All of JPHoughton's Comments + Replies

Agree a simple calculation as outlined wouldn't be hard.  That would effectively increase the cost-per-life-saved by 20%, say, which is noteworthy but not fundamentally changing things.

The real risk is the longer-term, hard-to-measure impacts which may hold back economic progress generally.  These are by definition hard to fit in to a cost-per-life saved calculation but that doesn't mean the impacts don't exist.  Knowing these risks exist and intervening anyway is a choice some donors will be comfortable with but others will not.

Thanks Aaron for this comment, and the longer one you made elsewhere.  

On the excerpt, I think there's anecdotal evidence for bullet points 1-3 in a few places.  GiveWell mention this in an old blogpost about people not buying subsidised nets:

Two different people told us in off-the-record conversations that they thought that this occurred because the mothers offered subsidized bednets believed that they would be able to acquire free nets at some other point. There have been periodic free ITN distributions in many sub-Saharan African countries ove

Thanks Lucas.  Agree that these may be reasons for 100% coverage to be a reasonable philanthropic target which would be unachievable through commercial means.  

Your first point includes the idea that some people may not purchase ITNs because they expect to be given them for free.  This reinforces the idea in the essay that there is an extra cost to such distributions as nobody can make a living from selling nets in areas where people think the price should be zero.

I don't have a specific intervention/opportunity in mind for the scenario where health spending is broken.  

I'm reminded of a survey of several poor countries which revealed many were not following best practice for treating complications in pregnancy and childbirth despite the treatments being cheap and well-known.  Digging into it showed there wasn't a single reason for this, so no single intervention would change things everywhere.  

If the underlying reality is locals make bad choices, as normal individuals and health practitioners and po... (read more)

Agree with your outline of first-order impacts.  My concern is with wider consequences not included in that view.   The world is a complicated place and unintended side effects can be significant and negative.  This is even possible under the assumption that such distributions are viewed as a good thing by locals.  For example:

For citizens, the lesson may be "Random acts from people far away determine my circumstances, so there's little I can do to improve my lot".  That would hold back all economic development in a way that outstr... (read more)

I like "steelmanning", so thanks for sharing that.

Sin taxes & behavioural nudges seem to support my point rather than work against it.  The US banned alcohol and discovered many people kept on drinking anyway, so now limits itself to talking a good game and collecting the extra tax income.  Most health professionals are very clear that alcohol is bad, and many claim if it were invented today governments would ban it like so many other drugs. Yet no government I know says "We've looked at the scientific evidence, this is a clear example of peo... (read more)

I agree that humans are generally bad at risk management for low-likelihood-high-impact events.  I think this is not limited to black plague era peoples or those living with malaria today.  I believe it's a feature of human brains which scientific knowledge helps mitigate.  

Despite this, we generally let people make their own decisions about things which affect their health.    People smoke, drink, over-eat, fail to exercise,  etc but societies rarely force interventions on their own residents to prevent this.  If a benev... (read more)

I agree that in theory you can measure such economic impacts.  In practice I don't believe anybody is.  

If a body of practical knowledge on this point exists, then it would be straightforward to quantify the economic downsides of bednet distibutions and include it in the GiveWell calculation.  I am confident GiveWell are intellectually honest enough to do such a thing.  I believe the reason they haven't done this is that the information isn't out there.

When the information isn't out there, all you can do is make general/theoretical poin... (read more)

I think we both agree that bednets give a 17% reduction in mortality.  The question is what mortality rate to apply this 17% to.

GiveWell say 11.9.

I say 7.7.

Based on your points I thought you were either saying (a) 7.7, agreeing with me, (b) an adjusted version of 7.7, which I calculate to be 7.66.  Either way we're agreeing here.

I agree for those that get a net it's a good thing and mortality is reduced, and also agree mortality would increase again if no further nets appeared.  However this point isn't material given the volumes involved.  0.4m children at a rate of 6.4 and 11.6m at a rate of 7.7 is an average rate of 7.66 for the 12m children in total.

Even this small effect is diluted further when you consider the 6.4 rate only applies for 1. 74 years after the 2014-2016 distributions, so much lower when you measure mortality in 2019.

Previous AMF distributions fall into the level of rounding error in this counterfactual, which is why I'm saying an uplift from 7.7 to 11.9 is unreasonable.

Thanks for your comments.  Agree with your suggested edit - there's now two references to the Democratic Republic of Congo.  Note that for the 2021 v2 model this is AMF's total distributions as no other countries were expected to get distributions but regardless it's worth stating explicitly.

Net distributions cover the whole community, they are not targetted at just under-5s.  Using GiveWell's figures, 16% of the population is under 5.  Scaled up 1.8 people per net this suggests coverage for 1.4m * 16% * 1.8 = 0.4m young children.  ... (read more)

Thanks Lucas.  You are right that there is a version 3 of this spreadsheet and my post is based on version 2.  I originally took a copy of the spreadsheet in late August and v3 appeared in September.  

The only difference I can see between the two is the allocation of funding across countries (row 7), with the rest of the values agreeing once this is adjusted.  Still, this is a vital difference for my essay, and agree v3 gets the average cost-per-life-saved pretty close to the stated range.

GiveWell don't justify this change in their chan... (read more)

Charles

No need to apologise for a lack of response. Thanks for sharing your high-level view here even with the limitation that explaining that view is hard.  It's much better for me to have that comment than none at all, and I know the feeling of struggling to explain a view.

For info, I agree with the thrust of what you were saying about real-world constraints on distributions, though agree this is not what the post was actually about.

Lorenzo, thanks for diving in here to help out.  I can confirm this is indeed the source I used.

Charles, I didn't expect anybody to bother checking my source values.  I described them as "GiveWell's 2021 cost effectiveness calculation" which I think satisfies your criteria on context, age & purpose of source.  Googling that gives me Lorenzo's first link above.

Inspired by your comment I've now figured out how to add in line links for sources so that will perhaps help address this type of comment in future.

Thanks for the comments Matt.  I've adjusted and improved the post based on your input.

I was aware of this info and assumed everybody else would be too, so I just took it as read. However, I agree these points are not clear enough in the original post above.

I've now changed the heading to add the clarity that it only applies to non-RCT/"real world" distributions.  I've also inserted a sentence in the final paragraph to make it clear the RCTs do show such evidence and this is the basis for GiveWell's recommendation.  

Thanks Charles for your detailed response.

I agree with your central point that it's very hard to use statistics to prove anything.  In particular, you need a huge amount of data and there is lots of noise as the real world is not a clean &  tidy place.

For bednets, we do have a huge amount of data.  The World Malaria Report 2011, used in GiveWell's macro review, says 145 million bednets were distrubuted in sub-Saharan Africa in 2010 alone [1].  That's theoretical coverage for around 30% of the population [2].  This is a massive ... (read more)

Thanks Linch, interesting thoughts.

To clarify, my point is not just there's no direct empirical evidence of AMF's specific distributions saving lives.  My point is that there is no direct evidence of any  non-RCT/"real world" distributions saving lives.

Further, this is not because nobody is looking for such evidence.  GiveWell's macro review of the evidence suggests every time somebody has looked for evidence of non-RCT/"real world" distributions saving lives they've come up with nothing.

I agree with your summary of the GiveWell argument (st... (read more)