Now that One Laptop Per Child has started Give One Get One XO sales, its time to revisit the cost of the XO-1 laptop. According to OLPC, its $200 just for the laptop, if you buy at least 10,000. Now that doesn't even come close to the Total Cost of Ownership, which is upwards of $1,000 per laptop when implementation and maintenance costs are added in.
OLPC Indonesia
But for argument's sake, let's go back to the base $200 laptop price. And let's take Sumner Lemon's example of Indonesia, one of the world's largest developing countries with a population of 235 million:
Indonesia has around 40 million students and buying all of them a laptop priced at $200 would cost $8 billion, a sum that is 3.3 times larger than the money set aside for Indonesia's mandatory 12-year education program in the government's 2007 budget.Now Indonesia is not alone in facing astronomical laptop costs when looking at a one-to-one distribution model when barely having a budget for current educational expenditures. So how can there be one learning laptop per child in such a populous country?
On this, I suggest that OLPC look to the wisdom of Intel's World Ahead program staff. Leighton Phillips, manager of Intel's World Ahead Program in Asia, introduced a simple, but effective idea to Sumner:
One possible solution is a monthly payment program where parents pay for subsidized laptops in installments over the school year. But families in developing countries are generally poorer than in other countries-- Indonesia's 2006 per-capita GDP was $3,900 compared to $43,800 in the U.S.-- and that calls for creative financing programs to cover the cost of the computers.Or imagine an alternate Give One Get One program. A global G25%G1 if you will, where relatively wealthy buyers of XO-1 laptops in the developed world pay a 25% mark-up on $200 laptops, with that $50 premium going to subsidize an Indonesian family's purchase of an XO on a payment plan."There are potential subsidies and there are different ways this is happening; some of it can be government-led, some of it can be corporate-led," Phillips said. With subsidized laptop programs, families could be asked to pay $10 per month in addition to existing tuition fees and receive a computer.
No, it's not a pretty as G1G1, but at the same time it could be significantly more effective. There would be exponentially more buyers of XO laptops and selling them in the developing world too can reduce unintended G1G1 consequences.
If XO's are given out free, there would be a propensity for recipients to waste computers - they expect they can just ask for another if the first one is lost or damaged or they could sell them for quick cash. But with a $10 per month payment plan for a $150 laptop, the XO would have a real value, yet only up to $250, which isn't enough price disparity to cause eBay XO sales to American buyers.
Parents + child + XO = learning
Now that would mean One Laptop Per Child would be expected to come up with a maintenance plan better than Humpty Dumpty on a million unit scale. In addition, there will need to be a cultural integration program for parents who've never seen a computer, much less the Internet.
Last but not least, they are also going to want proof that their child can "learn learning" more effectively with an XO vs. a Classmate, Eee PC, or a similar investment in One Teacher Per School.
My hope: OLPC's punt to the pilots develops all three.
TCO has been given a really bad name due to misappropriation by PR people (especially MS).
In the long run, such a TCO calculation should be done using local wage levels. If the labor costs are close to 25% of the per capita GDP, you are talking about a month at a teacher's salary (some random figures).
Anyhow, giving away the laptops can be hardly expected in countries where the parents have to pay for everything in school. It would also be bad for all the reasons you mention. So I have always expected that receipient countries will install some payment plans. I have vague recollections of having seen such suggestions somewhere in the early days of the OLPC.
Of course, the payment plans w/sh/could be complemented by subsidies for some classes of families.
I am surprised this was never explicitely mentioned elsewhere.
Winter
According the infallible wikipedia, TCO orginated for enterprise/corporate use. Not sure what effect MS's use had on its perception except for those who don't care for MS. It was applied to some education projects in other countries by staffers at the Academy for Education Development, and I learned about it there.
It's useful if only to get check signers and deciders to think about the complexity of the project they are paying for, as Wayan outlines for Indonesia.
"Not sure what effect MS's use had on its perception except for those who don't care for MS."
Just that MS has shown that you can use TCO studies to prove that expensive == cheap, cheap == expensive, and black == white.
TCO can be a form of numerology, where exact numbers are used to paste over fog and smoke. The problem is not limited to MS, they just came in the news most (eg, with their "Gut the Facts" campaign).
http://asay.blogspot.com/2005/09/commentary-of-tco-studies-and.html
Basically, general TCO is ill-defined in every respect. You can define a good TCO for a specific project. No doubt Wayan's example was done right. But any real TCO is extremely dependend on every last detail of the implementation. For instance, the cost of sys-admins depends crucially on the availabillity of people and applications. And a new targetted virus can cost you dearly.
A single application might become available this afternoon that halves the number of sys-admins needed. On the other hand, the availabillity of 10 million XOs might result in the training of thousands of sys-admins with a considerable decrease in salary, reducing your TCO.
In short,"Past TCO performance is no guarantee of future TCO results"
Winter
The paternalistic attitude of this discussion is disturbing. Why should people in the west be deciding how the laptops get paid for once they're in their destination country?
These computers aren't being sold to individual people, they're being sold to countries. The people of that country can decide for themselves if they think the laptops should be free from the government or paid for by the recipient - it's really none of our business.
There are innumerable cultural and political nuances in each OLPC country. The idea that we can come up with the one true model for how people should get a laptop is ridiculous. Free market ideals are very popular in the US, and US thinkers are always eager to apply them to struggling economies in the third world. This has very far from consistently positive results, owing mostly to an ignorance of - or indifference to - the actual workings of the local culture and economy.
What Winter said, and doubly!
"TCO" is an example of the idea that errors based on fuzzy math can expand (or be herded) geometrically, depending on the inclination and prejudices of the presenter.
For instance, many TCO figures include estimated "training costs," even when the organizations about which these figures are drawn in point of fact do very little actual training.
(Mini-rant: Small and medium-sized businesses often do none, and instead benefit from diffuse computer knowledge among their employees, and the day-to-day accidental cross-training that they do. Whether these companies *should* do more training may be a valid question, but it's not one with an obvious answer. "Training" is not a one-size fits all concept; I suspect most money spend on computer training is wasted, for a variety of reasons ;) In many cases, clearly, it's not, but there are local optimums which may defy easy or formulaic application.)
Details of what gets counted aside, TCO calculations are all speculative and provisional anyhow. In the case of the XO, I'm sure that the TCO as calculated in favor of the OLPC vision of the world would factor in and emphasize the cost of electricity, shipping (light weight vs. for instance using cheaper but bulkier desktops), number of wireless access points required, battery replacement need and expense (and when, by the way, will be see their new battery tech in other laptops?!), anti-theft provisions, etc.
I'm not saying that TCO is totally arbitrary; I *am* saying it's no more than an educated guess with considerable fudge factor, and that whenever someone mentions TCO it pays to examine their motives. Those motives might be anything but invidious, and yet lead to a very different end figure than those of another observer.
Anyone who can *truly* know TCO for years into the future shouldn't be merely selling computers, they should be doing all kinds of arbitrage based on discounted money values, and would (you'd think) be doing so very quietly, so as to maximize their take ;) It would require them to know a lot of information on oil stocks (affect price of plastic, shipping, and many other things -- and indirectly, *everything* else), future sources of electricity (if Mr. Fusion comes out tomorrow, who cares so much, for terrestrial applications at least, how many watts a computer requires?), political stability around the globe, etc.
In short, anyone who can accurately forecast TCO over the span of years for a new(ish) product is clearly Dr. Evil's smarter counterpart, and we should fear him.
timothy
"This has very far from consistently positive results, owing mostly to an ignorance of - or indifference to - the actual workings of the local culture and economy."
Although I generally agree, we shouldn't forget that countries like Indonesia are poor *because* they are unable to handle their affairs well.
On the other hand, there is no reason to assume WE know how to handle THEIR affairs any better.
As always, be cautious.
Winter
A possibility worth mentioning is that some very poor countries would not benefit from an OLPC program, givern the relatively high TOC.
Those countries could be better off investing their limited fund on improving their current education system and infrastructure:
1. More schools
2. More teachers
3. Repairing current school buildings
4. School supplies
It also important to remember that:
a. The OLPC program's effectiveness in education has yet to be determined
b. Computers are NOT a basic requirement for elementary education. It would be nice to give every student his own laptop as an early age, but it is not absolutely necessary. Even rich countries are not doing that yet, so there is no pressing need for poor countries to do it, either.
Algo, I don't think people here are trying to be paternalistic, as a matter of fact they tend to bend over backwards to avoid that.
I think they're trying to answer the question - "How can I afford to spend $8 billion on laptops?"
How OLPC is going to answer the question - "Why do we even need laptops?" I don't know. I'm assuming they're some evidence that this is the best way to spend the education dollars.
I am a bit confused by a few of the posts here:
Winter starts his comment talking about TCO -- not sure of the relevance of this comment to to the actual blog post. (It would seem more relevant in the extended TCO discussions in other posts, though).
Tim Lord appears to call into question the who TCO exercise.
"In short, anyone who can accurately forecast TCO over the span of years for a new(ish) product is clearly Dr. Evil's smarter counterpart, and we should fear him."
What's the alternative then? Simple belief that everything will work out? Seems rather irresponsible, especially when you're playing with public money. Even if TCO projections/estimates are fuzzy, aren't they a useful part of any budgeting exercise? And can't you, over time, actually collect data on true costs to better inform your TCO modeling going forward?
As for questioning the motives of anyone who attempts to engage in a TCO exercise -- well, shouldn't we always be looking at motives? Frankly, I find it more questionable NOT to include a serious discussion of TCO issues as part of discussion like these.
btw TCO was invented and popularized by Gartner. Steve's point questioning its connection to Microsoft in the first comment is well taken. (Gartner, by the way, worked with the USA-based Consortium for School Networking to create a costing tool quite widely used by school districts in the United States. It's a pity this tool isn't available for free use in developing countries.)
Maddie seems to have summed up the real question behind the whole OLPC value proposition quite succinctly!
Finally, the payment model from the Leighton Phillips quote sounds a lot like what Microsoft is using in Brazil, South Africa, etc. call "Pay-Go" (for pay-as-you-go, based on the pre-paid mobile phone model). One could imagine how Bitfrost could be used to aid in such a payment scheme.
The TCO referenced here is totally bogus. It makes any conclusions based on the given number bogus as well.
$108.00 Setup (1-time fee)
$27.60 Yearly training (multiplied by the tco period)
$135.00 Yearly internet access
The person who made up these numbers is comparing two very different models of computer deployment. He is quoting the cost of pilot projects and extrapolating this to mass deployment. The USAID projects are well known to be extremely costly in comparison to local efforts, due to the massive salary overhead costs of US personnel and bureaucracy. Countries choosing the OLPC are deploying them via their preexisting education ministries and universities. The most laughable addition to this cost structure is the Internet cost he imputes. Remember he is describing a cost PER LAPTOP in a massive deployment scenario. This can only be a plausible number if the build-out of a contry's network infrastructure is entirely imputed to the education system's use of these laptops.
The TCO question comes up time and again.
Below I will give a "plausible" TCO based on pure guesswork. To understand why these TCOs are so way off you need to understand the cost difference between wiring up a single school and training a dozen teachers by US personell, and wiring up 10,000 schools and training 100,000 teachers by local people.
An Indonesian teacher earns up to $1000 a month. A local trainer will be rich if she earns $2000 a month. Applying a US ex-pat might cost $10,000/month and more. Setting up 60 computers for the first time is hard work, setting up a million imaged XOs is a routine chore.
Below I gues that to come up to a TCO of $1000, the internet connection must cost around $5000/year. For one school, this is mostly correct. For 10,000 schools you definitely can get a better deal. Say, by a special deal with a mobile phone provider.
"TCO
$108.00 Setup (1-time fee)
$27.60 Yearly training (multiplied by the tco period)
$135.00 Yearly internet access
"
My own TCO guesstimates:
Millions of XOs, so fixed costs can be ignored. Only variable costs will add up.
6 year deployment of XOs
4 year average lifetime of XO (breakage and loss)
1 week training per teacher per year
30 pupils per teacher
1 server per 60 pupils, expected lifetime 3 years
1 School server $600 + $300 shipping + set up
Training:
1 Trainer earns $500/week, trains 25 teachers "locally"
1 Teacher earns $250/week
150% overhead for additional costs and on-line follow up
Training costs:
Training ($500/25 * 2.5) = $50
Teacher's salary = $250
----------------------------
added $300 * 6 years = $1800
Per pupil: $1800/30 = $60 TCO
Setup per XO: $5
2 servers: $1800 /60 pupils = $30
Together:
$200 1 XO
$100 0.5 XO replacement
$5 XO setup
$60 1 week training / year for 6 years
$30 School server
------------
$395 Computer infrastructure
Electricity:
30kWh/year per XO = 180kWh
10Watt * 10 hours/day * 300 days/year
(note that the charging power is higher than the consumption power, I guess in the order of 10W charging for average use)
The current production price of electricity on Java is around 200 Rp/kWh ~ $0.022/kWh
Thus, 180kWh/year * $0.022/kWh ~ $0.4/year
But let us take the cost at $10 a year on delivery for a total of:
$60 Electricity
Networking:
$10 6 mesh repeaters per 60 pupils in total $600/60
$X Internet connectivity, one connection per 60 pupils
Together:
$465 + $X TCO
To get at $1000, they need to pay around
60 pupils * $500 / 6 year $5000/year on internet connectivity for 60 pupils.
Winter
i need one of the laptops please.when i saw it on the news i waned to get some for my self.but what i dont know is if they will send some to ghana.but all the same i want some desperately.Please i want you people to send me one of the blue onse.my whole life i wanted to have a laptop but i dont get.I am 14 years in JSS 3.my adress is p.o.box ml335 mallam-accra ghana. please dont give a warrant for the post office guies to search or else they may take it and i may not revieve it.please you send me an email of what you will send to me.please dont spam me and send me one please.THANK YOU VERY MUCH