Home

Friday, August 23, 2019

Uncovering Causality III: Regression Discontinuity Designs

One of the nice things about running a blog is that you get to see which of your posts struck a chord and got more views than you’d anticipated. One of the less nice things is that you’re also able to see which posts went down like a lead balloon. Perhaps unsurprisingly, into this latter category falls my Uncovering Causality posts where I try to explain various statistical techniques in a way that’s easy to understand. Who would have thought Econometrics would be so unappealing?!

Not to be deterred, I’ve decided to continue my Uncovering Causality series because a) I think making statistics accessible is important and b) the nerd in me enjoys explaining statistics in a way that's intuitive. So, without further ado, here is a post on Regression Discontinuity Designs (RDDs) – one of the cleverest and most interesting statistical techniques you can come across.

The Selection Problem


Before understanding RDDs, we need to understand the selection problem they seek to address. Suppose we wish to understand the average ‘treatment effect’ of going to hospital – do hospital visits make us better or worse? One way we could do this is to compare the health status of people that have been recently visited hospital vs. those who haven’t. We can rank health status on a scale of 0-9, where 9 = perfect health and 0 = death’s door. Suppose we observe the following:


In words, the group of people that hasn’t been to hospital is healthier than the group that has. From this, can we conclude that hospitals are harmful to our health?

Thankfully, the answer is no. Almost de facto, people who visit hospital are less healthy than those who don’t, so it is unfair to compare these groups ex post. In an ideal world, we would want to compare the health status of the treated group (4.4) to what it would have been had they not been treated. Unfortunately, we cannot observe this ‘counterfactual’ – so we have to resort to clever statistical techniques to tease out what it might have been.

Overcoming the Selection Problem


The best way to do this is to ensure that treatment is randomised. Randomised Controlled Trials offer one option: by randomly deciding who gets treatment, we ensure that both treatment and control groups are as similar as possible ex ante, allowing them to be reliably compared ex post. However, for practical and ethical reasons, RCTs are not always viable (the ethical issues with randomising hospital treatment hardly need stating). Fortunately, there are other ways to generate randomness outside of an experimental setting.

Regression Discontinuity Designs


In October, 000s of 10-year-old kids will line up outside my old school to take the 11+: a gruelling 3-hour exam based on non-verbal and verbal reasoning. If they score above a certain mark, they are admitted to grammar school; if not, they go to comprehensive school[1] (for simplicity, assume there are no private schools in this world). To estimate the treatment effect of selective education, we might plot children’s 11+ mark against the number of UCAS points[2] they achieve upon completing secondary school.


From this set-up, how can we calculate the treatment effect of going to grammar school? It would be unfair to simply compare the UCAS scores of kids who got in to grammar school vs. those who didn’t. After all, kids who get into grammar school at age 11 are likely more intelligent and affluent to begin with – a classic selection problem.

However, what about the kids who just missed out vs. the kids who just about made it? Presumably these kids are pretty similar. Whether they got in or not may have depended on whether the right questions came up, whether they slept well the night before or whether they guessed correctly on a number of multiple-choice questions. These factors are largely random: enabling us to identify a ‘treatment effect’ of selective education – at least for the kids that scored near the cut-off.[3]


Early Childhood Development


Early childhood development matters – and assessing the effect of postnatal interventions is vital in informing effective health policy. However, in doing this, it is unfair to compare outcomes of babies that received postnatal treatment vs. those that didn’t – this is the classic healthcare selection problem outlined at the start.

However, what if there was an arbitrary cut-off that determined whether or not they got treated? For example, in Norway and Chile, new-borns below a 1500g cut-off get given additional respiratory and surfactant treatment, while those above it do not. Using the assumption that a 1490g baby is virtually identical to a 1510g baby, Bharadwaj et al. exploit this cut-off to estimate the ‘treatment’ effect of additional post-natal care. The results are displayed below: in both Chile and Norway, babies just below the 1500g cut-off perform significantly better in Maths exams that take place almost a decade later. 


Gaming the System


The biggest risk to RDDs is people gaming the system. The more people can exert influence on which side of the cut-off they fall, the less we can assume that treatment is random. While mother’s cannot meaningfully control their child's birthweight to ensure it falls below a certain cut-off, they might nonetheless be able to control which birthweight is recorded. Doctors in certain hospitals may be easier to convince to record a false measurement to ensure the baby gets extra care. If this happens, kids born just below the cut-off are no longer born in a random sub-section of hospitals – compromising our ability to infer a reliable treatment effect.

Despite this flaw, RDDs provide a clever way to untangle causality outside of an expensive experimental context. Their surge in popularity has no doubt contributed to the recent ‘credibility revolution’ in empirical economics. Valid RDDs are not easy to come by, however – like instrumental variables, identifying a good ‘cut-off’ requires imagination and ingenuity. This is partly why I like them: they force the researcher to shut the Econometrics textbook and think creatively about the world around them.  

Word count: 993 words

References

  • Bharadwaj et al. (2013), ‘Early Life Health Interventions and Academic Achievement’

For an example of a really creative RDD, check out this paper by Melissa Dell about the long-run impact of forced mining systems in Peru and Bolivia: https://scholar.harvard.edu/files/dell/files/ecta8121_0.pdf


[1] For my non-British readers: grammar schools are selective state schools (i.e. you have to pass a test to get in); comprehensive schools are non-selective state schools
[2] Again, for my non-British readers: UCAS points are what our grades are converted to when we apply to University, sort of like a GPA
[3] In the jargon, this is known as a Local Average Treatment Effect, since the effect is only estimated on a certain sub-section of the population i.e. those who scored near the cut-off

Wednesday, August 14, 2019

The Problem of Overambitious Curricula

Learning Profiles


One of the most curious features of education outcomes in developing countries is the presence of strikingly flat learning profiles. What do I mean by this? Figure 1 provides a good illustration. The x-axis plots students’ grade progression through school; the y-axis plots the proportion of students able to perform a basic competency. In this case, the graph plots what percentage of children are able to perform simple division (i.e. dividing a large number by a one-digit number) in four developing countries: India, Kenya, Pakistan and Uganda. In Pakistan, around 10% of students can do basic division by Grade 3; around 40% by Grade 5 and around 70% by Grade 7.

Figure 1: Learning Profiles in Four Developing Countries


This is strange – as Maryam Akmal, the author of this graph notes: we can’t really tell when division is taught in school. This flies in the face of what we tend to think about learning: that it occurs in stages, whereby mastery of ‘core’ concepts is a prerequisite for moving on to more advanced material. For instance, you can’t learn optimisation without calculus; and you can’t learn calculus without having mastered basic division. In an ideal world, all students would gain mastery of core competencies at the time it is taught, giving rise to a ‘steeper’ learning profile as overlaid in Figure 2.

Figure 2: A ‘Healthier’ Learning Profile

Source: https://www.cgdev.org/blog/are-we-wrong-about-right-way-organize-schooling

The Problem of Over-Ambitious Curricula


There are many explanations that try to make sense of these curves. For me, the most compelling one is the ubiquity of unsuitable and over-ambitious curricula in developing countries.

In many places, national curricula largely mimic those of more industrialised nations: but are delivered in schools that are less well-equipped, by teachers who are less well-qualified, to students who are less well-fed, well-mentored and well-rested at home. As a result, many students are left behind the official curriculum and are instead forced to pick them up at a much later date – or to leave school having not learnt them at all.

Learning in a Foreign Language


One of the most damaging ways in which an over-ambitious curriculum manifests itself is through an unsuitable language of instruction. In many countries, local languages are replaced by English or French long before most students have mastered proficiency. This, I think, is one of the biggest challenges facing the Zanzibar education sector, where English replaces Swahili in Form I (when students are 12 years old). Students and teachers are forced to interact in a language they are not comfortable with, leading to high rates of failure in Form II exams and subsequent dropouts soon after. The gap in exam performance between government and private students also widens at this point – a fact I think can be explained by them being unfairly discriminated against in Form II exams, which are written in a language they are less likely to be familiar with that their private school counterparts.  

In light of these consequences, administrators occasionally revert back to local languages. 2 weeks ago, the Chief Minister of Punjab, Pakistan announced that Urdu would replace English as the medium of instruction for primary education. Millions of schoolchildren will now be taught in a language they and their teachers easily understand – a great success, and an example that other countries would do well to emulate.

The Politics of Curriculum Reform


Unfortunately, unsuitable and over-ambitious curricula do not come about by accident. As Paul Glewwe argues, many national curriculums were originally designed to produce a narrow elite to work in colonial bureaucracies, with no remedial education offered to those who fell behind. After independence, enrolment expanded rapidly but control of the curriculum remained firmly in elite hands. These elites were likely to prefer an education system targeted to the needs of their own children – who were better-equipped to cope with an advanced curriculum due to their better schools, higher socioeconomic status and access to extra-curricular tuition.

This tendency to maintain the status quo was also exacerbated by the uncomfortable nature of reform itself. Curriculum reviews are highly politicised and contentious – often viewed by opponents as cynical attempts to indoctrinate children into the ideals of the regime (for more on this, read my Despots Guide to Education). As a result, reform often resembles a Pandora’s Box – something that’s better off unopened, delayed and left for the next regime to deal with. These dynamics are evident in Zanzibar: in theory, a curriculum review is supposed to take place every 7 years; in reality, only 2 have taken place in my lifetime: one in 1996 and one in 2010.

This inertia means curricula quickly become outdated. Below is a depressing excerpt from a Primary IV textbook in Nigeria in 2019 – something which would have looked positively Stone Age even when I was in school.  

Figure 3: A Primary IV Textbook in Nigeria, 2019


Source: photo taken by @Kabira_Namit

Terra Incognita?


The politicised nature of curriculum reform also makes it a difficult topic to study. For obvious reasons just discussed, governments are reluctant to allow researchers to tinker with different curriculums in randomised experiments – unlike with free school meals, teacher performance grants and conditional cash transfers. Natural experiments are also difficult to come across: reforms tend to be rare and, when they do occur, they tend to happen nationwide and all-at-once, making it hard to identify a distinct treatment and control group.

As a result, the body of evidence we have to inform ‘what works’ in improving the curriculum is pretty thin, especially compared to other education inputs like teachers, learning materials and textbooks. In fact, of all the inputs that go into making kids learn, I think the curriculum is the one where there is the biggest gap between what we can empirically measure and what we intuitively deem important. There is no obvious way out of this predicament: though looking at the effects of ‘localised’ reforms such as the recent language-switch in Punjab, Pakistan may be a promising place to start.

Word count: 997 words

References


The section on learning profiles was inspired by a great blog post by Maryam Akmal of CGD, available here: https://www.cgdev.org/blog/are-we-wrong-about-right-way-organize-schooling

For more on the problem of over-ambitious curricula, read:
  • Glewwe et al. (2009), ‘Many Children Left Behind? Textbooks and Test Scores in Kenya’
  • Pritchett & Beatty (2012), ‘The Negative Consequences of Overambitious Curricula in Developing Countries’

Wednesday, August 7, 2019

My Favourite Non-Fiction Books

Below is a list of my favourite non-fiction books I’ve read over the years, roughly arranged by theme. This was supposed to be a Top 10 but quickly became a Top 15. My current list includes:
  1. Guns, Germs and Steel – Jared Diamond
  2. Why Nations Fail – Daron Acemoglu & James Robinson
  3. The Bottom Billion – Paul Collier
  4. The Anti-Politics Machine – James Ferguson
  5. Peaceland – Severine Autesserre
  6. Sapiens – Yuval Noah Harari
  7. The Better Angels of Our Nature – Steven Pinker
  8. My Friend the Mercenary – James Brabazon
  9. Thinking Fast and Slow – Daniel Kahneman
  10. The Black Swan – Nassim Nicholas Taleb
  11. The Undoing Project – Michael Lewis
  12. Seeing Like a State – James Scott
  13. States and Power in Africa – Jeffrey Herbst
  14. The Uninhabitable Earth: Life After Warming – David Wallace-Wells
  15. Homo Deus – Yuval Noah Harari
Brief synopses and commentaries of these books are provided below. I plan to update this list as and when I read other books worthy of a mention. I hope you find it useful!

Development

Guns, Germs and Steel – Jared Diamond & Why Nations Fail – Daron Acemoglu & James Robinson


What is the most important geographic and historical fact of human history?

For Jared Diamond, the answer is simple: it is the rise
and subsequent dominance of Europeans in the modern world. If this seems short-sighted and Euro-centric, consider for a moment the sheer scale of the European colonial project. In North America, British colonialists bullied and forcibly removed Native Americans; in Latin America, Spanish conquistadors eviscerated indigenous populations via unfamiliar weaponry and disease; in Africa, an entire continent was divided and divvied up by French, British, Belgian and German colonialists with breath-taking indifference; and in Australia, aboriginal Australians were literally exterminated by the British colonial regime. In each of these places, European colonialists imposed their language, religion, technology and institutions, almost all of which have persisted in some way up to the present day.

These facts give rise to one of the Really Big Questions of Development Economics: why did Europeans come to dominate the globe, and not Africans, Native Americans or Aboriginal Australians?

Until not long ago, a fairly common and uncontroversial response to this question would be to appeal to Europeans greater intelligence, or their stronger work-ethic instilled by their Judeo-Christian values. According to Diamond, the pervasiveness of this view was not due to there being any evidence that either of these things were true (there wasn’t), but because Economists, Historians and Geographers had collectively failed to put forward a convincing alternative explanation.

This changed with his publication of Guns, Germs & Steel in 1997. In it, he put forward a new explanation that emphasised the importance of historic geographic and ecological endowments in determining a nation’s wealth – not the underlying characteristics of its people.

For example, people in areas with an historic abundance of domesticable animals such as cows, oxen and pigs were able to transition from a hunter-gatherer to an agrarian society more quickly. Control of crops and livestock generated food surpluses, enabling people to specialise in activities other than subsistence farming. Increased specialisation brought about additional technological advancements, enabling early societies to support ever-larger populations. This, in turn, sufficiently incentivised the build-up of primitive state apparatus – a crucial precursor to modern industrial development.

Despite significant fanfare when the book was published, Guns, Germs and Steel left a number of unanswered questions. Geography is remarkably unchanging: yet human history is characterised by the rise and fall of different civilisations. The Aksum, The Angkor and The Mayans were all once among the most advanced societies of their time, yet their descendants now all reside in countries towards the bottom of the global income distribution. Why is this? The geography of these places has not changed in thousands of years; there must be another piece of the puzzle.

Acemoglu & Robinson largely pick up the baton of this question in Why Nations Fail. In it, they argue that institutions – the rules of the game of society – are the most crucial determinant of a country’s success. Institutions are deep-rooted but not immovable, which allows for the numerous ‘Reversal of Fortunes’ that punctuate human history. It also explains how two neighbouring cities with identical geographies and the same name – Nogales – can have such divergent economic, health and education outcomes today. The answer, they argue, lies in the the different institutional quality of the different countries they inhabit: Nogales Sonara lies in Mexico; Nogales Arizona lies in the United States.

Guns, Germs & Steel and Why Nations Fail are often viewed as two books antithetical to each other: you are either in the geography camp, or the institutions camp, but not in both. I think this is wrong – both books contain important insights and much can be gained from a synthesis of the two. For this reason, I include them together in my list.

The Bottom Billion – Paul Collier


I first read The Bottom Billion when I was 17 years old on a
voluntourism trip to Bolivia (I know). At the risk of sounding sentimental, this was the first Economics book I ever read in the first developing country I ever visited – at a time when I was still planning to study History at University. Reading this book in this setting made me confront questions I hadn’t given much thought to before: why are some countries so much poorer than other? And is there anything we can do about it? By and large, these remain the questions I find most interesting today.

Collier has his own answers to these questions which he lays out thematically in separate chapters. He argues that countries are poor because they fall into one or more of four development ‘traps’: the conflict trap, the natural resource trap, being landlocked with bad neighbours and the bad governance trap. Once there, a confluence of self-reinforcing factors makes escape especially difficult. The net result of this is a ‘Bottom Billion’ – a billion or so people that are poor; and stuck in countries that are going nowhere.

To remedy this, Collier proposes four solutions. First, the aid industry should take on more risk and increasingly concentrate itself in the most difficult and desperate environments. Second, appropriate military interventions (such as the British in Sierra Leone) should be encouraged, particularly to guarantee democratic governments against coups. Third, international charters are needed to encourage good governance and provide prototypes. Finally, trade policy in both developed and developing countries needs to encourage free trade and give preferential access to Bottom Billion exports.

Did this book save me from being a loquacious Arts student? Possibly. Are the methodologies Collier uses now out-dated? Probably. Has he said some pretty stupid stuff since? Definitely. But is this book still one of the most accessible and engaging introductions to Development Economics? Absolutely. Read it!

The Anti-Politics Machine – James Ferguson



The Anti-Politics Machine is an excruciating post-mortem of
an almost unbelievably ill-conceived World Bank project that took place in Lesotho from 1975-84. The project, dubbed the Thaba-Tseka Development project, sought to ‘modernise’ the Basotho economy by encouraging a switch to cash crops, introducing sustainable livestock practices and building roads to connect rural farmers to markets. The project was woefully unsuccessful: by 1984, crop output remained low, new livestock practices were completely ignored and agricultural exports remained largely unchanged. On top of this, the project’s tree-planting project was sabotaged, fences delineating the project area were torn up, and a project manager’s car was set on fire.

How did it go so wrong? Ferguson argues that it all stemmed from a fundamental misconception of the Basotho economy, epitomised by the project’s laughably inaccurate inception report. The report portrays Lesotho as a backwards, isolated, and predominantly agrarian society whose development had been thwarted by lack of reform efforts by both colonial and post-independence leaders.

About 5 minutes of research shows this to be completely false. Lesotho was not, and has never been, a predominantly agrarian society: at the time the report was written, less than 10% of household income came from farming. Instead, over 60% of Basotho men were employed in industrial jobs in neighbouring South Africa – which formed by far the biggest contribution to Basotho GDP. This reliance meant that Lesotho was – and always had been – deeply embedded in the regional economy, and not the aboriginal ‘island’ economy the report sought to portray.

This misrepresentation was no accident. A truthful portrayal of Lesotho’s situation would reveal that nearly all determinants of economic life in Lesotho relied on factors outside the nation’s borders. However, such a representation would leave no role for the World Bank. Instead, by emphasising the importance of agriculture and obsessing over government policy as the most crucial variable in Lesotho’s success, World Bank economists had carved a role for themselves as indispensable gatekeepers in Lesotho’s road to prosperity.

I find this book so interesting not least because I see similar dynamics at play in Zanzibar. Here, ‘underdevelopment’ often seems attributed to people doing things wrong – things that more learned, or more capable people simply would not do. Accordingly, government non-compliance is put down to a lack of understanding or capacity, rather as a strategic reaction by individual agents with wide-ranging and often incongruous incentives. The result is to fundamentally misconstrue the state as an apolitical entity – an ‘Anti-Politics Machine’. This reduces ‘development’ from a social and political problem to a largely technical problem – one in which consultancies, aid agencies and international organisations are perfectly poised to help solve.

Peaceland – Severine Autesserre


Why do aid and peace interventions so often go awry? James
Ferguson provides one answer to this question; Severine Autesserre provides another. Peacekeepers, aid workers and diplomats occupy a different social space to the places in which they work – a space she dubs ‘Peaceland’. This space is characterised by its own distinct beliefs, behaviours and rituals. Examples of these include socialising exclusively in expat circles, not speaking the local dialect, upping sticks every few years, favouring technical know-how over regional expertise and abiding by unnecessarily stringent security procedures. These practices often seem innocuous at face value – but together they create a distinct sense of ‘otherness’ that hinders contextual understanding and often exacerbates tensions between interveners and local populations.

I read this whilst in ‘Aidland’ in Zanzibar – not quite the same as ‘Peaceland’, granted, but a social space with striking resemblances. I read it with a mixture of intrigue and embarrassment as I realised I conformed to almost all the stereotypes of the boujie aid worker in Tanzania: not speaking Swahili ; drinking at the 6 Degrees and the Slow Leopard; subsisting on a diet of sushi and Huel . I don’t think there’s a problem with this per se – but when it comes at the expense of all other social and cultural engagements then it begins to produce the consequences that Autesserre so expertly outlines.

Since reading Peaceland I’ve definitely tried harder to escape the ex-pat bubble. I’ve joined a local football team; I’ve visited local colleagues’ homes and invited them to my own; I’ve doubled-down on my shaky Swahili. So yeh – I’m still a boujie aid worker – but at least now I’m trying harder.

Sapiens – Yuval Noah Harari


I remember getting the District line from Earls Court in 2015 and 
there being about 10 people in my carriage all with their noses in this book. While suit-clad commuters are hardly a representative sample, this book seemed to strike a chord with almost everyone: from University Professors to Love Island contestants. I think this is mostly to do with the book just being so damn readable: I remember buying this in Delhi airport and then having almost finished it by the time I touched down in Heathrow.

This is not marketed as a book about development, but in my mind it’s about human development in the really, really long-run. In charting the rise of the human race, Harari sets himself an even deeper question than Diamond, Acemoglu & Robinson: not just why Europeans came to dominate other humans, but why Homo Sapiens came to dominate other species.

Harari’s answer to this question rests not on our ability to stand upright, use tools or communicate via language, but on our unique ability to believe in things which don’t really exist – that is, things with no direct correspondence to the physical world. For example, chimpanzees are able to tell each other if there is an eagle circling them overhead – a very real phenomena in the physical world – but are unable to discuss abstract concepts like God, democracy or monkey rights.

Humans, of course, can imagine these things – and our capacity to do so enables our uniquely human ability to cooperate flexibly and in large numbers. Chimpanzees are able to cooperate flexibly but not in large numbers, as all cooperation must be underpinned by personal relationships. This sets an upper limit on chimpanzee troops of about 150 members.

Ants and bees can cooperate in large numbers but not flexibly. Their social relations are defined by their DNA – not by their shared beliefs. For worker ants to rise up and overthrow their queen, this would take millions of years of evolution. By contrast, humans in 18th century France were able to transition from a monarchy to a democracy virtually overnight, as their shared myths about the natural and just social order were quickly and violently rewritten.

All human accomplishments can eventually be traced to this ability for large-scale and flexible cooperation. The fruits of this are hard to overstate: the ability to walk on the moon; to eradicate deadly disease; to create weapons powerful enough to destroy civilisations. By the end, you are left to marvel at our voyage from insignificant apes in the corner of East Africa to the undisputed masters of Planet Earth. Very humbling indeed.

War & Conflict

The Better Angels of Our Nature – Steven Pinker


Like Sapiens, this was another book that littered bookshelves, conversations and coffee tables upon its release in 2011. Pinker’s central thesis is that – despite the seemingly incessant stream of bad news that fills our TV screens – the prevalence of violence has unambiguously declined throughout human history. This decrease has been universal in scope: wars have declined; genocides are less frequent; homicides have decreased; the acceptability of torture has fallen; sexual violence has decreased; even terrorism – so often perceived as the defining feature of the post-9/11 world – has remained at barely perceptible levels as far as the entire species is concerned.

Pinker identifies five historical forces that have facilitated this decline. First, the rise of the modern nation-state has enabled the monopolisation of the legitimate use of force, reducing the temptation for individuals to enact extra-judicial revenge. Second, the forces of globalisation have enabled the trading of goods over longer distances, bringing distant societies into contact with each other and making each, in the eyes of the other, “more valuable alive than dead”. Third, women have become more emancipated in both political and domestic spheres, and have tended to be more doveish in both their voting patterns and social behaviour. Four, the rise in literacy, mobility and mass media has led to the State being fundamentally reconceptualised as a protector against violence to a guarantor of human rights. Finally, Pinker points to the Escalation of Reason in exposing the sheer pointlessness of war and in reframing violence as a problem to be solved rather than a contest to be won.
  
Of course, there is a lot to disagree with. I think his basic argument is essentially correct – violence has declined over the course of history, and any far-reaching historical, political or social theory of human history has to somehow take this into account. However, what bothers me about Pinker – and indeed the New Optimist crowd in general – is the sleight of hand deployed in moving from the empirical claim that violence has declined to the subjective claim that the world is now more secure. It is perfectly consistent to claim that the world is both much less violent and much less secure – and hence hardly the stuff to be celebrating.

For example, our success in stymying ancient forms of violence does nothing to assure us against the relatively new threats posed by nuclear war, environment degradation, artificial intelligence and bio-hacking. If things go wrong now, they go really, really wrong – a fact that Pinker largely dances past in his follow-up book, Enlightenment Now. This makes me reluctant to join Pinker in his New Optimist camp – no matter how compelling his argument in The Better Angels of Our Nature may be.

My Friend the Mercenary – James Brabazon


My Friend the Mercenary is the captivating memoir of James
Brabazon, a war journalist who was the only person to film the LURD rebel group inside Liberia during the Second Liberian Civil War (1999 – 2003).

The first half of the memoir recounts Brabazon’s friendship with Nick du Toit – a South African mercenary-cum-arms dealer who Brabazon hires to be his bodyguard for his tour of Liberia. Together, the two make contact with LURD rebels at the Guinean border and join them on their subsequent march to Monrovia. Their journey is marred by severe illness, near-impenetrable rainforest and frequent encounters with government soldiers – some of which end in execution, torture and ritual cannibalism. Much like Marlow’s journey to the Heart of Darkness, Brabazon’s journey to Monrovia reflects a journey to the      depths of the human soul – and the depravity of what it’s capable of.

For all this vividness, Brabazon – to his credit – frequently cautions against the stereotype of the blood-thirsty African rebel. He makes an admirable attempt to form close bonds with the soldiers and allow their personal stories to echo through his pages. In doing so, Brabazon helps ‘humanise’ the conflict, helping the reader appreciate the sheer depth of circumstances that turns otherwise ordinary citizens into such desperate acts of violence.

After the Liberian War ends, du Toit invites Brabazon to film ‘regime change’ in Equatorial Guinea: an outrageous coup attempt financed by extremely shady British establishment figures including Mark Thatcher, Margaret Thatcher’s son. By an amazing fluke, Brabazon misses his flight to Equatorial Guinea and is forced to watch on the news as the plot is uncovered and defused. Du Toit gets arrested and accused of high treason, culminating to a life sentence at Black Beach prison, notoriously known as the most brutal prison in the world.

Brabazon began writing his memoir largely as a testimony to his friend and former bodyguard, whom he safely assumed he would never see again. However, in another extraordinary twist of fate, du Toit is released just 5 years into his sentence. This book concludes with the pairs’ rendezvous, with du Toit recounting further gruesome tales of survival in Black Beach.

By the end of all this, you aren’t really sure what to make of Brabazon. Despite his attempts otherwise, you can’t help but think that he is perpetuating negative African stereotypes – at least in the minds of his more lazy and prejudiced readers. He’s also a bit of a mercenary himself: barging into conflict zones uninvited to harvest footage for his award-winning documentaries. With this being said, Brabazon certainly has balls – what’s more, his work undoubtedly shines a light on conflicts which may otherwise be forgotten. Whatever the final judgment – I’d certainly like to go for a beer with him.

Risk, Uncertainty and Behavioural Economics

Thinking Fast and Slow – Daniel Kahneman


The unquestionable Bible of Behavioural Economics, Thinking
Fast and Slow chronicles the most famous findings of Danny Kahneman’s extraordinary research career. Along with Amos Tversky and Richard Thaler, Kahneman helped transform Behavioural Economics from an eccentric, nagging, but ultimately ignorable thorn in the side of mainstream Economic theory to one of the fast-growing and important sub-fields of Economics today.

For a large part, Behaviour Economics is concerned with how humans make decisions. Before the subject came to be, orthodox economic theory was built on the assumption of Rational Economic Man: the cool, calculating machine that abided by several axioms of rational choice. Sure – Rational Economics Man occasionally made mistakes, but these mistakes were so random and infrequent that they could be safely ignored for the purpose theory and public policy.

Behavioural Economics essentially took a sledgehammer to this assumption by showing that, in many contexts, humans make predictable and systematic mistakes. Rather than axioms of rational choice, humans are instead governed by two ‘Systems’ of thought: a fast, emotional and instinctive System 1 and a more cool-headed, deliberative and logical System 2. While traditional Economics would have you believe that System 2 was always in control, Kahneman and his collaborators showed that System 1 often intervened automatically. This gave rise to many ‘biases’ – behaviour at odds with textbook rationality. Examples of these include the halo effect, loss aversion, confirmation bias and the narrative fallacy.

What I like so much about this book is that it’s peppered with actual thought experiments to test your own susceptibility to System 1 thinking (I fell for almost all of them). My favourite of all these experiments involves estimating the odds that a car from either a Blue Cab Company or a Green Cab Company was involved in an accident. I won’t spoil the punchline here – but needless to say this example bamboozled me then and still bamboozles me today. I know what the right answer is; I know the mathematics to get to the right answer; but for some reason my brain still screams at me the wrong answer. The power of System 1 thinking!


The Black Swan – Nassim Nicholas Taleb



Next on this list is a book by loud-mouthed, Ferragamo-tie-hating,
former derivatives trader & Econ Twitter über-villain Nassim Nicholas Taleb – or NNT, as he often abbreviates. The book focuses on Black Swans – unpredictable, high impact events that we tend to ignore and downplay by applying simplistic explanations retrospectively. Examples of Black Swans include 9/11, the collapse of the Soviet Union and the rise of the internet. These events dominate our world – yet by their very nature, were impossible to predict in advance.

To cope with a world defined by these events, Taleb advocates a number of strategies. First, forecasters are to be ignored – especially those without Skin in The Game. Second, the Bell Curve – with its focus on averages rather than extremes – should be abandoned, except in a few very specific contexts (e.g. human height). Third, our entire epistemology must be overhauled: focus should move from knowledge (the stuff that we know) towards anti-knowledge (the stuff that we don’t). Fourth, we should become Anti-Fragile: distributing our eggs across many baskets to ensure that no single event can ruin us completely. Finally, we must make peace with the unavoidable chaotic of the world and adopt a stoic, detached stance – like Buddhism, says Taleb, but with a “F*** you!” to fate!

In making his case, Taleb takes us on a breath-taking voyage through the history of scepticism and the philosophy of science and social science. He devotes numerous pages eulogising the early purveyors of anti-knowledge and unpredictability: David Hume, Karl Popper and Benoit Mandelbrot. He also makes sure to slag off those who systematically over-simplify and under-estimate the world – ‘fragilistas’, as he calls them. Into this category falls Karl Marx, Carl Gauss and even Plato – no one, it seems, is safe from his arrogance or ire.

For all his bluster, NNT can certainly walk the walk. Each time I read one of his books I come away viewing the world through different lenses. Some of his advice is undoubtedly wise (become Anti-Fragile; ignore those without Skin in The Game); some of his advice is, well, eccentric (never run for a train, never read the newspaper and never drink anything from the past 4,000 years). He’s like a controversial uncle around the family Christmas table: rude, obnoxious, but undoubtedly entertaining and at times mesmerising. His online presence is also legendary: from posting videos of his deadlift sets on Facebook to calling Steven Pinker a fraud on Twitter, NNT is truly the gift that keeps on giving.

The Undoing Project – Michael Lewis



In 1969 two men met on a University campus.
They were different in every way. But both were obsessed with the human mind – and both happened to be geniuses.

So reads the blurb of The Undoing Project – the remarkable biography of Daniel Kahneman and Amos Tversky. In chronicling their lives and careers, the book is ostensibly about the founding of Behavioural Economics – but it’s also about so much more. It’s a tale of two types of genius: one arrogant, unfettered and seemingly unlimited; the other measured, perfectionist and wracked with self-doubt. It’s a tale of a young Jewish boy in Nazi-occupied Paris, and his death-defying escape to Israel. It’s a tale of human fallibility; and our inability or unwillingness to accept this fact. It’s a tale of unorthodoxy triumphing over conventional wisdom. It’s a tale of friendship and the love, laughter and envy that this encompasses. It’s a book about human happiness, and how it can be misunderstood. Finally, it’s a tale of the finitude of human life, and a reminder that even the brightest stars eventually fade away.

Of all the books on this list, this one is undoubtedly the most moving. Read it!

Political Science

Seeing Like a State – James Scott


If you want to understand Politics, you need to understand the State. And if you want to understand the State, you need to understand its modus operandi.

For Scott, the central project of the State boils down
to one thing: legibility. Any substantial state intervention in society – vaccinating a population, mobilising labour, enforcing sanitation standards, taxing people and their property, conducting literacy campaigns, catching criminals or conscripting soldiers – requires the invention of units that are visible.

To do this, primitive states had to demarcate and standardise a rich and varied tapestry of localised norms and social structures. Surnames were invented to make individuals identifiable and their lineage traceable through time. Standardised weights and measurements were introduced to make regional goods commensurate, and to ensure that their production could be monitored and taxed. Universal languages were imposed to ensure disparate groups could be understood, educated and instructed. Entire cities were built around perpendicular axis and geometric squares to ensure their populations could be counted, regulated, spied upon and organised.

This ‘high modernist’ planning gives States’ the unparalleled ability to move and manipulate groups of people en masse, but it is not without its costs. It becomes risky when it supersedes local practices that have adapted to local contexts over several centuries. And it becomes downright dangerous when it is combined with an authoritarian state and a weak civil society.

Take, for example, the ‘ujamaa’ project of Julius Nyerere’s Tanzania. Taking the reins of power in 1964, Nyerere was one of the foremost champions of the high modernist belief that it was the state – and the state alone – that knew how to organise a more satisfactory, rational and productive life for its citizens. He embarked on a compulsory villagization programme: smallholder farmers were forcibly relocated to massive collective farms that could be readily monitored and dictated from above. The results were disastrous: agricultural output collapsed and Tanzania became heavily reliant on food imports between 1973-1975. The blunt, unvaried and clinical logic of the State was no match for the nuanced local practices that had centuries of adaptation to fall back on.

This book has notable overlaps with two others on this list: The Anti-Politics Machine and The Black Swan. Like Scott, both Ferguson and Taleb caution against blanketing a highly complex social world with crude, abstracted and over-simplified explanations. Of course, as practitioners and students of social science, it’s our job to understand this world – something which can’t be done without simplification and abstraction. So how should we proceed? Scott gives four pieces of advice: first, take small steps: intervene moderately, stand back, observe, and then plan the next small move. Second, favour reversibility: prefer interventions that can easily be undone if they turn out to be mistakes. Third, plan on surprises: choose plans that allow the largest accommodation of the unseen. Finally, plan on human inventiveness: always assume that those involved in the project will have or develop insight to improve upon the design. I would add a final one: be humble – acknowledge the limits of social science and that our world is often too complex, chaotic and unpredictable for us to fully understand. But don’t let that deter you from trying!

States and Power in Africa – Jeffrey Herbst



Why has effective state building proved to be so difficult in Africa?

To answer this question, Herbst makes an exhilarating
journey through African pre-colonial, colonial, and post-colonial history. The answer begins, Herbst says, in the unique complexion of war in pre-colonial Africa. Unlike in Europe and Asia, pre-colonial wars of territorial conquest were extremely rare. The reason for this was two-fold: first, unlike in Europe, land was extremely abundant – by 1975, Africa had only reached the level of population density that occurred in Europe in 1500. Second, unlike in Asia, land was extremely cheap – the plough never reached Africa, meaning that even royal villages would move periodically as the soil became exhausted. This was in stark contrast to Asia – where the painstaking labour that went into constructing tiers of rice paddies meant that land was worth defending – and conquering.

This different dynamic of war had a profound impact on early state formation. In Europe and Asia, the allure of territorial conquest prompted elites to secure their borders and develop extensive tax frameworks – both of which required them to develop close connections with the hinterland. In Africa, by contrast, rural populations remained virtually uncaptured by the state apparatus.

The next part of Herbst’s explanation charters the transformation of the African state under colonial rule. Colonialism effectively grouped together previously disparate and transient populations into arbitrary geographic blocs. Despite this new cosmetic, European occupiers generally chose not to enhance the reach of the state much beyond the centre. Doing so was expensive – particularly due to Africans vast geography and lack of navigable rivers – and hence did not show up favourably in the cold cost-benefit calculus of colonial rule. The exception, of course, was if an area was abundant in natural resources – which could be extracted and then funnelled through the centre.

At independence, this urban bias was perpetuated by inaugural elites for largely pragmatic reasons. Urban workers could strike, riot or concoct a coup d’état; rural peasants, by contrast, were a lot less threatening to the regime. As a result, surveillance and spending were disproportionately biased towards the centre – again leaving the countryside relatively uncaptured.

However, while elites had little incentive to tighten their grip over rural populations, they had little incentive to let them go, either. Doing so would relinquish some of their newly found power: not only would they lose land, resources and (albeit limited) tax revenues, they would also lose credibility in the international arena.

By 1950, the principal of state sovereignty had become a virtually unassailable juridical norm. This can be traced to the horrors of WWII and the partition of the Indian sub-continent, both of which made vivid the human cost of boundary change and territorial ambition. The New World responded by a near-universal embrace of the nation state. Irredentist movements were almost always condemned and new organisation like the World Bank, the UN and the IMF identified nation states as the only legitimate actors they would do business with. To access this New World, African states had to remain unified: even if it meant holding on to bizarre and incoherent blocs handed to them by colonial rule.

What I like so much about Herbst’s argument is that it makes modern-day aberrations like state failure and civil war a lot less mysterious. When you consider that many African states are characterised by illogical borders, uncaptured rural populations and disparate ethnic and linguistic groups, the lack of effective policymaking hardly seems surprising. And when you consider the immense role that war played in state formation in Europe and Asia – and the bizarre way in which most African states came into being – then civil war doesn’t seem like such a stupid thing.

Climate Change

The Uninhabitable Earth: Life After Warming – David Wallace-Wells


It is worse, much worse, than you think. If you’re unconvinced of this, the first 12 chapters of The Uninhabitable Earth should set the record straight. Vast swathes of the equatorial belt are set to become literally uninhabitable: reaching temperatures that the human body is not adapted to survive in. Global food chains will be eviscerated by unprecedented weather patterns and drought. The megacities of Jakarta, Shanghai and Mumbai are set to become underwater within my lifetime, while the nations of The Maldives and The Marshall Islands may entirely cease to exist. Devastating wildfires, floods and hurricanes will become the new normal. Entire populations will be uprooted and dislocated, producing – according to median estimates – 200 million climate refugees by 2050. This, coupled with heightened competition over increasingly scarce resources (e.g. freshwater, cropland, forests), will likely exacerbate local, regional and international conflict. As Wallace-Wells says, it is probably worse – much worse – than you think.


While the first half of the book outlines our likely trajectory absent an urgent response, the second half explores how the human condition may change in the face of impending disaster. How will climate change mould our collective conscience and the stories we tell our children? How will our political, economic and moral systems adapt in the face of unprecedented upheaval? How will death, ecological disaster and economic collapse affect our perception of the journey we’ve taken on Planet Earth?

This is not a cheery read. No punches are pulled in outlining the enormity of the crisis we face. There are very few silver linings. The book does not end in a flurry of optimism about human ingenuity and technological progress – something I find morbidly refreshing.

Despite this, The Uninhabitable Earth really is essential reading. Millions of people have already been killed by climate change – whether via wildfires in California, flooding in Bangladesh or conflict in the Sahel. The impacts of climate change in the next 100 years are going to be bad – but there is a massive degree of scope in terms of how bad things will be. It matters whether we heat the Earth by 2°C or 3°C. It matters whether millions or billions of people die. Wilful ignorance of the risks is not a morally defensible strategy. If, like me, you’re not versed in the current scientific consensus, The Uninhabitable Earth is a great place to start.

Technology


Homo Deus – Yuval Harari


Homo Deus is the equally-hyped follow-up to Harari’s
ludicrously successful debut work, Sapiens. Sapiens ends with a wild prophesy: that Sapiens, via our scientific advancement, are cooking up the next epochal revolution – the technological revolution – that will overshadow the previous three (the cognitive, the agricultural and the scientific). Homo Deus is the full exploration of this prophesy.

Harari begins with a brief survey of human progress. Up until now, human meaning and progress has been largely defined relative to an ongoing struggle against the historic enemies of mankind: plague, famine and war. Now, for the first time, these scourges are falling under our control. Diseases such as smallpox, river blindness and polio have all but been eradicated. Famines have become sensational, not routine. Major war has become seemingly obsolete.

As we elevate ourselves above beastly survival struggles, Harari speculates that we will divert more time and energy towards upgrading the base human condition. Indeed, this trend is already visible: we outsource biological algorithms to our smartphones; we use aids to improve our hearing and sight; we take anti-depressants to alter our brain chemistry; and we have plastic surgery to makes us beautiful. Our ability to do this makes us already God-like, at least by the standards of most of history. However, if we continue, even more God-like abilities may be on the horizon: the ability to cheat death, customise embryos or even create life itself.

However, just like the three revolutions that preceded it, this revolution will come with a price. Sophisticated human upgrades will likely only be available to those who can afford it. The implications this has for global inequality may be unprecedented. Today, for the most part, the global superrich are physiologically indistinguishable from the masses – we have the same size brains, the same fleshy carcass and the same reliance on nutrients, water and oxygen. What if this were no longer the case? What if the super-rich, via bodily upgrades, were able to become faster, smarter and stronger than the rest? This is a scary prospect indeed – and one which our current systems of politics, laws and ethics are not equipped to deal with.


Things go from bad to worse. Even if we avoid this threat and create a super-human future for everyone, we risk propelling ourselves into a world devoid of any real meaning. How will we cope in a world where any economic, political or cultural endeavour can be done by algorithms better than by ourselves? We will need a new religion to make sense of all this: the powerful combination of science and humanism will not work in a world where the sanctity of being Human has lost all meaning.

Homo Deus is, of course, extremely broad-brushed and speculative in its treatment of the future, placing it in a broader category of books that are seemingly in vogue at the minute. In my eyes, Homo Deus is the best of the bunch – with an honourable mention to Future Politics by Jamie Susskind, which narrowly misses out on this list.

Unlike other books I’ve read, Harari treads a careful line between starry-eyed optimism and doom-mongering fatalism – probably the most sensible stance, given both the scale and unpredictability of our technological charge. He also sees the bigger picture better than most – teasing out the implications of technological advancement on philosophical themes as broad as human experience, individualism, human emotion and consciousness. If you’re interested in where we might be headed – or simply in what it means to be human – this book is for you.