Evaluating the effect of Network Neutrality on content provider - PowerPoint PPT Presentation

Evaluating the effect of Network Neutrality on content provider business models Maurizio Naldi, Giuseppe D’Acquisto, Bruno Tuffin, Patrick Maill´ e ARC MENEUR meeting, Paris, May 2011 M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 1 / 6

Objectives Consider a competitive context involving content providers in competition. Some providers offer content for free, while others offer pay-per-use content. M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 2 / 6

Objectives Consider a competitive context involving content providers in competition. Some providers offer content for free, while others offer pay-per-use content. Investigate the effect of network neutrality on the results of that competition M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 2 / 6

Objectives Consider a competitive context involving content providers in competition. Some providers offer content for free, while others offer pay-per-use content. Investigate the effect of network neutrality on the results of that competition Does network neutrality favor pay-per-use, or other business models? M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 2 / 6

Objectives Consider a competitive context involving content providers in competition. Some providers offer content for free, while others offer pay-per-use content. Investigate the effect of network neutrality on the results of that competition Does network neutrality favor pay-per-use, or other business models? Also, this can have consequences in terms of incentives for content creation. M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 2 / 6

The model M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model F P The actors: � one (F) offering free content, 2 content providers: the other (P) pay-per-use content M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model F P N The actors: One network (access) provider M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model F P N U The actors: Users willing to access content M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model F P N U The money flows: M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model v F v P F P N U The money flows: Content providers benefit from advertising... M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model v F v P F P p S N U The money flows: ... and possibly by charging users for content M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model v F v P F P t F t P p S N U The money flows: Content providers have to pay to the ISP M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model v F v P F P t F t P p S N a U The money flows: Users pay a fixed fee... M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

The model v F v P F P t F t P p S N a p N U The money flows: ... and a volume-based fee M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 3 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. We assume a linear demand, of the form D P = D 0 P max[1 − α ( p S + p N ) + β p N , 0] α p N + ˆ D F = D 0 F max[1 − ˆ β ( p S + p N ) , 0] M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game Π P = ( p S + v P − t P − s P ) D P Π N = a − k + p N ( D F + D P ) + ( t P − n P ) D P + ( t F − n F ) D F . ���� ���� ���� OPEX OPEX OPEX M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game Higher level: Payments t P and t F from content providers to the network provider are determined M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game Higher level: Payments t P and t F from content providers to the network provider are determined Four possibilities ◮ Objective ⋆ maximize Network provider revenue ⋆ maximize social welfare ◮ Situation with respect to Network Neutrality, where here neutrality means t F = t P M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game Higher level: Payments t P and t F from content providers to the network provider are determined Highest level: a regulator may limit the amount of available free content M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

How the game takes place We model those interactions as a multi-stage game Lower-lever game (shortest time scale): users decide on their demand level on each content provider. Intermediate level: Network provider and pay-per-use content provider play a pricing game Higher level: Payments t P and t F from content providers to the network provider are determined Highest level: a regulator may limit the amount of available free content Game studied by backward induction M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 4 / 6

Pricing game among content providers Best-reply strategy of the content provider P : � � 0 , 1 + ( β − α ) p N − α ( v p − t P − s P ) BR P ( p N ) = max . 2 α Best-reply strategy of the Network Provider can be computed (5 cases to distinguish) M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 5 / 6

Pricing game among content providers Best-reply strategy of the content provider P : � � 0 , 1 + ( β − α ) p N − α ( v p − t P − s P ) BR P ( p N ) = max . 2 α Best-reply strategy of the Network Provider can be computed (5 cases to distinguish) analytically heavy, turn to numerical analysis M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 5 / 6

Setting the parameter values Values provided by G. D’Acquisto and M. Naldi � D 0 F = 50GB/month (per user) Demand max. levels: D 0 P = 30GB/month (per user) Fixed user fee a = 20 e /month  k = 5 e /month (per user, cost of providing access)    n F = n P = 0 . 1 e / GB/month (transport costs, OPEX: supported by the network)    s P = 0 . 2 e /GB/month (unit cost of the content for P ) � v F = 0 . 75 e /GB/month Advertising revenues v P = 0 . 25 e /GB/month M. Naldi, G. D’Acquisto, B. Tuffin, P. Maill´ e (2011) Modelling Net Neutrality MENEUR meeting, May 2011 6 / 6