Our Trendy Ideological Narcissism doesn’t help Africa

Ok so it doesnt quite roll of the tongue as the Invisible Children campaign!

I’m not going to do what countless others have done already and rip the piss out of Invisible Children, their Kony2012 Campaign and the obvious dodginess about  their methods/operations. Although I’ll let this little musical number (that they came out with back in 2006) speak for itself. It is simply breathtaking in sheer zaniness (especially about 2 minutes into the video, truly bizarre). On the main however, I find it hard to feel too indignant about them, as they are not the problem itself but only a symptom of a broader set of cultural idiosyncrasies. A fact which is exemplified in the success of this slick PR campaign-y approach which straddles both politicians and celebrities alike, in a bid to ‘change the world for the better’ where no one has to make the tough calls.  The details and nitty gritty is proudly de-emphasized.

But as I said, KONY2012 deconstruction has been done to death by now and it would be tedious in the extreme, to indulge in that exercise again. Instead what I want to focus on is the work of a group of people who are as passionate and interested in tackling the various issues surrounding Africa (& the developing world); but with an empirically informed approach. An approach which borrows from evidence based medicine and employs Randomized Control Trials to answer questions of whether specific policy interventions (on poverty) actually deliver successful results on the ground.

While it is laudable that people are willing to help out in terms of their money and time, the real action is in the quest to figure out what actually works and (just as importantly) what doesn’t work in that context. Frankly, being happy with ‘doing something instead nothing’ strikes me to very ambiguous and unambitious mindset. And it has always perplexed me as to why this mindset is so pervasive in society. Though Adam Curtis produced a short film which has a stab at explaining why this kind of thinking that permeates through popular thinking on Africa and the world’s poor

The failings of the Aid Debate

Meanwhile, in slightly less bonkers circles (not involving Invisible Children), the debate rages on whether Aid to Poor countries actually works or not. In the pro-Aid camp are people like Jeffrey Sachs, Bob Geldof and Bono (The Live-Aid crowd) ; who claim that it is the responsibility of the rich nations to pledge money towards helping build and rebuild the poorest in the world as they face some of the harshest challenges on the planet. They believe that the world’s poor simply CANNOT be expected to meet those natural obstacles and come up to speed with the developed world.

In the Aid-sceptic camp are people like Bill Easterly and Dambisa Moyo; who claim that International Aid actually makes the situation worse for Africa’s poor,  as it empowers corrupt leaders, creates this ‘Desperate Africa’ image which means people are hesitant in investing into Africa thereby reducing its prospects of progress. Moreover they argue that if the free-market system was allowed to operate, it would be much more effective in tackling the problems of poverty, healthcare, nutrition and education.

Both sides have a semblance of validity in their arguments. However, once you go through their work, you’d be hardpressed not to notice that both sides are quite fond of making grandiose claims about the challenges faced by the poor. Also, the definition of what works seem to vary from person to person and argument to argument. It is remarkable to see, how certain they seem to be, about the assumptions they’v made about the situation and the prescriptions they offer, to address the endless challenges faced. Problem is,  most of them, tend to be high on anecdotes and rhetoric, but low on hard evidence. Sure, a smattering of statistics is ever-present in highlighting the various issues of poverty, malnutrition etc, but there is no real way of knowing whether these approaches/policy recommendations are grounded in any  good quality evidence- something  which frames these problems a scientific manner.

You could say, that since they are dealing with ‘macro-issues’ and it is unreasonable to expect them to frame/discuss these in  micro-economic terms  – and of course that is a legitimate point to raise at first glance. But, what inevitably ends up happening, (as I hope to to show a little later on)  is that their ‘macro’ prescriptions for the global economy perfectly map onto their ‘micro’ prescriptions on single challenge issues like Malaria prevention Strategy; where, this same debate is rehashed – and therein lies the problem.

What are Randomized Control Trials (RCTs)?

RCTs form the heart of evidence based medicine. There is a hollywood notion of medicine that has people in labs, tinkering around with chemicals and coming up with novel formulas to cure diseases. In reaility it could not be further from what actually happens. And there is a very good reason for it also. Any drug or treatment has to stand the test of trials which are basically experiments of sorting out the signal from the noise. Making sure that alternative explanations (or confounding variables) are accounted for, before we give the green light for a drug or a treatment as a viable choice for a better  health outcome. What is repeatedly found, is that patient responses to interventions, are complex matters, and it requires a careful systematic approaches to sort out the causes-effect relationships – or lack thereof. That’s why people selected in trials, are randomized; which is a fancy way of saying that the study is not biased as a result of selecting different groups of people.

For example, if your running a trial of a drug for a particular medical condition, and you select 2 groups; giving 1 of them the new drug and the other, the control (which could either be a placebo or other alternatives depending on the context of treatments available for said condition). Suppose then, that the first group is made up of younger people who recieved the new drug (note: just for argument sake, assume younger people get healthier quicker, which may not necessarily be the case all the time); and the 2nd group, comprising of older people, received the ‘control’. I dont think it would be difficult to deduce that the younger set of people are likely to get better quicker anyways and to attribute the success of their recovery solely to the drug/intervention, just on the basis of this study, would be erroneous. The design of this study would  therefore be termed as biased towards producing a positive result for the drug and lacking in proper ‘Randomization’.

RCTs are basically a way to account for such inherent systematic challenges in data aquisition which can contiminate the cause-effect analysis in order to iron out what assumptions have sufficient grounding in evidence –  to be considered for further scaling up. They are very effective methodological tool for clearing out the clutter and systematic flaws which can lead spurious false positives or negatives results for a given hypothesis. And conceptually, they do not require a drug or a pill to study these cause-effect cases.

An organisation called Abdul Latif Jameel Poverty Action Lab (or J-PAL) employs this tool in addressing the myriad issues linked to poverty and development. Many people (like Sachs and Easterly) have ideas as to how we ought to proceed, but in reality, none them have really demonstrated the willingness to test them out in conditions where the clutter and noise is accounted for. Abhijit Banerjee and Esther Duflo who co-founded J-PAL have also co-authored the book called Poor Economics (Its in the Quackonomics Bookstore, among other fine selections for your reading pleasure) which looks at this very problem. Another one called More than Good Intentions (authored by Jacob Appell and Dean Carlan) also looks at similar issues compliment the work of Banerjee and Duflo. Both advocate the deployment of a rigourous Evidence-based anaylsis of development and poverty related issues before actions/interventions are scaled up. The fact that mainstream political and economic policy thinking has been slow in incorporating this approach is more of a comment on their way of working, than any difficulty associated with costs, expertise or reliability of RCTS.

So lets move on to some concrete and substantive examples of their work.

Pricing Insecticide-treated Bednets for Malaria Prevention

Insecticidal Bed-nets to sleep under at night (via ensnewswire.com)

Malaria is one of biggest life-threatening diseases  in Sub-Saharan Africa. It was responsible for the estimated deaths of a 1 million people each year (mostly in Sub-Saharan Africa). More pertinently, it is a preventable disease which makes the current situation even more tragic. Whats more, there are really well established solutions (in the context of Africa) for Malaria prevention; like the use Insecticidal Bed-nets which have been shown to not only protect people sleeping under them but to also reduce the risk for the rest of the population. Pretty much everyone is in agreement that there are huge benefits to be had from the widespread adoption of these bednets, and the need for comprehensive subsidization and distribution.

But what is the best way to achieve that optimum distribution. And how do we price them in order to ensure maximum effectiveness? Should we give them away for free? Because the benefit is likely to be far higher than any cost of the bed-nets themselves. Or should put a price on them? because giving them away for free would mean people wont value them and use them for other things.

Not surprisingly, those who want the bednets to be given away for free are the  Jeffrey Sachses of this world and those who want to put a price on it and let the market decide are the Bill Easterlys  and the Dambisa Moyos. The Sachs’ argument is that the productive windfall from a malaria free population would comfortably override the cost of providing the bed-nets for free. And if we start to price them, people would not be able to afford them, and the demand would fall rapidly which will lead to a decrease in their use. The Bill Easterly and Dambisa Moyo crowd, on the other hand, feel that if they are given away for free, people will not value them, use it for other stuff like fishing or as wedding veils (they provide some anecdotes to that effect) and other things, not for sleeping under them – thereby defeating the purpose of the bednets themselves to begin with. Again, both parties, like the AID debate, make confident and conflicting claims as to what will happen and they make policy recommendations based on that – and these carry alot of sway with influential people. So what happens when people test out these assumptions?

Pascaline Dupas and Jessica Cohen, are two researchers who did exactly that. They decided to evaluate a bednet delivery organisation called Together Against Malaria (TAMTAM) in Kenya and how different ways of subsidizing affected the purchasing of the bednets.In 2006, they looked at around 20 prenatal clinics in Kenya with 20,000 pregnant women (hence access to 20,000 families) . Then, they randomly selected 16 clinics, where the nets were distributed at varying rates. Moreover, to check whether (and how) these nets were being used, they sent people/supervisors on unannounced visits to check whether these nets were being used for their purpose (which is easy enough to do). What they found was, that there was indeed a correlation between pricing the nets and their acquisition. The demand did indeed fall, and thus concerns raised by the Sachs camp were indeed justified in the data. Moreover (in a further torpedoing of Easterly camp) the data showed that people were NOT ‘more likely’ to use the nets for other things, if they had bought them for free compared if they paid money for it.

Predictably the Sachs camp were very quick to point to studies such as these as vindication for their ideas. I’m not too sure about this truimphalism. I think the main lesson to be learnt here is, that assertions without a good evidence-base are not a good idea – and that is something we cannot entirely exonerate the Sachs crowd of either. What these studies also showed was that once people got these bednets for free, they were more likely to buy them the following year and so they were NOT ‘getting used to free things’; rather they were getting used to the NETS.

Immunizing the Poor

Randomistas: Banerjee (L) and Duflo (R) authors of Poor Economics (via mit.edu)

Leaving aside the nutty anti-Vaccine crowd, it is a well established fact, that Vaccines have been one of the most successful public health interventions of all time. It is easy to forget – especially in the oasis of prosperity that we inhabit – how certain diseases like polio, mumps, measles simply don’t kill children anymore. The simple fact of our world is, that the richer we are (and this applies internationally as well as within country); the more,  our basic needs and necessities,  are taken care of. So when we get a little uppity and start pointing out “how irrational these people are, for  not bothering to immunize their children or boiling their water?”; we forget that all of these decisions are basically taken out of our hands. Immunizations are easy-to-access and compulsory in the rich world, and the water we drink comes out clean from the tap. I don’t think that none of us are in a position to play the rationality card here.

So how do we approach the problem of low vaccination rates in the developing world. We need to first figure out why we think immunization rates are so low? Is it because of a supply-deficit? Is it because of ‘health professional’ absentee-sim? Is it because the cost of immunising a child is too high for some people? Is it because there are some deep-rooted local cultural reason why people are refusing to vaccinate their children? Or Is it simply because people just cannot get round to having their kids immunized? How do we sort these questions out because knowing which of these factors are more important than others is key to finding out what policy intervention will best optimise the vaccination rates in poor areas.

Around 2004-2007, Esther Duflo and Abhijeet Banerjee et al.(pictured above) decided to team up with an NGO in Rajisthan (India) attempted to answer this very question. Immunization rates there were around a paltry 6% at the time. What they did was to select random villages in the district of Udaipur, and proceeded to compare and contrast different incentives with the vaccination schedule while at the same time improving the reliability of healthcare services to ensure supply. Obviously an increase in immunization rates was to expected. So to compare and contrast, they offered incentives for vaccinations in some of the villages whereas others followed the conventional method of simply offering immunization services. In this case, the incentive was a 1kg of Lentils, to the families. who were getting their kids vaccinated. (Does not have to be that in all cases)

Now some people find this OUTRAGEOUS, claiming that this is tantamount to bribing these people to do the right thing by their kids and that is just MORALLY UNACCEPTABLE – forgetting of course, that we’r essentially bribed to do the same thing ourselves in the West also.  They found that the immunization rates in those districts (where the provided lentils) rose to an astonishing 38% whereas those without the lentil only manage to rise to 16% (the rise is attributed to improvement in the supply of health care services, which turned out to be a valid concern as expected). So it basically vindicated the incentive view of immunization that, people just had to be compensated for the cost that they were incurring for having their kids immunised. Its not that they were culturally opposed to vaccines, just that they could not get round to doing it. It was – in the end – a rearrangement in their ‘to-do’ list and a little prompting was all it took to get alot of people to get their children vaccinated.

Perhaps, the most astonishing part of this was that the introduction of lentils actually drove down the costs of the program. Previously, the average cost for immunization of a child was $55.83. With the lentils incentive, (even though they’r were cost associated with purchasing the lentils) the average cost of immunization of child fell to  $27.94 – almost halved. And it begins to make sense (after the fact) because, while prior to the lentil incentive, people were not showing up, but the health care providers (nurses) had to be paid a fixed to immunize a smaller amount of children. With the lentils, more people were coming through the turnstiles so to speak, which meant that the nurses in the same time were able to immunize many more kids in the same time. That is an astonishingly counter-intuitive way of cost-saving, which can really come from empirically evaluating the situation. And that is not even include the monumental public health benefit, vaccines have on the whole of the population long term. This doesnt mean that we must now provide lentils in every case, but it does validate the effectiveness of using incentives to boost immunisation rates. And that is something empirically confirmed – atleast in Rajisthan, India.

The trouble with the ‘White-Guilt’ Brigade

I hope I’v done more than enough to convince you why I think Kony2012 is bizarre campaign which is out of touch with the realities in Africa at a time when there are genuinely exciting things like JPAL going on. However one thing that  troubled me was part of the critical response to that the campaign got from some quarters. Many were quick to jump for the “White-man’s burden” line on this which – beyond knee-jerk – seemed to me to be a bit of an outdated response to what is increasingly a [pretentious word alert!] globalized narrative. By-gone thinking continues to exist in the West, but I’m not sure any input from the West should  necessarily be saddled with that label as articles like this and this, are keen to stress. I found this excerpt to be particularly interesting

 “African solutions for African problems”isn’t just a State Department slogan, and it isn’t about promoting African leadership, although that’s certainly important. Africans are already leaders. There are many reasons for Africa’s amazing rise over the last ten years, but one of the biggest has been African leadership. It’s not a coincidence that the 200 years of Western leadership in Africa were some of the continent’s worst. Africans have proven time and again that they’re better at fixing African problems

There are a couple of problems with such polemics. Firstly who speaks for the African poor? I think there is a high probability (note: probability does not mean everyone) the people typing up these articulate ripostes are quite privileged themselves, and for all intents and purposes NOT in the same boat as (most of ) those they seek to represent – who lest we forget are, still, quite poor people. This is not to say that you reject their viewpoint, outright. Of course not! But I think we have to be skeptical of the viewpoint of people of privilege (within the developing world) when or if they start to drone on about how they speak for the rest of ‘their people’ – especially if they offer less than full disclosure about their own privilege. Dambisa Moyo herself, possibly falls in that camp a little bit as well. Also, I think there is something very dubious about the point of view, that somehow if your white, you’r not allowed to have an opinion on Africa. Its a complete non sequitur. Ultimately – as with the aid debate –  it boils down to the evidence you can summon in support of the arguments and the tenuous conclusions that you draw from them, and whether they stand up to critical scrutiny. Rhetoric and anectdotes are just that – rhetoric and anecdotes. They are not Data.

A more pernicious form of this kind of thinking can seriously endanger many people. My last post was about AIDS-Denialism; the conspiracy theory which states “that there is no link between HIV and AIDS and this is just a con by Western Pharmaceuticals to keep Africans sick so that they can sell them their anti-retroviral drugs”. Tragically, the South African government under the stewardship of Thabo Mbeki, was a subscriber to this rotten Denialism.

And alot of it, was expressed in the “we will not accept this western science imposed on us” attitude. They denied their population (which has the highest rates of HIV infected people) vital life-saving anti-retrovirals, resulting in what is estimated in the deaths of 300,000 preventable deaths. That is a phenomenal price to pay for an indulgent attitude (which may have grounding in legitimate historical grievance). But 300,000 people dying because of ‘politicising’ what was a scientific question and should have had a scientific approach to resolve it, is unimaginably tragic and disgraceful. Thankfully (if very belatedly) the South African government under Jacob Zuma has had a change of heart over Mbeki’s policies, and are now working to have ARVs available for South Africans.

Perhaps the fundamental difference between our decadent lives in the West (or the privileged world), is that we have the safety nets which shield us from the consequence of bad ideas, and thus it allows us to pander to our preconceived notions/ideas/indulgences of the world, without incurring much of a downside. The poor don’t have that luxury and it is therefore even more important for Critical thinking and systematic empiricism to take center stage. We need to make sure that subject our ideas to rigorous scrutiny before we advocate for them to be scaled up. Because, the lives of the poor are a great deal more complex than we give them credit for.

This entry was posted in Africa, AIDS, Development, Economics, Politics, Poverty, RCTs, Statistics, Vaccines and tagged , , , , , , , , , , . Bookmark the permalink.

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