How Blockchain Could Change The Management of Identities and Passwords and Prevent Occupational Fraud. Network Effect in Detail.
Before the Internet, the network effect had very limited applications, because scalability was not as easy as it is with technology. For example, if a restaurant were to become really popular, then adding more tables in at existing locations would cost much, but opening a new location, signing a commercial lease, setting up the space in compliance with all the local laws, doing remodeling and design would cost much more than just adding tables in an existing location. This same principle and example applies to almost all traditional brick-and-mortar businesses. There is a point up until which a business can scale, but the potential of network effect is capped by physical constraints such as floor space.
With technology, the game has changed dramatically because scaling became much easier.
However, not everything about network effects is simple. Gary Becker, an economist who has won a Nobel Prize in economics in 1992, studied the network effect in his now classic paper “A Note on Restaurant Pricing and Other Examples of Social Influences on Price” (source: https://www.unc.edu/~fbaum/teaching/PLSC541_Fall06/Becker%20JPE%201991.pdf )
Potential negatives of the network effect
In the paper, Becker discusses the following issue: if a restaurant becomes really popular and there’s a line of people waiting to get in, why wouldn’t the restaurant simply raise prices? Supposedly, if it were to raise the prices, the line would thin out, the restaurant would still be full, yet it would be making more money. In this scenario, everybody would win: customers would win because they would not have to wait in a long line and those who get in would be able to afford the food. The restaurant would win because it would be making more money due to higher prices. Finally, the area around the restaurant would also win because a shorter line would mean fewer people driving to get in line, which means less traffic, fewer people on the street even if they don’t have to drive, and so on.
While it may seem that a paper from 1991 about restaurant lines has nothing to do with blockchain technology, in reality the findings of Becker apply perfectly to blockchain networks in general and Bitcoin network in particular. For instance, the same logic about pricing used in the example above could be applied to a cryptocurrency network or a network that stores self-sovereign identities: if a network is becoming really popular, then it could raise transaction fees (in the case with Bitcoin) or it could impose conditions on users (in the case with self-sovereign identities) that users will abide by because of the popularity of the network. Becker’s work and history of the Bitcoin blockchain show that these assumptions are faulty and do not work.
As it turned out, there are several factors that contribute to the network effect.
Perception of fairness and reasonable requirements as necessary conditions of network effect
First, in the case of a restaurant, people need to perceive the pricing as fair and there are typically points at which people would refuse to pay for a product or a service. This is one of the reasons why restaurant chains with multiple locations, such as McDonald’s, test their prices and figure out what people would pay for the products.
With Bitcoin, the same issue happened with transaction fees on the network. In 2011, the fee to participate in a transaction on the Bitcoin network was equal to a fraction of a penny. By 2013, the fee went up and could at times equal to several pennies. From 2014 to 2017, sending funds on the Bitcoin network typically cost fewer than 50 cents. In 2017, as the network started becoming really popular and blockchain technology and Bitcoin in particular have entered the conversation in the media, the fee started growing exponentially. In July of 2017, it surpassed $4. On certain days of December of 2017, the fee to send funds on the Bitcoin network was over $50. (You can the chart with Bitcoin network average transaction fees starting with the inception of the network in 2009, here: https://bitinfocharts.com/comparison/bitcoin-transactionfees.html )
Obviously, nobody would use a network to send money to a friend for a cup of coffee if it costs $50 to send $3. This is one of the issues that organizations working on making self-sovereign identities a reality will need to solve.