Proof that cost differentials drive SMS usage

It's widely known that SMS usage in the Philippines exceeds that of any other country, but I haven't noticed a concise explanation of what's going on.  The fact that SMS started out as a free service is occasionally mentioned, but that was long ago — as SMS adoption soared, charges were rapidly introduced.

Now Babar Bhatti points to a study, Teleuse at the Bottom of the Pyramid, by LIRNEasia which includes some compelling data, e.g.

.

.

Of course there are multiple contributing factors, but the perceived cost of voice calls is surely one of the most important. Yes, there are issues of literacy, local language interfaces and other effects, but when considering poor people at the bottom of the economic ladder, cost seems paramount.

Comments

[Ayesha Zainudeen]

I agree with Mr Brough's comments, that although other factors are at play (language, literacy, etc), cost is probably the most important. However, when considering the 'cost' of an SMS, it is important to distinguish between the actual cost of sending one message, versus the relative cost of sending an SMS. For instance, Pakistan has been seen to have the lowest SMS charges among the five countries considered in the above-mentioned study, even lower than the Philippines, the 'SMS capital of the world' (see http://www.lirneasia.net/wp-content/uploads/2007/06/pricebasketsnewsreleasejune-2007_final.pdf). What matters really is the relative cost of sending an SMS; for instance, in the Philippines, for a Smart TnT prepaid mobile user, a one minute call is about 5.5 times more expensive. In Pakistan, for a Jazz Budget prepaid mobile user, the ratio is about 2.1. In India, where SMS use at the BOP was seen to be the lowest (among the countries studied), the ratio was 1 – i.e, a one minute call and an SMS are the same price (so why would you bother SMSing if that were the case?). In fact, the correlation between the relative cost and SMS use at the BOP be in Pakistan, India, Sri Lanka and the Philippines, was found to be positive, and very strong. Whereas the correlation between the actual cost of an SMS and SMS usage was found to be very weak.

[Dimuthu Ratnadiwakara]

It was discovered in this study (T@BOP) that sms usage is not influenced by the actual cost of an sms but influenced by the relative cost of a call to an sms (i.e. call charge per minute/charge per sms)

A very strong positive relationship was observed between the sms usage and the relative cost of a call to an sms. i.e. higher the relative cost of a call, higher the sms usage. The regression analysis indicated an R-square figure of 99.9% for the above relationship, which implies 99.9% of variability of SMS use is explained by (or due to) the relative cost of a call.

But the relationship between the sms usage and the actual cost of an sms was found to be very weak and the R-Square figure is close to 0%.

[Philippine data entry]

That’s great that you are teching technology to kids! I just found your blog and it’s really good! I’m going to read more posts!