This blog aims (among other things) to explain the methods we employ in our research at Giving What We Can. In that spirit, we’d like to start by explaining an important way in which we differ from many charity evaluators.
In the same way that there are many persistent myths about aid, there are also persistent myths about how to evaluate charities. One of the most prevalent, and most damaging, of these is that we should just look at the amount the charity spends on advertising, or salaries, or broader administration in order to decide which charities to donate to – supporting those which spend most of their revenue directly on the programs they operate.
The grain of truth in this is that, if we assume the program has a certain value, every dollar spent on admin is indeed a dollar not spent on the program. But the value of programs varies dramatically, and dollars not spent directly on programs can be used to improve the program, or raise even more money: we need to take these into account.
Firstly, the quality of the program is not constant. Some programs are thousands of times more effective than others. If one charity A spends slightly more on administration than charity B, but operates a program that's ten times as effective, then overall it will still be much more effective. As overall effectiveness is what we care about, it would be misleading to focus on overhead costs.
The variance in charity effectiveness means there's great room for improvement. Improving the efficiency of its operations means a charity can achieve much more - but this requires spending money on gathering data, analysing it, producing a new strategy and implementing it. This is extremely beneficial, so it doesn't make sense to punish the charity for it - and yet that's exactly what the 'overhead' metric does.
Secondly, money raising is a valuable activity! If my $100 donation funds a media campaign that raises $200, then I've helped twice as many people as if my money was spent directly. As advertising generally raises more money than it costs, it's no bad thing for a charity to be undertaking. Obviously some of the money has to be spent on the program, but as long as investments in advertising is yielding more money to spend on interventions, it's a very good thing.
So while there is a grain of truth to the anti-administration argument, there are powerful considerations on the other side. The solution is to look at an overall metric, like QALYs/ dollar. Since ultimately we want to help as many people as possible, and only care about the charities as a means to this end, it makes sense to look at QALYs instead. QALYs/dollar takes into account both the internal structure of the charity, their relationship with donors and the impact their intervention has. It allows us to avoid the messy question of judging the organisational structure of charities, and focus instead on helping as many people as we can.
Read more about using QALYs here.