More interesting than what charity programmes achieve is what they don’t achieve

Much effort in the charity world goes into understanding what programmes achieve. Which is fine and well and good, but doesn’t indicate anything about whether funding a particular programme was any good.

Let’s take an example. In India, there is a programme which improves school attendance by paying parents when their children show up at school. It works: it costs about $1000 per child per year. Fantastic achievement. But here’s the rub. There are other programmes which improve school attendance for just $4 per child per year. So if a donor spends $1000 on the first programme (properly called a ‘conditional cash transfer’ programme), fully 24 children miss out. That is to say, the programme only achieves 4% of what it could have achieved; put another way, what it didn’t achieve was fully 96% of what it could have.

How is it possibly meaningful to look at what a programme did achieve without knowing what it could have achieved? It isn’t.

The example above, though extreme, isn’t isolated. Here’s a variant. Getting households in Kenya to put a few drops of chlorine in their drinking water dramatically reduces incidences of diarrhea. Population Services International had been selling bottles of chlorine for this purpose, and had great success. Fantastic – nice achievement. But a randomised control trial found that halving the cost of chlorine increased usage from 5% of households to 10%, and that giving it for free increased usage right up to 70%[i]. So whatever Population Services International had been achieving with donations of, say, $1000, it’s not hard to imagine that they could have achieved a whole lot more if they’d dropped the price.

This analysis really comes alive when we turn to corporate giving. Why? Because a company has many, many options of how to spend, say, £1000. It could give it to Charity A or Charity B, one of which might be better than the other. Or it could use it to ‘buy’ the time of some of its staff to do voluntary work. That can be massively influential: the International Finance Facility for Immunisation (IFFIm), a complicated bond issue which Goldman Sachs structured on a pro bono basis, is thought to have saved half a billion lives. I’d happily wager that’s more than any of its charitable programmes will ever achieve. Or put it another way. If the company creates some ‘bad’, such as carbon emissions, it could put that £1000 towards reducing its emissions, towards developing technologies which reduce emissions, or research into how to reduce the ‘bad’. Or towards developing or delivering products or services which are ‘good’ socially or environmentally: subsidising delivery to the poor, subsidising R&D of products or services which wouldn’t be economical on their own… or other projects which use the full gamut of the company’s unique skills and assets. “Money is the least valuable social change asset” said Kurt Hoffman, then Director of the Shell Foundation, and companies have many more assets at their disposal.

So yes, it’s great that your programme achieved what it did. But I – and I dare say those 24 children – are interested in what else it could have achieved. We therefore also need to analyse the missed opportunities.


[i] From More Than Good Intentions, by Dean Karlan and Jacob Appel

This entry was posted in Impact & evaluation and tagged , , , , , , , . Bookmark the permalink.

4 Responses to More interesting than what charity programmes achieve is what they don’t achieve

  1. Pingback: MoneyBox Live: good questions to ask | Giving Evidence

  2. Pingback: MoneyBox Live: good questions to ask | Caroline Fiennes @carolinefiennes

  3. Pingback: Designing the poor out of poverty in Myanmar | East Asia

  4. Pingback: Selfies, charities, and the importance of critical analysis | Patching Reality

Leave a Reply

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

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

Facebook photo

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

Connecting to %s