Photographer: Tomohiro Ohsumi/Bloomberg
Photographer: Tomohiro Ohsumi/Bloomberg

My post on Bitcoin yesterday triggered the predictable burst of outrage; Bitcoin seems to be the Linux of the 2010s. One question I got, over and over, was why we had to define success as some huge number of users. My answer is that network products need to grow, or they'll die. But it’s worth unpacking that a bit.

First, though, let’s define what we mean by Bitcoin succeeding. This means different things to different people, of course -- the enthusiastic technologists mostly want to see its neat solution of the Byzantine Generals' Problem used to make the world a better place, while the speculators, who seem to be a substantial fraction of actual Bitcoin owners, want their Bitcoins to allow them to retire in comfort.

I’m not going to make predictions about whether Bitcoin’s ledger technology will become central to the world of commerce -- I’m not a technologist, so that question is outside my purview. I’m interested in Bitcoin itself -- both as a currency and as a new technology product. Success to me means widespread adoption of Bitcoins as a medium of exchange (I can use them to buy lots of goods and services) and as a store of value (I can save money in Bitcoins and be reasonably sure that if I come back in a couple of years, I will still be able to buy almost as much stuff with them in two years as I can today).

Bitcoin is obviously not there yet on either front -- and no, please don’t tell me about the bakery on your block that takes Bitcoins. In any given place, most stores do not take Bitcoins, which makes it a lousy medium of exchange. And because the price in other currencies can swing quite wildly -- and because everyone has to do most of their spending in other currencies -- it’s not a good store of value, either.

That could change if Bitcoin gets millions of users. But it really needs to get a lot of users in one place -- so many that businesses will be willing to accept Bitcoins the way they currently accept Visa.

That’s what makes it a network business. One way to think about network businesses is that they are the embodiment of the chicken-and-egg problem: Lots of businesses will want to take Bitcoins if lots of people use them, and lots of people will want to use Bitcoins if lots of businesses accept them as payment. In the beginning, this is a hurdle for a network product; once it's established, it’s a competitive advantage.

So the problem for Bitcoin is to grow -- especially because a lot of its users are currently speculators who are counting on that growth to make their Bitcoins more valuable. If they decide that Bitcoin is a good way to lose money, they will exit.

(Why would they suddenly decide this? Why haven’t they decided this already? I dunno -- why did people think was a steal in 1999 and a dog in 2001? People are herd animals.)

Network products that stall below “widespread adoption” tend to eventually die -- think Friendster or Laserdisc. Why? Because all such networks experience natural attrition. Right now, Bitcoin owners are essentially hobbyists. Most hobbies experience attrition over time -- people need the cash for something else, or other things in their lives become more absorbing. If those users aren’t replaced, this can be a real problem for collectors, because suddenly their precious collections lose value and, possibly even worse, become difficult to sell at all.

But at least they have the intrinsic joys of the objects themselves. Bitcoin collectors don’t. Electronic fiat currency is the ultimate network product; pretty much all of its value stems from the network.

Theoretically, you could spend a lot of time admiring the block-chain histories of your particular Bitcoins, but that’s not how I see people talking about Bitcoins: They talk about them like they're money or stock certificates. Most new users seem to be interested because they think that, eventually, the network will be fairly ubiquitous. If they stop thinking that, then new people will stop joining the network. Which doesn’t mean the network will stabilize. It means that the network will begin to shrink -- potentially quite rapidly, if the speculators stampede.

So the network has to get a lot bigger before it will have reached critical mass. As I said yesterday, I think Bitcoin’s problems get even larger if it does reach critical mass, because governments will step in to kill or co-opt it. But that’s a post for another day. Right now, I just wanted to explain why Bitcoin needs very high growth levels for a long time if it is to become a serious contender as a payment system.

The Rise of Bitcoin

To contact the writer of this article:
Megan McArdle at

To contact the editor responsible for this article:
Brooke Sample at