Bitcoin and the Network Effects Problem

“There are steps entrepreneurs can take to help reduce the network effects problem. It is, nonetheless, an obstacle bitcoin must overcome to achieve widespread adoption.” ~ William J. Luther

Editor’s Note: The following excerpt is adapted from The Gold Standard: Retrospect and Prospect edited by Peter C. Earle and William J. Luther.

Bitcoin proponents differ in the extent to which they expect the upstart cryptocurrency to circulate in the future. Some see the potential for bitcoin to replace the dollar. Others think bitcoin might serve as the primary currency for cross-border commerce, while reasonably well-managed national currencies like the dollar continue to circulate domestically. Still others expect bitcoin use will be largely limited to countries like Venezuela, Zimbabwe, and Argentina, where national currencies have been mismanaged in the past. But most have some notion of widespread bitcoin adoption in mind.

The market capitalization of bitcoin suggests there is some chance it will achieve widespread adoption in the future. Peter Hazlett and I have argued that the market capitalization of bitcoin conveys the extent to which bitcoin is useful for making transactions today or is expected to be useful in the future. As events suggest bitcoin is more likely to replace the bolivar (Venezuela’s currency) or capture market share in global remittances from Western Union or otherwise be more widely used in the future, people are willing to pay more for bitcoin today. The market capitalization rises as a result. On the other hand, when events suggest bitcoin is less likely to be used so widely in the future, people are willing to pay less for bitcoin today and the market capitalization declines. 

At present, the market capitalization of bitcoin is on par with that of many national currencies. There are around 18.8 million bitcoin in circulation today worth around $50,000 each, for a total market capitalization of roughly $0.93 trillion. For comparison, the market capitalization of the U.S. dollar is around $5.25 trillion. The market capitalization of the Chinese renminbi is $1.4 trillion, while that of the Canadian dollar is only $0.352 trillion. Since bitcoin is not widely used today, its market capitalization suggests those placing their money where their mouth is think it will be more widely used in the future.

Are they right? Perhaps. But those expecting widespread adoption would do well to recognize that bitcoin faces at least four obstacles: network effects, government opposition, a suboptimal supply constraint, and limited transactions capacity. In this essay, I discuss the network effects problem.

Network Effects

It is important to recognize that bitcoin faces entrenched incumbents. Since we are already using monies, choosing to use bitcoin means choosing to use some incumbent money less. It is a portfolio allocation decision. One wishes to hold a given percentage of her wealth in highly liquid assets to make transactions. Prior to adopting bitcoin, she is only holding some incumbent money (e.g., dollars, euros, baht, etc.). Increasing the allocation of bitcoin in her portfolio means decreasing the allocation of the incumbent money. This is a problem for bitcoin because network effects favor the status quo.

The demand for money is unlike the demand for many goods and services in an important way. I like cheeseburgers. Perhaps you like cheeseburgers as well. A lot of people like cheeseburgers. But the extent to which I enjoy taking a bite of a medium-rare, freshly-ground brisket burger with bacon and cheddar is completely unrelated to anyone else’s enjoyment of cheeseburgers. That is not the case with monies. We use money to transact with other people. If more of my trading partners choose to use a particular money, that money becomes more useful to me. My demand for a particular money, in other words, depends on my trading partners’ demand for that money, and vice versa. We do not just care about the characteristics of the money. We also care about its network of users.

Incumbent monies, like the dollar, enjoy a large network of users. Virtually everyone in the United States––and many outside its borders––will happily accept dollars today. Upstart monies, like bitcoin, do not have such a large network. Only a few people accept it. 

Suppose everyone independently recognized bitcoin to be superior to their incumbent money. Would recognizing that fact induce them to switch to bitcoin? Not necessarily. Perhaps they would be willing to switch if they were convinced a sufficient number of their trading partners would switch as well. But, absent such assurances, the safest strategy is to continue using the money you and your trading partners have been using. And that inertia can sustain the status quo, even if the available alternative is superior.

The coordination problem inherent in choosing which money to use is similar to that involved in choosing whether to drive on the left or right side of the road. In the abstract, it probably does not matter much whether we all drive on the left side or we all drive on the right side. The important thing is that we all drive on the same side! In the United States, we drive on the right side of the road. But suppose a highly-respected think tank were to circulate a very well-designed and well-written study to every American that clearly indicated driving on the left side of the road is superior to driving on the right side of the road. We might all recognize that driving on the left side would be better. But, even setting aside the legal issues, it would be foolish to start driving on the left side of the road before anyone else does. Absent coordination (and, likely, a change in the laws), we would continue to embrace the status quo.

Much the same could be said about monies. No one wants to use a money that no one else uses. Few people are willing to switch to a money that few people are using. Even if everyone were to recognize that bitcoin is superior to incumbent monies, they might continue using those monies if they are not convinced other people will switch as well.

I have explored the network effects problem more formally elsewhere. But the key takeaway is straightforward. It is not enough for bitcoin to be better than the available alternatives. It must be sufficiently better to warrant the costs of switching, which includes the costs of coordinating that switch with others. The network effects problem is not an insurmountable obstacle. There are steps entrepreneurs can take to help reduce the network effects problem. It is, nonetheless, an obstacle bitcoin must overcome to achieve widespread adoption.