One Laptop Per Child (OLPC) is not going to happen in India.
The Human Resources Development (HRD) ministry of the government of India recently decided to just say no to the $100 laptop that Prof Negroponte of MIT Media Lab has been furiously peddling. He wanted the government to buy, oh, about 1,000,000 of those at the modest cost of $100,000,000 and give it to school children. Mind you, noble intentions motivate this: so that no child is left behind and the digital divide is bridged and all the kids will become computer savvy and what not.
The HRD explained that according to some American psychologist “any sustained use of computers may lead to a disembodied brain and bring about isolationist tendencies in social behaviour” and that the “pedagogic effectiveness of this initiative is not known.”
Not just that, it went on to warn that “Both physical and psychological effects of children’s intensive exposure implicit in OLPC are worrisome. Health problems of our rural children are well known; personalised intensity of computer-use could easily exacerbate some of these problems”.
I bet the good folks at the HRD ministry are not as careful when it comes to their own children playing with laptops and PCs in their government provided flats in New Delhi. The reasoning behind promoting OLPC in poor countries is flawed (as I had written earlier: Formula for milking the digital divide); but the reasoning behind the HRD ministry’s rejection of the OLPC is worse. I am not surprised.
However, the Secretary to the Ministry, Sudeep Banerjee wrote to the Planning Commission and argued that instead of spending on laptops, funds should be allocated to univeralizing secondary education. Good point, Mr Banerjee. Still, Banerjee said that OLPC “may actually be detrimental to the growth of creative and analytical abilities of the child”. Not at all convincing.
My opposition to the OLPC revolves around the notion of opportunity cost. First, let’s briefly consider the total cost. There’s the direct cost of a laptop, which was first advertized to be $100 but now has been pegged at $140. Add to that the operational costs. They will include the cost of maintenance. Assume that over its lifetime, given that it is a new piece of hardware, it is a conservative 25 percent, or $35. Then there are the “use costs.”
Use costs are incurred because the laptops are used by people. Predictably, people–especially children–drop things, misplace things, get things stolen. So what happens then? Does the government replace those laptops? Who pays?
Then who gets those laptops? There are, I estimate, about 100 million school-going children in India. Can we afford to buy laptops for them all? If not, who then will be favored? Will there be “reservations” for laptops so that favored religious and caste groups be given preference? Who decides? Will those in charge of handing out the laptops make a bit on the side, either directly or indirectly, through their power to deny or grant a shiny new gizmo to thousands of people. Power in the hands of people invariably corrupts them.
Who owns the laptop? The child or the parent? What does ownership mean? Will the parent be held liable for the cost of the laptop if the laptop is “lost”? Will a very poor family be able to shoulder that liability? Remember that the cost is $140, which is about 30 percent of the per capita income in India. Who pays for routine maintenance? If the user is not responsible, then there is the problem of moral hazard: the user will not be sufficiently diligent in caring for the object.
The total costs then is the sum of the direct costs ($140), the maintenance costs ($35), the use costs ($25, say): $200. Let’s say that India buys only 2 million of those cute green machines. The cost: $400,000,000.
Now on to the reason why I oppose the OLPC: opportunity costs. Some time ago, I had explore the notion of opportunity costs in “Casting Spells to Fix a Broken Car.”
The proponents of OLPC argue that spending hundreds of millions of dollars on laptops will empower many children, educate them, make them cross the digital divide. You will not get any argument from me against that. Some unknown percentage of those who use those laptops will benefit from them; some unpredictable percentage will get computer literate. Those things will happen because of the OLPC. My concern is with things that will not happen because of OLPC.
This point is worth stressing. It is not just that we make A happen; we have to also recognize that we have to forego the opportunity of making B happen. The important thing is to weigh the benefits of A against the benefits of B. Only if the former out weighs the latter, can we convincingly argue for making A happen.
Spending a few hundred million dollars will help some children, and also enrich the manufacturers of the laptops (Chinese manufacturing), and all the middle-layers that will be invovled in the selling, maintenance, and support. Compare that to the alternative use of the same money.
Tens of millions of children don’t go to school, and of the many who do, they end up in schools that lack blackboards and in some cases even chalk. Government schools — especially in rural areas — are plagued with teacher absenteeism. The schools lack even the most rudimentary of facilities such as toilets (the lack of which is a major barrier to girl children.)
Attention and funds need to be directed to those issues first before one starts buying laptops by the millions. Fact is that we need basic education (literacy, numeracy, etc) and secondary education. These have been provided very successfully without computers around the world. Every one who went to school and became educated more than a mere 30 years ago–in the entire history of human civilization, billions of people in all–did so without having ever seen a computer. What they had was much less expensive than PCs: they had teachers and an environment conducive to learning.
Here is an analogy. By pushing OLPC, what they are trying to do is to increase the capacity of a tub made of staves of different lengths. How much water the tub can hold is then dictated by the length of the shortest stave. If one were to pour water into the tub, the water level will continue to rise but only uptil the level reaches that of the shortest stave, when it starts overflowing. To increase the capacity of the tub, you will have to lengthen the staves. But lengthening any of the staves except the shortest stave will not increase the tub capacity. And even lengthening the shortest stave beyond the length of the next shortest stave is wasted. So the strategy for increasing the tub capacity is this: lengthen the shortest stave(s) first to match the length of the next shortest stave(s), and repeat.
The shortest stave in our tub is the will and commitment of our policy makers.
Post Script: My friend Ethan Zukerman commented on this over here.