Operator billing, continued

Two factual errors, one serious critique and a bunch of new facts and figures about operator billing.

Factual errors

It turns out I made two factual errors in my original piece.

Crazy Frog

Crazy Frog’s price was not 9cts, as I originally said, but more like 99, as two people with actual experience with the ringtone mailed me. This was pointed out in several emails and I added an update to the original post. Must. Quote. Sources. Or I can’t find them again when I need them most.

This means that the half billion the ringtone earned is to be divided among ten times fewer people. That’s math of a different order, but it’s still 500 million people, and I continue to believe that a purely credit-card-driven system would not have reached so many.

Credit cards

I estimated the number of credit cards in the US at 62 million, following an article that has meanwhile disappeared. A reader mailed me to point out the existence of this study (PDF), which puts the number significantly higher.

In my defense, I searched for “number of US credit cards” and found the 62 million article, but not the PDF. That only goes to show that sometimes a web search isn’t enough to gather such data.

Also, it turns out that calculating the number of credit cards is pretty hard. From the PDF, p. 56:

For example, to estimate the total number of credit card users in the U.S. from the percentage of U.S. consumers who had a credit card in 2008, one would take the percentage of U.S. consumers who had adopted credit cards from Table 4 (78.3 percent), divide by 100, and multiply by the population in 2008 (225,852,350) to get 176.8 million consumers. Similarly, to estimate the total number of credit cards held by U.S. consumers in 2008, one would multiply the per capita number of adopted credit cards, which is reported in Table 7 (2.7 cards), by the same population count to get a total of 609.8 million credit cards held by U.S. consumers.

I choose to go with the lower 178 million; 609 million seems FAR too much. (Maybe it includes corporate credit cards that are shared among a lot of employees. I’m not sure.)

So there are 178 million credit cards in the US, which amounts to about 24% of the (adult?) population not having access to one. That’s significantly less than my earlier 47%, but it’s still not even close to zero.

And I wouldn’t be surprised if the US has the highest credit card density in the world. In any case, people in the developing countries don’t have access to one and need a mobile/online payment system that doesn’t involve credit cards.


Dean Bubley wrote an interesting critique of the general idea of operator billing. I’d like to give a few partial answers, and concede a few points.

The critique also treats the disappearance of cash. I never said cash is going to disappear any time soon. Tomi Ahonen did, and I think it’s an interesting idea, but I’m not sold on this happening right now.

I do think credit cards are going to be threatened in their near-monopoly of web/app purchases, though I’m less sure about their future role in other purchases.

With that said, let’s go to a few specific criticisms. (I left out those having to do with cash.)

The average mobile prepaid credit balance is something like $5 (much lower in developing countries) and is frequently zero. Not idea for doing your shopping, or even buying a coffee.

A cup of coffee doesn’t cost $2 in a country like Kenya. The question is not how much money you have on your phone, but how much purchasing power. (I don’t know the answer, but I do know $5 buys you a lot more in Kenya than in the West.) So I respectfully reject this argument.

Even some of the most enthusiastic NFC-promoting operators don't believe in the idea that customers will charge purchases to their phone bill or prepay accounts. [...] Few customers want a fridge or airfare appearing on their bills at the end of the month. [...] the [phone] bill is fine for small digital goods, but not major purchases.

To be honest I never thought about fridges or plane tickets when I wrote my piece, and I agree there would be a psychological barrier. There is no such barrier in smaller web or app purchases.

And if people were to buy fridges with their mobile phone, they would have a subscription, not a prepaid account.

Still, there is no fundamental difference between the purchase of a fridge being charged to your credit card or your mobile phone bill. You have to pay up anyway, and you have to do your administration anyway. It’s just that people aren’t yet used to seeing those amounts on their mobile bill. Psychological barrier, no real barrier.

Governments rather like cash and seem unlikely to want to permit or encourage its death. It's a bit difficult to do quantitative easing via mobile phone.

Big-time banking will not be done by mobile phones any time soon. But it isn’t done by cash or credit cards, either.

But I admit I have been too quick in assuming banks would be seriously threatened by operator billing. I wrote the original piece when I hadn’t yet found the description (PDF) of M-PESA, and it turns out banks still play an important role.

In the SafariCom example, M-Pesa accounts are distinct from the mobile phone airtime account. It's basically a bank with a mobile-phone front end. Which makes a lot of sense, especially for the "unbanked" who would otherwise have cash stuffed under the bed & would be unable to make payments. But it's not the mainstream view of a "mobile wallet" substituting for cards or cash in developed economies.

I don’t really understand this point. M-PESA is a front for banking, but so is a debit card. Having a debit card makes simple, practical banking processes such as buying something a lot easier, and again I don’t see the difference with mobile payments from a consumer’s perspective.

Of course there must be financial communication between your bank account and your mobile account, but they already exist: pre-paid or montly bill. (Or your phone account could become your bank account, but that is still futurology at this point.)

There are no "payment portability" or "mobile banking portability" laws yet. Given that most (all?) operator mobile payments linked to an access account, users unlikely to lock themselves in & embrace huge switching costs when they want to churn phone provider. Obviously operators want to reduce churn, but assuming your customers are gullible and stupid is not a longterm winning strategy. Loyalty isn't the same as lock-in.

Interesting point that I hadn’t thought of yet. There are several solutions, such as having two overlapping accounts for one month, and paying only from the old account while receiving money only on the new one, or sending the money from your old to your new account, but operators won’t like it and will likely have to be regulated here.


  1. Mobile payments will initially be about small amounts.
  2. Mobile payments will not replace cash any time soon (and I never said they would).
  3. Mobile payments threaten credit cards more than they do banks, especially in web-based purchases. I should have been more clear on that point in my original article.
  4. From the consumers’s perspective mobile payments are best compared to debit cards.

Facts and figures

There are a few other stories about operator billing that caught my eye:

This is the blog of Peter-Paul Koch, web developer, consultant, and trainer. You can also follow him on Twitter or Mastodon.
Atom RSS

If you like this blog, why not donate a little bit of money to help me pay my bills?