Skip to main content

INSIGHTS BLOG > First Game Theory Group Meetup


First Game Theory Group Meetup

Written on 23 October 2014

Ruth Fisher, PhD. by Ruth Fisher, PhD

Group Introductions

Introduction of HDTV Technology

Definition of a Game

Insights into Technology Games and New Technology Adoption

How Do You Change the Environment?

Mapping Out a Game

How to Determine Which Choice Will Prevail

 

October 22, 2104

 

Group Introductions

The 7 attendees were seated around a large table. Before the talk began, people introduced themselves and gave a short description of their background. The audience included gamers and software and systems people. Several people commented that they were interested in applying game theory to “negotiations” with co-workers, that is to figure out how to better get things done. Others were interested in applying game theory to gaming.

 

Introduction of HDTV Technology

I started my talk with a description of the introduction of the HDTV technology into the market in the 1990s. At that time, when digital technology was introduced, TV programs were being broadcast in analog. The government wanted the transition from analog to digital to proceed quickly, because the government planned to use the analog spectrum for emergency services and other government services.

However, users did not actually end up adopting the new technology very quickly at all, because (i) the TVs were expensive, and (ii) there was very little programming available in digital. Before users would adopt the new technology – by purchasing the expensive HDTVs – they wanted there to be plenty of programming available to be able to watch.

At the same time, the broadcasters refused to switch much of their programming over to digital, because (i) the resources used to provide digital programming would then not be accessible to, and thus generate revenues from, the majority of viewers, that is, people with analog TVs, and (ii) too few users had adopted HDTV to make it profitable for broadcasters to provide them with digital programming.

So what we have is

Users wouldn’t adopt HDTV because there was too little programming available.

Broadcasters wouldn’t make programming available because there were too few users to access it.

This is a classic chicken-and-egg problem that comes up often when new technologies are introduced into the marketplace, and it inhibits timely adoption.

 

Definition of a Game

The situation of HDTV adoption constitutes a game, because the payoffs to the players – HDTV Manufacturers, Program Broadcasters, and HDTV Viewers – are affected by the actions taken by each of the other players. Since each player’s payoff depends on what the other players do, it behooves each player to understand what motivates the other players, and thus what actions the other players might be led to take. Only with such an understanding will any particular player, say Player A, be able to change the environment to motivate other players to do what will end up being in Player A’s best interest. In other words, if Player A doesn’t like what he thinks the outcome of the game will be, given the current environment, he must change the environment to get the other players to take actions that are more beneficial to Player A.

More formally, a game consists of

•  A set of players: HDTV Manufacturers, Program Broadcasters, and HDTV Viewers

•  A set of actions available to each of the players:

°  HDTV Manufacturers choose how many HDTVs to manufacture and what price to charge each period;

°  Program Broadcasters choose how much of their programming to switch over to digital each period; and

°  HDTV Viewers: choose if and when to adopt (purchase) HDTVs.

•  A set of payoffs for each player, where each player’s payoff is a function of the actions taken by the other players:

°  HDTV Manufacturers: Profits are increasing in (i) the amount of HDTV programming made available by Broadcasters and (ii) the number of Viewers who have switched over to HDTV;

°  Program Broadcasters: Profits are increasing in (i) the number of Viewers who adopt (purchase) HDTVs each period, which is increasing in (ii) the amount of programming made available in digital broadcast; and

°  Viewers: Viewers switch over to HDTV when (i) the price of HDTVs is low enough and (ii) there is enough digital programming available.

 

Insights into Technology Games and New Technology Adoption

 

1.  Perspectives Contributing to the Adoption Process

The adoption process is influenced by a combination of factors provided by three distinct sets of perspectives:

•  Economic: When considering if and when players will adopt a new technology, the economic perspective considers such factors as the profits or utility of each of the players, the costs of purchasing the new technology, the switching costs associated with the new technology, any network effects that tie players to older technologies, etc.

•  Marketing: When considering how to motivate players to adopt new technologies sooner rather than later, the marketing perspective considers the price of the new technology, together with the amount and type of advertising as choice variables to use to influence adopters’ decisions.

•  Sociological: When considering if and when players will adopt a new technology, the sociological perspective considers the inter-relationships among people in society; cultural norms, attitudes and beliefs; etc., to determine whether or not the new technology is consistent with people’s beliefs and if so, how to use such beliefs and inter-relationships to influence the pace of adoption.

A review of studies on technology adoption suggests that analysts tend to consider the adoption problem from a single one of these three perspectives. And while each perspective has something very valuable to contribute, no one perspective alone is very satisfying. Rather, it takes a combination of all three perspectives together to provide a complete profile of the adoption process.

 

2.  Technology Systems That Exhibit Network Effects

A system can be described as exhibiting network effects when the value to each system user increases as the total number of system users increases. There are two distinct types of network effects

Direct Network Effects (DNE): A system exhibits DNE when technology users are able to interact or otherwise share resources with other users. For example, telephones and fax machine exhibit DNE because when there are more users with telephones or fax machines, then the technology is more valuable to any given user, because he is able to contact more other users.

Indirect Network Effects (INE): A system exhibits INE when complementary products and services to the technology base provide value to users. In this case, having a large base of users provides a larger profit opportunity for suppliers of complementary products and services for the base technology. Having a larger base of users adopt the base technology, attracts more accessory suppliers to the system. In turn, having a larger base of suppliers leads to an increase in the availability of complementary products and services. And having a greater supply of accessories available makes the technology system more valuable to users. For example, the iPhone exhibits indirect network effects, because a larger base of iPhone users attracts more App developers to provide Apps for the iPhone market. Having a larger supply of Apps available increases the value of iPhones systems for iPhone users.

When considering (adoption of) technology system games that involve network effects, it turns out that the universe of systems can be partitioned into four (relatively) distinct categories: (those with Low or High DNE) x (those with Low or High INE), with examples of each illustrated in Figure 1.

Figure 1

When considering the games formed by technology systems, the strategies associated with each system will be different, depending upon which of the four types of network effects combinations the system exhibits, (Low, Low), (Low, High), (High, Low), (High, High).

 

How Do You Change the Environment?

Much time was spent emphasizing that if a player doesn’t like the probable outcome that will result from a game being played in particular environment, then it is up to the player, say Player A, to change the environment to better suit him. In other words, Player A has to change the environment so that the actions that other players are led to take in the new environment are precisely those actions that benefit Player A. The only way Player A can do this is if he truly understands what it is that motivates the other players.

As a specific example, a player’s profit or utility function will often look like this:

Net Utility from Adoption = Utility or Profit from Adoption

       – Cost of Purchasing New Technology

       – Costs of Switching from Old to New Technology

Switching costs are any and all costs incurred by users associated with disentangling themselves from their old technology systems and adopting the new system. Switching costs include, for example, any costs associated with disposal of old system hardware, breaking contracts associated with old systems technologies, learning how to use new systems technologies, porting content from the old system into the new system, etc.

To the extent that New System Suppliers can decrease the switching costs to Users associated with adopting the new system, they will be able to speed the rate of new system adoption by Users.

 

Mapping Out a Game

One attendee asked how you go about mapping out a game. Do you start with a particular formulation and mold the Game into that formulation?

I used the example of Uber to illustrate how I go about mapping out a game.

Figure 2 is the diagram I used for my analysis of peer-to-peer (P2P) platform systems.

Figure 2

I started with Uber as the P2P Platform (orange node).

Next, I linked in Uber Drivers (light green node) and Users (yellow node).

After that, I linked in Taxi Drivers (dark green node) as the existing competitors.

I finished by adding in the Regulators (blue node), who currently regulate Taxi Drivers.

I briefly explained what each player wants. I indicated that Regulators currently increase the costs to Taxi drivers, which gives Uber Drivers a cost advantage over Taxi Drivers. Then I noted that over time, much of the regulations that currently apply to Taxi Drivers will eventually be forced upon Uber Drivers, which will erode Uber Drivers’ cost advantage.

I noted, however, that what drives a permanent cost advantage that Uber Drivers benefit from, but that Taxi Drivers do not, is the fact that Taxi Drivers must recover their costs of capital (the costs of their cars), whereas Uber Drivers don’t have to (see my analysis of P2P systems for a more complete explanation).

I also noted that when reading about the industry, I discovered that Uber Users seek a sense of community, that is, they seek a personal connection with Uber Drivers.  Knowing this would enable Uber and/or Uber Drivers to provide (emphasize) such an environment for Users, thereby speeding the rate of adoption of the Uber platform.

 

How to Determine Which Choice Will Prevail

One attendee was kind enough to demonstrate how he uses his own method to come to an unbiased conclusion as to which choice will prevail, when faced with a number of possibilities. He used my possibilities for my next Meetup location as an example. I have to choose between holding my next Meetup session in either (i) a community center, (ii) a hotel, or (iii) at some local startup office.

Here’s how the attendee explained his algortithm (illustrated in Figure 3):

Start by listing the choice of venues, together with the important attributes of each (see rows and columns of Figure 3).

Second, for each venue, rate the desirability of each of the attributes it has to offer (see numbers in the inner matrix of Figure 3).

Third, weight each attribute in terms of importance (see row of numbers above attributes in Figure 3).

Fourth, weight each venue in terms of desirability (see column of numbers to the left of venues in Figure 3)

Finally, take a weighted total for each of the venues across each of the attributes (see column [F] in Figure 3).

So, an unbiased analsis of the possible venues I have to choose from for my next Meetup, together with their respective attributes, would lead me to choose a community center for my next venue.

Figure 3