Why saving lives is complicated (and costly)
My own personal journey of effectiveness-minded giving began with the UK based nonprofit Giving What We Can, whose members pledge to give 10% of their lifetime earnings to charity.
On a perusal of their website, you'll find a tool into which you can enter your salary, which then neatly pops out an estimation on how many human lives you have the potential to save were you to take their pledge.
Now, admittedly, at the time I was working as a full-time jobbing musician - not exactly “raking it in”. But I was nonetheless disappointed to learn that even by donating 10% of my income to charity - something I considered to be a substantial commitment - I would annually be saving not 1000’s, not 100’s, nor even 10’s of lives. I could expect to save one.
It is perhaps a necessary caveat at this juncture to acknowledge that the concept of saving lives is somewhat a western interpretation of doing good, and that altruism is of course not limited to this metric. However for the sake of argument and consistency, I’ll stick with it throughout this piece.
So why my surprise?
The honest truth naivety.
Having grown up in privileged surroundings in a developed country, without really digging into the vastly complex area that is international development, it can be tempting to imagine that helping people on the other side of the world is easy. To imagine that their needs are uncomplicated and inexpensive, and so all that's needed is to throw money at the problems they encounter. Right? Not so, as Nick Jenkins (Moonpig founder) discovered himself; seen here speaking at our forum last year.
The biases that I held were likely influenced by the marketing of some of the more established charities, whose campaigns might feature messages along the lines of “for the price of one coffee, you could save a life in Africa.” Whilst such organisations may be doing incredibly important work, and could perhaps be forgiven for using messaging such as this in order to attract donors, it nonetheless led to a wildly inflated sense of confidence in me. And I imagine I wasn’t alone in that.
So why is the estimated cost of saving a life so much higher than the price of a cappuccino?
Firstly, there is the simple truth that doing good and helping people - especially at scale - is difficult. The myriad of challenges faced in low-income countries are a result of a complex mix of historical, political, structural, environmental and economic factors.
Another reason for our misconceptions about the cost of social impact might be our tendency to conflate outputs and outcomes. The reality is that giving a child a school uniform does not guarantee them a good education, any more than immunisation against a fatal disease equates to a long and healthy life. Finding out how to achieve the ultimate outcomes we care about is a learning process that is far from complete. More on this further down.
Thirdly, we have a tendency not to consider the alternative. However, in order to achieve a true estimation of our contributions, we must consider impact counterfactually. What is the marginal impact of our donation compared against the scenario in which we make no donation?
Let’s take the example of anti-malarial bed nets: the cost of one net is $5. Does this mean that through a donation of this amount you could save a life?
Of course not. Since this is a preventative measure, we can’t know for sure whether the recipient of the net would have contracted malaria without the net, let alone lost their life to it.
The situation is further complicated when we consider the idea that a donation to a particular organisation may in reality just be a “reshuffling of the pack.” That is to say, by funding the implementation of one intervention, are we by default directing resources away from another? Or - international development being the crowded space that is is - what’s to say that the work you're supporting wouldn’t have been carried out by a different organisation anyway - perhaps even with more efficiency?
Finally, there is so much more to effectiveness than simply the cost of the materials involved in an intervention. Quoting Peter Singer from his book The Life You Can Save “oral rehydration treatment for diarrhea may cost only a few cents, but it costs money to get it to each home and village so that it will be available when a child needs it, and to educate families in how to use it.”
So then, what is the ACTUAL cost of saving a life?
Charity evaluator GiveWell’s current estimation for the cost per life saved by Against Malaria Foundation - one of it’s top recommended charities - is $7,500. It is essential to state that this is simply a best (albeit well educated) guess. GiveWell are transparent about this fact, and go into much depth on why cost-effectiveness is so difficult to measure, and why we should be wary of attributing too much weight to these figures when considering our giving options.
$7,500 might seem like a high figure, but, it's important to put it into perspective: consider that in the UK, the National Health Services consider it cost-effective to spend £20,000-£30,000 in order to gain one year of healthy life. Even thinking conservatively and interpreting “saving a life” as adding another 30 years of good health to a person's life, with some quick number crunching we can see how these figures suggest that it is potentially over 100 times less costly to save lives in low-income countries than in the western world.
The good news
The good news is that as smart donors, we can help make it easier to save a life. As mentioned above, even a well-thought out social program may not lead to the ultimate desired outcomes (some can even cause harm), because in so many areas, we simply don’t know what works and what doesn’t, or do not have enough evidence to back our assumptions. But with research, evidence collection and thoughtful analysis, we can find out.
For example, organisations such as J-PAL and IPA work with charities to conduct Randomised Controlled Trials (RCT's) that aim to work out the true causal link between social programs and outcomes. For example, an RCT might compare malaria inflicted fatality rates over a given time, comparing regions where bed nets have been distributed by a particular organisation to areas where they haven’t. The result would be a better estimate of the counterfactual impact of that particular programme.
It is important to state that RCTs are by no means a silver bullet, and there's been lots of interesting and valuable discussion in how to measure the impact of interventions that are hard (if possible at all) to measure this way, such as policy change or campaigning, or funding innovation. In any case seeking more evidence and making use of the existing evidence is highly valuable.
By supporting evidence-driven and transparent organisations that prioritise trials and evaluation, you can help increase and fine-tune knowledge within international development. As such, you have the opportunity to have impact reaching far beyond the immediate good achieved by funding effective interventions; you will be helping figure out what works, and so ultimately be bringing down the cost of saving lives, making it easier and easier for others who want to give back to do so strategically and impactfully.
Originally published on 8 May 2018