adopt
restrictive contractual restrictions
thereby raising
antitrust concerns
Very dynamic sector
Enabler - digital revolution
Competion Policy
Definition
Applying rules ti make sure businesses and companies
compete fairly
with each other
Goal
Ensure and encourage
Low prices for all
Better quality
More choice
Innovation
Better competitors in the global market
EU Competition Policy
Mergers and acquisitions
Antitrust
Cartels
In the EU - Article 101(1)
Prohibits restrictive agreements and concerned practices that prevent, distort or restrict competition
Distinguishes between horizontal and vertical agreements
Horizontal - increase market power and reduce competition
Vertical - ambiguos effect
Whistleblower communication tools
Leniency policy
Hand-over proof - total immunity / reduction of fines
Very deterrent effect on cartel formation by seeding distrust and suspicion among members
Examples
Truck producer cartel - 2016 (Man, Volvo, Daimler, Iveco)
Algorithms, ML and collusion
Price-settings algorithms
It only know the aim of the exercise and learns from experience
Antitrust - these may independently learn to avoid price wars
They might
learn to collude
even if they have not been instructed to do so ad do not communicate with one another
Growing evidence for tacitly colluding
1 more item...
Monopoly
State Aid
Competition Policy in the Digital era
Digital economy charactheristics with huge impact on competition policy
Network externalities
The role of data
Extreme returns to scale
Abuse of dominant position
EU - Art 102
" Any abuse.. shall be prohibited as incompatible with the common market in so far as it may affect trade between Member States"
Abuse of market dominance
Exclusive discounting
Predatory Pricing
Margin squeeze
A dominant vertically integrated firm in the upstream market prevents its downstream competitiors from achieving an economically viable orice-cost margin
Margin = Retail price - Wholesale Price < Downstream Costs
Bundling
Profitable when
Customers have heterogeneous preferences and their demand are negatively correlated
Price discrimination is not feasible
Type
Pure
Two or more products are only offered as part of a package
Mixed
Two or more products are offered within a package or sold individually
Antitrust
Abuse of monopoly power if this strategy deters entry in at least one of the markets served by the firm
Leverage Theory - Multi product firm use monopoly power to hinder entry in a complementary market
Entry- deterrence effect
Bundling as pro-competitive
Example 1
Dropbox dominant with network effect
Google provided range of products that were complementary and could enter
Online-storage market
Example 2
Amazon entering online video streaming market
Dominated by Netflix
This market has indirect network effects
AMazon gives free one-day delivery etc
Complementary products- customers save time
Entry into markets with network effects - platform envelopment
Teacher research
Complementaries can help a firm enter a market with strong network effects and incumbency advantages?
Main results
1 more item...
Dominant positions in online markets
Currently no international agreement on how nd when authorities should intervene.
Platform bsiness model presents specific features that render antitrust investigations
Platforms may offer consumers a great number of their products for free
Case studies
Microsoft
Back then
Eu fined them for tying WIndows Media Player to WIndows
Bundled Internet Explorer wih Windows 7 E
Now
Complaint from Slack
Bundling Teams with its Office 365 packages
They must show the price difference for with or without Teams
Intel
CPU Market
For laptop makers give big discount for exclusitivity
For wholesalers gives discount if exclusivity
Exclusive discounts per se illegal
If its in favor of customer in terms of efficiency then its fine - Intel won
Google
Eu fined them
Pre-install Google Search and Chrome as a condition for licensing Play Store
Pays manufacturers to preinstall Google Search exclusively
Restricts development of new open source versions of Android
Amazon
Using sales data to gain unfair advantage over smaller sellers on the Marketplace platform
Buybox?
To adress
Data - not to use no-public data for its retail business
Buybox - to treat all sellers equally whn ranking the offers for the purposes of the selection of the Buy Box winner
Apple
Spotify
Mandatory use of Apple's own in app purchase system
Anti-steering provision - preventing developers to inform about cheaper purchasing possibilities
Dominant positions for distribution of music streaming apps in App Store
EPIC
They bypasses the App Store payment system
Apple blocked the game
Epic filled a lwasuit and failed
Because of DMA they have to accept sideloading - security concerns
DMA - changed commision structure
Price Parity Clauses - PPCs
Types
Wide
This forces the client hotel to charge the lowest price on the contracted platform
Narrow
Allows seller to set lower price when selling through other platforms
But in both cases sellers cannot charge lower price through its direct channels
EU against PPCs
Online Travel Agencies - OTAs adopt PPCs mainly
EU say that it can eliminate competition for low end rooms and prevent entry if alternative platforms
Current situation
All rate parity clauses banned
Italy
France
Austria
Ban announced
Belgium
Switzerland
Ban for som OTAs
Germany
Sweden
The DMA prohibits very large platforms from using all types of PPCs - 2023 may
Not only OTAs companies like Amazon can also apply this
Removal of PPCs
Short term effect - Significant price decrease -2.6%
Medium term - limited effect -1.6%
The complete removal of PPCs produced only limited effects
Chain and highly rated hotels registered more relevant price reductions 12-10%
Effect in France
Non statistically significant effectson prices both on OTAs and on hotel websites
It might be not that effective since
Sellers might practice price parities to reamin in good terms w the platform
They might be afraid of dimming
Suggestion - make the recommendationalgorithm more visible
Alternative solutions
Peitz suggest that platform can simply ajust its recommendation algorithm to increase the visibility of hotels that generate high conversion rates
Regulatory interventions
Examples
US cities - comission caps
New York - third party delivery services cannot charge more than 15%
Canada - caps of 15 to 18%
Platform may refuse to operate
Uber quit South Asia because of new regulations f ride-hailing
Apple threatened to terminate Apple Pay in Australia - NFC
Amazon forced to remove products in India to protect small retailers
Alternative solution
Tests to assess platform contribution to producer/consumer surplus
Cloud Computing
The economics of cloud computing
Definition
Cloud computing is a model for enabling ubiquitous, convenien, on demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction
Main characteristics
Access to computing resources
which are used to process workloads and store data
On demand self-service
hardware and software are available without requiring human interaction with each service provider
Broad network access
private or public internet connection
Measured service
cloud systems automatically control and optimize resource. Resource usage can be monitored, controlled and reported
Rapid Elasticity
hardware and software are avilable on a scalable and elastic basis
Services
Paas
Enhanced computing resources to develop, test, deploy and run
Supplier controls the infrsastructure and the platform, customer controls apps and data
Saas
Completed applications
Suppliers control infrasteucture the platform and all apps
Iaas
Raw computing resources to process workloads and stoe data
Control - Supplier controls the infrastructure, customers controls everything else
Deployment model
Private Cloud
Single tenant - computing resources are dedicated to a single customer
Hybrid Cloud
Combining aspects of both piblic and private cloud
Public cloud
Multi tenant : computing resources are shared among multiple customers
The growth of the cloud and its most imprtant players
AWS 2006 -> Microsoft Azure 2010 -> Google Cloud Computing 2011
"hyperscalers"
Many business rely on these services - Netflix, Spotify, banks
Key advantage - flxibility
Smaller players
IBM
OVHcloud
Oracle
Anticompetitive concerns and policy initiatives
there is potentially a lot of competititon
while degree of concentration is very high
High entry costs, increasing returns to scale, strong network effect
Natural oligopoly situation
High egrees fees
Definition
The fee of the switch from one cloud provider to another
Imperfect information about different offerings, limited foresights->
locked-in effect
Customers are often unable to anticipate their future need in terms of data and traffic
Alternative solution
Banning egrees fees
Can lead to higher storage costs penalizing customers
Inefficient transfer of data
Commited spends discounts and cloud credits
Definition
Units of virtual currency required to perform certain tasks on the cloud
Large customers may agree to spend a predetermined amount with a cloud provier in return for a percentage discount
These discounts and credits can benefit customers in the short run due to lower prices and ease of payment
They can encourage some customers to use a single cloud provider
Technical restrictions on interoperability
Some technical restrictions may be engineered by leading cloud providers to defend their positions
AWS and Microsoft Cloud present interoperability limits
Data Act
European Council in 27 November 2023
Article 29 - Egrees fees are forbidden - 3 year transition period
Article 30(1) and 35
Requirements for technical aspects of switching between data processing services
Cloud providers must guarantee functional equivalnce after switching - minimum level of functionality