Hey, can we talk about opportunity costs for a second?

I’m pretty sure the concept of “opportunity cost” is covered in the first week of every introductory economics course. It forms the foundation of more complicated econ concepts. But it’s like we just completely forget about it after that first lecture. The concept is tragically underutilized in public discourse.

In case your econ 101 class was a long time ago (or you never took one), here’s how Wikipedia defines it:

Opportunity cost is the cost of any activity measured in terms of the value of the best alternative that is not chosen (that is foregone). It is the sacrifice related to the second best choice available to someone, or group, who has picked among several mutually exclusive choices.

So if you’ve got $100 to save, you could just stick it under your mattress. That course of action seems to cost you nothing. But the opportunity cost would be the interest that you would earn from putting it in the bank. Applying the concept is straightforward when both your inputs and outputs are measured in dollars (or shillings, or yuan, or whatever).

It gets a little more complicated when you have other factors involved. For example, suppose you’re choosing between cooking dinner or ordering delivery. Cooking dinner is usually cheaper, but it also takes more of your time and planning to have ingredients on hand. Your personal preferences matter too: I enjoy cooking, but you might find it frustrating.

Does it seem like I’m belaboring the point? Well, there’s a reason I am:

The aid discourse seems to have little grasp of opportunity costs.

Which is amazing considering that economists dominate this field. Of course, the choices are inherently harder in aid than in the rest of economics. Our outputs are rarely measured in dollars. It can be very hard to compare outputs within a given field (say, health: what’s the relative value of stopping HIV transmission to 10,000 people, versus providing treatment to 10,000 who are already HIV-positive?) and even harder to compare across fields (say, how do you compare educational outcomes versus health outcomes?).

But the one place we should have little trouble is with the inputs. Those are almost always denominated in dollars, or in something that can be converted to dollars. Yet when we talk about committing resources to a course of action, we rarely talk about how those resources might otherwise be spent. Let me give two examples.

Opportunity cost of SWEDOW — Stuff WE DOn’t Want.

Last year, the anonymous blogger J. offered the aid world a new term: SWEDOW. In the post where he introduced it, he described his frustration:

This whole GIK (gifts in kind) thing is really driving me around the bend, lately. More than normal, even.

I’ve been annoyed with Soles for Souls for some time already. And although it came as no real surprise when Jessica Simpson signed on early to their unbelievably ill-conceived “50,000 shoes in 50 days challenge”, I admit that I was genuinely a little disappointed to see Wanda Sykes jumping on board, too. (Girl… you’re funny, but this idiotic shoe thing isn’t.)

Then, I became aware of Pedals for Progress and thought we’d reached a new all-time low on the pseudo-humanitarian aid GIK front. I mean, with some kinds of particularly high-tech or special application equipment that is not widely available, I suppose under duress I can see how professionally refurbishing “lightly used” pieces and then providing them as GIK in very limited and specific situation might not be wholly inappropriate.

But bicycles? Seriously? Is there a country in the world that is running out of bicycles?

The standard SWEDOW scheme involves collecting a bunch of old, used items (shoes, bicycles, yoga mats, whatever), getting some donated money to cover shipping, and then distributing the items in an impoverished community somewhere overseas.

Although J. never used the term “opportunity cost” in his post, that’s how I’ve always thought about SWEDOW. The question we have to ask when evaluating any SWEDOW scheme is this: If we sold all of those used items and combined that with the cash donated for shipping, what would the community in question do with that money? Would they choose to buy the shoes/bicycles/etc. that we had been planning to ship there? Or would they rather have something else? That “something else” is the opportunity cost.

The fact that a recipient expresses gratitude for the free items is not enough to prove that this was the best use of the resources. Who’s going to turn down free stuff? Who’s going to say, “thanks for the bikes, but next time please fix the road”? If you offer me a free slice of pizza, I’ll probably eat it — despite the fact that I’m mildly lactose intolerant — because it’s there and it’s free. (This is not an exaggeration. I can’t turn down free food.)

Opportunity cost of stuff I DO want

Pictured: a box. Not pictured: a school.

Not every scheme involves sending things we don’t want. Apparently there is now a plan to send something that I would love to have: an iPad.

This morning, Wayan Vota commented on a proposal to send a “School in a Box” (SIAB) to rural schools in Zimbabwe. The box (shown at right) comes complete with an iPad 2, solar charging panel, speakers, projector and more.

There are two points to make in response to this idea. Vota makes the first point:

A school is the summation of many parts, almost all of them human. Teachers, students, administrators, parents and the surrounding community all working together to educate children and lead them to adulthood.

There is no way all that can be squeezed into a box or expected to come out of one.

In other words, this isn’t a school. It’s just a box. An awesome box. But a box nonetheless.

The second point is to think about opportunity cost. The SIAB site hilariously compares the cost of the box to the cost of an Interactive Whiteboard, as if that were the alternative to this package. SIAB has a per unit cost of about €1200 — or $1600. I was curious if this whole thing would only be $1600 so I did a little back-of-the-envelope calculation. I came up with $1440 for just the parts. (The cheapest version of the iPad 2 costs $500; another SIAB document says the solar power system costs $390; the projector shown on their website costs at least $270; the speakers shown will cost about $30; the case shown with padded dividers would be another $250.) That leaves $160/unit for things like overhead, distribution, training and maintenance.

The opportunity cost question is this: What else could we do with $1600? Suppose we just limit ourselves to education in rural Zimbabwe. How else might we use $1600 to benefit Zimbabwean kids in a rural school? Well, we could hire them another teacher. Teachers in Zimbabwe make $220/month. At that rate, SIAB would be worth just over 7 months of a teacher’s salary. However, $220/month is well below the poverty line in Zimbabwe, which is why the country suffers from a shortage of teachers. For the past two years, the government has been asking parents to make up the salary difference with monthly payments to teachers. Such direct payments are not only unfair to poor parents, but they also warp the incentives for teachers. Clearly, there are other needs in Zimbabwe’s education system.

Zimbabwe isn’t the only country under consideration for SIAB. The politics and economics will vary by location. Yet I still feel pretty confident stating that in most developing country education systems, you can find a better use of $1600. The problem is that if you get enamored with the idea of an iPad in a box, you forget to look for those other uses. You forget to look for the opportunity cost. It’s the same thing that happens when you decide to donate old shoes, but it’s a lot more expensive with Apple products.

3 thoughts on “Hey, can we talk about opportunity costs for a second?

  1. Hi Dave,

    I have to say that I’ve agreed with just about all your posts so far.  Also agree with the general of this one, except that your target of choice is off.

    Yes, opportunity cost should be considered, especially for misguided GIK solutions.  But used properly, the SIAB or something similar but a bit cheaper is a steal for ROI.

    I don’t know why they chose and iPad2 or an iPad at all, as there are cheaper options, but the concept of empowering rural teachers with digital content powered  by solar is the one of the best values you’ll find for your development dollar. 

    There are many caveats, some of which I have discussed with Wayan in the past.  I agree that there is more to a school than hardware– something OLPC seems to disagree with given their latest helicopter scheme.  Most of these elements are human– teachers, parents, pta, training, maintenance, and on and on.  So robust and sustainable human systems need to be put in place around any technology solution.  This is one of our core focuses in Zambia where we are building a school system around such a concept.

    Also, curriculum is far more important than hardware.  I don’t know what SIAB is using but it will need to be available offline, comprehensive, culturally appropriate, language localized, be approved ideally by the MoE, updateable, include interactive activities and on and on.   ipad apps will not cut it.

    But back to opportunity cost, with a few cheaper parts you are looking at around $1000 bucks for a system that should last around 3 yrs to deliver (ideally) a great curriculum to around 240 kids, which is under $2 per student!  But to be fair, add the 3 local teachers needed to run the curriculum at $300/yr or so each (zambia and zimbabwe probably are similar here) so that is another $3 or so per student.  Then add some ongoing professional training for these teachers for another few dollars per student but still under $10/student/yr.

    This total will still be much below if you just hired 3 professional teachers at $220/mo.  (in zambia its more like $300+).  In this model, cost is more like $35/student/yr.  More than 3x the cost!  While we of course need to run an evaluation to test and compare quality, the quality of this sort of model run in government schools in Zambia is pretty low.  Early indications are that we should be able to match or beat this at 1/3 the cost.

    Also if you compare opportunity cost of a projector based e-learning model against things like big unsustainable computer labs in rural areas, which is being done in some places, it isnt even close on the ROI side.

    I’ll stop there for now.  The point is that there are a lot of things with the SIAB to be cautious about, improve, and build systems around.  But done correctly, dollar for dollar, this is something that should be supported.  It will be the most efficient use of gov tax dollars in these countries to deliver education to the 38%* of kids out of school in subsaharan africa.

    Dan

    *usaid

  2. Dave

    Nice one. Two good examples of opportunity cost failures, and there are plenty more where they come from when it comes to aid spending choices.

    A small nuance though in how you are applying this. In the second example – improving educational outcomes requires investment in a number of different areas including teachers, books, buildings and also things like cash transfers for poor families, latrines for girls etc. although taken by itself one of these may seem more cost effective than another, at a certain point you will also need to invest in those with lower returns to make the whole system improve, and so you can’t really look at individual interventions alone.

    In the first example (and maybe even the second) the opportunity cost is not necessarily compared against doing something else, but instead against doing nothing. That’s because an organization that collects #SWEDOW is not that easily able to be persuaded to just give cash instead – because that might not be feasible if say collecting t-shirts is easier than collecting cash.So the question might well be is it better to get free t-shirts or no t-shirts. The answer to this is not as obvious as it seems. While I (like you) might not want to turn down free food, it is not without a cost in terms of my time, and also my wasted calories eating free unhealthy food I don’t like that much versus providing something delicious and nutritious myself. So even though I might be thankful for a free t-shirt 3 sizes too big, I might have ultimately been better off paying for or paying for something that I actually needed.

  3. Ian and Dan, thanks for adding some nuance. My examples were meant to be illustrative of the opportunity cost principle, rather than comprehensive analyses of SWEDOW or ICT4E.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s