The Pundit’s Fallacy Fallacy

In the wake of Donald Trump’s election, I and many others have drawn attention to what we believe is one of many factors the left has lost political ground over the past several years: far-left political ideology, particularly within the culture of higher education and the media (including radical factions of the social justice movement, Black and Brown identity politics, third-wave feminism, college campus censorship culture, political correctness, etc).

Our criticism has drawn the ire of many on the far left who are exceptionally quick to accuse us of committing the Pundit’s Fallacy. The Pundit’s Fallacy is a type of cognitive bias – the tendency towards believing that the best way for a politician to improve their political standing is to adopt a position more like ours. Of course, which policy positions are most likely to win an election are not necessarily the same as the policies we personally would like to see implemented, so it’s suspiciously convenient to believe that they happen to be identical.

So, isn’t it a bit suspicious and convenient that I not only oppose far-left political ideology but also think it helped Donald Trump win the election? Let me be the first to admit that the answer is yes. It is a bit suspicious.

But there’s an important mitigating factor that critics of the anti-far left are missing: Half the reason we opposed far left political ideology in the first place is that we predicted it would be a political disaster.

To illustrate how this works, imagine the following conversation between two store managers:

Manager 1: ‘Hey, I don’t think we should scale up our production. There isn’t enough demand and we’re going to end up losing money.’

Manager 2: ‘I disagree.’ *Scales up production, and the company loses money*

Manager 1: ‘See, that’s why we can’t scale up production. It’s going to lose us money.’

Manager 2: ‘Well, isn’t it a bit suspicious of you to say that? I mean, isn’t it rather convenient that you both believe we shouldn’t scale up production and that the company would be better off if we adopted a position more like yours?’

Actually, no, it’s not suspicious at all, because the reason Manager 1 didn’t want to scale up production to begin with was that he predicted it would be an economic disaster.

And yet, this is more or less how I feel talking about far-left politics.

Me: ‘Hey, I don’t think we should be pushing a far-left agenda. People don’t like it, and we’re going to end up losing the election.’

Far left: ‘I disagree.’ *Pushes a far left agenda, and we lose the election*

Me: ‘See, that’s why we can’t push a far left agenda. It’s going to lose us political capital.’

Far left: ‘Well, isn’t it a bit suspicious of you to say that? I mean, isn’t it rather convenient that you both believe we shouldn’t push a far left agenda and that our politicians would be more popular if we adopted a position more like yours?’

This a plainly absurd thing to say. I call it the Pundit’s Fallacy Fallacy: the tendency to view anyone who believes their stated political positions would improve their party’s political standing as inherently suspect. The main problem is that sometimes one’s stated political positions are what they are because they believe it would improve their party’s political standing.

In other words, there’s a specific causal link between what improves a party’s political standing and one’s stated political positions. So no, it’s not that suspicious, and it’s not a mere coincidence.

Now, there is one important caveat: I (and many others) oppose far-left politics for other reasons as well. I personally am unsure whether far-left politics was decisive in the most recent election, but even if it made no difference whatsoever, I would still consider it harmful (though to a much lesser extent). So yes, it is a bit suspicious that I also think far-left politics may have contributed to Donald Trump’s victory.

But it’s worth noting that most of the reason I and many others opposed far-left politics in the first place (or at least, opposed it so vocally) is precisely because we thought it might lead to this scenario. We predicted – months or years in advance – that openly and frequently mocking White people, calling people bigots for having reasonable concerns about Islam, creating institutional barriers to prevent conservatives from entering academia, etc. – would alienate millions of people and lead them to do something outrageous and reactionary. Something like, oh I don’t know, elect Donald Trump as the 45th president of the United States.

At that point, something will crack. The nonsuburban electorate will decide that the system has failed and start looking around for a strongman to vote for — someone willing to assure them that, once he is elected, the smug bureaucrats, tricky lawyers, overpaid bond salesmen, and postmodernist professors will no longer be calling the shots. …

One thing that is very likely to happen is that the gains made in the past 40 years by black and brown Americans, and by homosexuals, will be wiped out. Jocular contempt for women will come back into fashion. … All the resentment which badly educated Americans feel about having their manners dictated to them by college graduates will find an outlet.

-Richard Rorty, Achieving Our Country, 1998



Election Night Disaster

To say that I am ashamed, embarrassed, sick, and horrified at the man my country elected tonight would be a great understatement.

These emotions – and more – are no doubt shared by many of my fellow liberals. If you’re anything like me, you’ll be tempted to spend the next four years finding fault in our political opponents: their sexism, xenophobia, ignorance, and hypocrisy.

I want to suggest a more constructive approach. The Republican party acted shamefully in this election cycle, but there are things we as a Democratic party can improve as well. Here I’m going to suggest three things we should focus on over the next four years. This is by no means an exhaustive list, nor are these necessary the most important things we can do. They might increase our polling numbers by 1% cumulatively – but then again, 1% is often the difference between winning and losing an election.

First, stop treating ‘White Men’ like dirt. Even the phrase ‘White Men’ has become almost a curse word among certain factions of the left – especially on college campuses.

Now, some of my fellow liberals will be tempted to respond to my suggestion by accusing me of ‘crying White tears’. Some will even delight in that description. Others will make the very legitimate point that minority ethnic groups have been treated so horrendously that they can’t really be blamed for expressing frustration at the rampant racism in our society. After all, we can excuse someone for ‘punching up’, can’t we?

Perhaps you’re one of those who take pleasure in the misery of White men, in which case, I don’t blame you for your frustrations (in fact, I share many of them). But what you have to understand is that the mockery and derision of White men doesn’t make me or anyone else more likely to vote for your candidate.

We liberals need to start policing the anti-White anti-male sentiment that has arisen among certain factions of our party – not because minority ethnic groups don’t have a right to be outraged, not because we don’t recognize the legitimacy and moral necessity of fighting for a racially equal society, but because this sentiment has the potential to alienate 40% of registered US voters. White men vote, and whether we like it or not, we need them on our team if we want to win.

Second, take the problem of Islam seriously.

When I say ‘the problem of Islam’, I don’t mean, ‘kick all Muslims out of the country’ or any such nonsense. I mean we need to acknowledge that there are certain widespread, scripturally plausible interpretations of Islamic doctrine which adversely impact our world.

In this election, voters were faced with the choice between one candidate who proposed to ban Muslims from entering the country and another who until about 8 weeks ago couldn’t bring herself to utter the words ‘Islamic terrorism’ in the same sentence.

Both of these positions are transparently outrageous, and any sane human being should be able to see so. The common sense, middle ground position is to appreciate that there is a problem within Islam today – that ISIS really is Islamic, and that certain interpretations of Islam have inspired acts of terrorism and gross violations of human rights more generally. At the same time, we need to acknowledge there are also plausible and widespread interpretations of Islam which are either benign or benevolent, and that Muslims who embrace these versions of Islam should be welcomed into our country with open arms.

I take this position to be fairly self-evident, but unfortunately neither candidate was able to articulate it. Moreover, leftist political commentators and in many cases leftist politicians were on a hair-trigger to shout ‘racism’ at anyone with enough common sense to observe that there is a problem with Islam and enough courage to say so out loud.

We on the left need to start articulating this middle ground position – a position which implies that we should be concerned about the religious elements of terrorism and Islamic theocracy, but which at the same time acknowledges that most Muslims don’t adhere to these religious elements or operate according to them.

More importantly, we need to stop demonizing those who take this middle ground position – not only because it’s the correct position, but also because slandering voters for appreciating the reality of Islam isn’t going to win us the election.

Third (and my final suggestion for this evening), make the Democratic party the party of reason.

We on the left are quick to ridicule people like Newt Gingrich for saying that he values the voters’ feelings over facts, for claiming that whether or not national crime statistics show an upward or downward trend in violent crime is a ‘matter of opinion’.

But we on the left do the same thing. All the time. The only difference is that John Oliver doesn’t mock us when we do it.

We’ve apotheosized personal experience over statistics and data, glorified the preeminence of feelings over rationality in the spheres of altruism, politics, and morality, and, when our perception of reality clashed with the facts, retorted ‘well, that’s just your opinion’.

This is unacceptable. We need to start creating a societal norm which says that facts are facts, that personal experience does not equal expertise, and that our uncertainty about the facts is not a license for treating everyone’s ‘opinion’ as equally valid.

Bobby Jindal once said to the Republicans, ‘we must stop being the stupid party’. We on the left need someone to say to us, ‘we must stop being the irrational party’.

Now, we may not be any more irrational that our conservative counterparts. In fact, I would bet that liberal political views tend to correlate with rationality quotients and similar measures. And, you might object, people don’t seem to respond to reason and evidence. Look at Donald Trump.

But let me be something of a neoclassical for a moment when I claim that supply can drive demand in this instance. I would speculate that if we create a supply of political discourse fundamentally grounded in reason and evidence – an atmosphere in which emotion is our tailwind but rationality is our helmsman – the electorate will demand more of it.

People are tired of being lied to and manipulated. So let’s stop. If we treat people like rational agents rather than emotional children, they will respond.

Measuring Wellbeing

About a month ago, I set up a rule for myself: whenever the same controversial topic comes up three times independently in a single week, I write a blog post about it.

This week’s controversial topic: measuring wellbeing.

A common question (which often sounds more like an objection) to effective altruism, utilitarianism, and other consequentialist philosophies which rank outcomes in terms of the wellbeing of humans and nonhuman animals is, but how are you measuring wellbeing? Doesn’t everyone measure wellbeing differently? If so, how can we say that certain charities or public policies are better than others?

As with many questions, the question of measuring wellbeing can take many forms, running the gamut from very serious and important at one end of the spectrum to silly and potentially dangerous at the other.

Starting with the serious and important questions: There are many unresolved philosophical and empirical issues when it comes to measuring wellbeing.

Some of the philosophical issues include things like, Should we weight the wellbeing of all people equally, or should we assign people different weights based on things like deservingness? Perhaps we think some people are more or less deserving of happiness than others. For instance, we might intrinsically care more about the wellbeing of a vegan who donates 98% of their income to charity and works tirelessly for the betterment of others than we do about the wellbeing of a serial killer.

There are psychological issues when it comes to measuring wellbeing as well. Some happiness studies ask participants about overall life satisfaction, while others ask participants about moment-to-moment experience, often with very different results. Which of these two measures more accurately reflect wellbeing? I don’t think we yet have a clear answer.

Here’s another measurement issue from the field of environmental economics: Economists often use GDP per capita as an admittedly imperfect proxy for average wellbeing within a country. But this measure doesn’t account for environmental degradation and other externalities of resource extraction and industrial production. So perhaps maximizing wellbeing as measured by GDP per capita in the near future comes at the expense of maximizing wellbeing as measured by GDP per capita over a longer time horizon. If so, how do we project the future costs of contemporary environmental degradation?

These are just a few examples among many of the philosophical and empirical issues we run into when attempting to measure wellbeing.

Certainly the aforementioned measurement issues are as interesting as they are pressing. But let me now shift focus to the other end of the spectrum: the (non)issues in measuring wellbeing which I consider not only silly but also immensely dangerous.

To caricature these (non)issues only slightly, I often hear (mostly from liberal arts university students after a couple of glasses of wine) some version of the following: But isn’t wellbeing totally subjective? Other minds are so inscrutable that we can’t possibly know how others feel about anything. Wouldn’t it be awfully presumptuous of us to attempt to compare the wellbeing of two different people?

Let me try to shatter this line of thinking with a thought experiment.

Imagine a woman we’ll call Alice. Alice has a husband with whom she’s madly in love (or at least, so she says) and to whom she’s been happily married for 20 years. She has two children who are just beginning to embark on successful careers, who she sees several times a year and chats with weekly on the phone. Alice works a modest 35 hours a week at a job which pays $55,000 per year and which (she claims) she finds deeply fulfilling. She has several close friends and volunteers at her local homeless shelter every week. Moreover, Alice has no serious health issues – mental or otherwise. By all accounts, including her own subjective reports, Alice feels a deep sense of wellbeing in her life.

Now imagine a woman we’ll call Yana. Yana lives in war-torn Syria. Last week, Yana watched government forces execute her husband and younger brother for their participation in the rebellion. She was forced to flee with her daughter to a refugee camp where they sleep in a small crowded room. Her daughter is running a severe fever, but because of triage the doctors probably won’t be able to see her until next week. Yana is still uncertain about the fate of her son, who was taken prisoner by Assad’s forces last year. If Yana were to see a psychiatrist, she would be diagnosed with severe depression and suicidal tendencies. By all accounts, including her own subjective reports, Yana is in a chronic state of suffering and misery.

I think any reasonable human being would agree with me when I conclude that Alice is probably experiencing a greater state of wellbeing than Yana. When I put it like this, it seems flatly obvious that interpersonal comparisons of wellbeing are possible, at least in certain cases.

Why is it important to acknowledge the possibility of interpersonal comparisons? Because it allows us to get our priorities straight. It lets us realize that certain problems are causing more suffering than others, and leads us to conclude that we should be working first to solve those problems which cause the greatest amount of suffering.

Reality Check: Africa is a continent that needs our help

Take a minute and watch this video.

It’s the sort of thing that’s meant to change your perspective on Africa. Look at all those statistics and figures. One could almost come away with the impression that things really aren’t so bad in Africa; with their shiny trains, strong middle class, and rapidly growing economies. Equally amazing is how Mehdi Hasan managed to make this point without a hint of condescension.

That is, until you realize that almost everything said in the video is either false or technically true in a way that completely distorts reality. Here’s a point by point breakdown of what he got wrong:

The ‘Lazy Image’ of African War, Famine, Corruption, and Disease

Hasan begins by criticizing what he calls the ‘lazy’ image of war, famine, corruption, and disease in Africa, and this seems like a good place to start. Here’s a non-lazy, accurate image of disease in Africa:

The child mortality rate in Sub Saharan Africa is more than twice the global average and 16 times that of the US and Europe. And in case this seems like a problem for the developing world in general, consider that the African child mortality rate is twice that of other developing nations including India and several countries in Southern Asia.

Life expectancy in Africa is just ¾ of life expectancy in developed nations, and 10 years lower than in developing nations outside of Africa. The majority of African deaths are due to disease (62% of total deaths), compared with just 5% in developed countries and 28% in India and Southern Asia.

Obviously not all of Africa is afflicted by disease, nor is disease solely an African problem. Still, it would be dangerous for us not to acknowledge that African countries in general are unique in their level of disease burden, and that this is a critical obstacle to overcome on the path towards prosperity.

22 Nobel Laureates

Quite frankly, number of Nobel Laureates is a terrible indicator of a region’s welfare or accomplishments. But since Hasan brought it up…

Let’s do a rough calculation to see how 22 Nobel Laureates compares to the rest of the world. If we just look at current population size as a proxy for population size since the Nobel Prizes came into existence, we can see how many prizes per capita were awarded to Africa relative to the rest of the world.

This gives us a figure of 0.02 Nobel Prizes per million people in Africa. Compare that to the rest of the world at 0.14 per million. For some perspective, this means that the average African is about seven times less likely to win a Nobel Prize than everyone else.

4 of the 10 Fastest Growing Economies

This is a prime example of a statistic which is technically true, but which completely distorts reality.

The four countries Hasan refers to are Côte d’Ivoire, Mozambique, Ethiopia and the Democratic Republic of Congo, with growth rates ranging from 7-10% per year in recent history (compare this to the typical US annual growth rate of about 4%). But now let’s take a closer look at the economic reality in these countries.

After adjusting for purchasing power, we see that the per capita GDP of these economies ranges from $3,500 in Côte d’Ivoire down to $800 in Congo. For comparison, the US poverty line is $11,000 per year, with an average GDP of $56,000 per year. In other words, the average person in the wealthiest of these ‘four of ten fastest growing economies’ is three times as destitute as impoverished Americans.

If anything, rapid growth is a hallmark of an unhealthy economy. Rapid growth – such as we see in these four countries, as well as in places like India and China – is an indication that a nation is finally starting to catch up with the rest of the world. These figures should inspire optimism about the future of African economies. But simultaneously, we should be concerned that at present, these nations are so far behind developed countries that they have to scramble to catch up.

It’s also worth noting that many African countries have staggeringly low growth rates (2-3% annually), in no small part due to disease burden.

One in Three Africans are Middle Class

This is undoubtedly the most egregious mistake of Hasan’s piece. Unfortunately, I couldn’t find the source of his figures, but a few minutes combing over World Bank data shows that this claim can’t possibly be true in any meaningful sense.

For most African countries, there is no data on GDP share by income bracket. That is, we don’t know how much of the country’s generated wealth goes to the top, middle, and lower third of its citizens. A few countries, however, did report GDP share by quintile – that is, top, upper middle, middle, bottom middle, and bottom 20%.

Of these, I decided to assume (in Hasan’s favor) that as large a share of Africa’s GDP went to the top 20% as I could by extrapolating from Cameroon (with the highest rate of inequality). That is, I generously assumed that 51.6% of Africa’s GDP went to the top fifth of its population.

Combine this figure with the fact that African GDP PPP per capita is $3700, and we can calculate the average income of the top quintile of Africa’s population: $9600. So by American standards, not even the top 20% of Africa’s population would be ‘middle class’, much less the top 33%. In fact, the average person in the top fifth of Africa’s population would be classified as impoverished.

So where in the world is Hasan getting this ‘one in three’ figure? My guess is that he’s disingenuously playing with the term ‘middle class’.

To illustrate how this trick works, if you’re ever planning on running for office against an incumbent, this is a great line to use: The poverty rate in this country hasn’t improved one bit in the last 8 years! And it will always be true.

How can it always be true, you ask? Because most countries define their own poverty line as whatever the watermark is for the bottom 20% of our population. Even if everyone in your country became 50 times as wealthy in real terms as they were 8 years ago, it would still be technically true that the poverty rate – the percent of the population living below your country’s poverty line – is the same, because by definition it’s fixed at 20%.

My best guess is that Hasan is doing something similar. He’s playing a trick whereby what it sounds like he’s saying is ‘one third of Africans are middle class by anyone’s standards’, when what the facts actually indicate is that ‘one third of Africans are middle class by African standards’, which are abysmally low compared to the standards of developed countries.

By the way, take a look at the suits the African people are wearing when Hasan talks about the ‘middle class’. Each one of those suits cost more than the average middle class African makes in four months.

One of the Fastest-growing Markets for Mobile Phones

Like the ‘four of ten fastest-growing economies’ point, Hasan’s argument about mobile phones has it exactly backward. A rapidly growing market for mobile phones isn’t indicative of a healthy economy. Just the opposite. 98.6% of all human beings on earth have a mobile phone already. In Africa, one in four do not.

Why is there such a burgeoning market for mobile phones? Because so many Africans don’t already have mobile phones.

Again, this is a good sign of things to come. Either it means that Africans are finally becoming wealthy enough to afford basic communication devices or that mobile phones are becoming cheaper to produce. Still, at the present moment, the emerging market for mobile phones is an indication that many Africans are just now able to afford technologies that the rest of the world has been using for decades.

Look at that Shiny Railway

Great, now we’ve seen definitive proof that at least one African railway car looks better than at least one American railway car.

Now let’s take a look at Africa’s railway system as a whole. On the continent, there are about 59,000 km of railway for 30.3 million square kilometers of land – or about 0.2 km of railway per 100 square kilometers of land. In the US, there are 229,000 km of railway for 10 million square kilometers of land – about 2.2 km of railway per 100 square kilometers of land.

Adjusted for land area, Africa has 11 times less railway than the US. (Adjustments for population yield similar figures).

That Hasan can show us a picture of one nice railway car in Africa has absolutely no bearing on the fact that the continent as a whole desperately needs improved infrastructure, especially to transport goods from major ports to its large landlocked regions.

The Two Things He Got Right

So, I’ve spent most of this post ripping apart everything Hasan got wrong. Maybe I can spend a bit of time on the two things he got right at the end.

First, he’s correct about the fact that the political participation of women in African governments often rivals that of more developed nations, or in some cases surpasses them.

Second, he’s very correct that foreign aid plays a small role in Africa’s economies, which brings me to the point of writing this post.

We cannot – absolutely cannot – understate the challenges faced by African countries. It is entirely unacceptable to make a video giving the impression that things are more or less fine in Africa, barely acknowledging the problems of poverty and disease and even then in only a hand-wavy sort of way.

Distorting facts to make it seem as though African countries are doing better than they actually are – presumably to push some political agenda – is disgraceful. It prevents us from appreciating the nightmare of poverty and disease in which so many Africans live, and absolves us of our responsibility to do something about it.

There is more human suffering and misery in Africa than anywhere else in the world. The fact that we’re doing so little to help the African people – in terms of foreign aid, making equitable trade deals, debt forgiveness, etc – is shameful. We need to be raising awareness about these problems, not obscuring them; because only by acknowledging reality will we be motivated to change it.

Academic Luck

Two days ago, I was inducted into the economics graduate program at Cambridge University. This is a dream come true for me. A year ago today, I would never have thought that I would be studying the subject of my choosing as a graduate student at one of Britain’s top universities.

On the first day, the students in my program – the Graduate Diploma – had a lunch organized by the department with a few of our professors. Out of curiousity, I asked one of them how many students had applied for the Diploma. He said about 150-200. Out of those, 25 of us were selected to study here – somewhere between a 12-17% acceptance rate. Cambridge selected only the best, he told me.

This evening, I met another graduate student at my college who’s starting his PhD in history. It wasn’t his first choice, he told me. He’d studied history of philosophy and economics throughout his undergraduate and masters, but what he really wanted to study was economics. Of course, he said, he didn’t have a strong enough economics background to go straight into an MPhil or DPhil, so he had applied to the Graduate Diploma program – my program – here at Cambridge. And been rejected.

We had a nice chat over dinner, and I couldn’t help but think that this young man was at least twice as qualified for my program as I was. He was clearly very knowledgeable on economic history, on the strengths and weaknesses of econometric modelling, and on the role of macroeconomic policy-makers in Europe (and probably many more topics that didn’t happen to come up over dinner). I couldn’t help but think that if Cambridge had spent even 5 minutes interviewing us, he would easily have taken my spot.

It wasn’t that Aaron (let’s just make up a name for him) was ungrateful or unhappy with the program he got into. Just the opposite – he seemed thrilled to be here. He was lucky to have had a great backup option. But it does make me wonder how many other Aarons there were who applied but didn’t get into my program, and who had to settle for, well, something else.

Last year, I applied to over a dozen graduate programs and over 30 economics research assistantships. Of the research assistantships, I made it to the first round of 2 interviews, neither of which called me back afterward.

During the months following my applications, I got a series of rejection letters, one after another, to the point where I was terrified of getting rejected from literally every program I had applied to. Finally, I received an acceptance letter from a wonderful school. It was the only one I would get.

Cambridge may select only the best, but it does not select all of the best. The same holds for every top university, competitive job position, scholarship, etc. There are simply too few spots to fill, and too many applicants with near-perfect credentials who want them. I didn’t get into Cambridge because I was better than everyone else – I got into Cambridge because I was luckier than everyone else.

I’m not sure what message I wanted to convey here, other than this: Those of us who have studied or are studying at top universities should recognize how fortunate we are – that we are not necessarily smarter or harder working or better qualified than the majority of candidates who were rejected, but only luckier. Those of us who didn’t manage to get into these programs should remember that rejection isn’t a judgment about our competence or potential. Keep working, and keep applying – perhaps you’ll be luckier this time.

Putting things in Perspective

At the moment, there are three serious contenders for what should be the world’s highest priority: extreme poverty and preventable disease, animal welfare, and global catastrophic risk. I believe the jury is still out as to which of these really is the number one priority, and perhaps other serious contenders will emerge over time.

I, for the most part, tend to believe that extreme poverty is the greatest moral crisis of our time. Often when I make this claim, I can expect a number of complaints. Isn’t it just my opinion that extreme poverty should be one of the world’s top priorities? Who am I to say that global poverty is worse or easier to mitigate than poverty in the developed world?

But the direction of causality assumed in these questions is exactly backwards. The reason I’m so passionate about extreme poverty in the first place is because of the overwhelming evidence showing the depth and breadth of human misery for which it is responsible.

Here I’m going to share just a few key statistics to put the problem of global poverty in perspective; the statistics that inspired me and continue to inspire me to study development economics. Before I begin, let me just clarify that all of the figures you see are adjusted for purchasing power.

For example, when I note that 900 million people worldwide are living on $1.90 per day, a common response is, sure, but $1.90 goes pretty far in developing countries, so that’s not a fair comparison. But no, the $1.90 per day figure already accounts for price differences between the US and whichever region of the world we’re talking out. That is, 900 million people are living on the equivalent of $1.90 per day in Miami, Florida.

With that being said, here we go:

The poorest quintile of the world’s population is 10 times as destitute as the poorest quintile of the US population.

The child mortality rate of Sub-Saharan Africa is 16 times that of the developed world.

We see someone in the US living on $12,000 per year and consider them below the poverty line. That person is wealthier than 85% of the world’s population and 8 times as wealthy as the global average.

We see someone in the US living on $55,000 per year and consider them middle class. That person is wealthier than 99% of the world’s population and 40 times as wealthy as the global average.

42% of the world’s income in concentrated in the hands of 10% of the world’s population (i.e. everyone earning above $17,000 per year). The poorest 10% of the world’s population (i.e. everyone earning below $700 per year) have just 1% of the world’s wealth. Source

The leading cause of child (under 5) mortality in the US is unintentional injury, which kills 5 in 100,000 US children annually. The leading cause of child (under 5) mortality in Sub Saharan Africa is malaria, which kills 550 in 100,000 African children annually.

This is in spite of the fact that we can save a child from dying of malaria for $3,500.

The number of deaths from suicide and violent crime, war, and terrorism combined is just 60% of the number of deaths from preventable disease each year. One important difference is that half of those who die from preventable disease are children under the age of 5.

Americans donate a mere 0.09% of their income to developing countries.

Less than 1% of our federal budget is spent on foreign aid.

Still, there is cause for hope. The number of people living in extreme poverty today is just half of what it was two decades agoMoreover, global inequality has been steadily falling since the 1960s.

I believe we can see the end of extreme poverty within our lifetime. It’s time to make this the world’s top priority.

Defending the DALY

New rule: Whenever the same controversial topic comes up three or more times independently in casual conversation within a week, I have to write a blog post about it.

This week the topic was the disability-adjusted life year, or the DALY. The DALY is a standard metric in healthcare policy employed by the World Health Organization, MIT’s Poverty Action Lab, the Center for Disease Control, the National Healthcare Service, and many other top-ranked institutions with fancy acronyms.

Philosophically, the DALY is a flow measure of quality of life over time. To start with, take a person (let’s call him Fred) living a full quality life – not a perfectly happy life, but a ‘full quality life’, meaning his basic healthcare needs are met and he doesn’t have any debilitating diseases. Each year of Fred’s life is worth 1 DALY.

Now take a person (let’s call him Harry) who’s suffering from a horrible, debilitating disease. For the moment, let’s forget about what exactly this disease is – call it disease D if you like. The point is that everyone, including Harry, would strongly prefer to be free of D if he could be. Harry feels like he lives a worthwhile life, but he estimates that the quality of life has been diminished by a factor of ½. This is the “disease burden” of disease D. Each year of Harry’s life is therefore worth ½ DALYs.

Organizations like the WHO, the UN Development Programme, etc, standardize this metric across diseases; estimating the disease burden for every ailment under the sun, from blindness to malaria to cancer. They then employ this metric in estimating the cost-effectiveness of healthcare interventions. Suppose it costs $6000 to cure Harry of D. After being cured, Harry is going to go on to live an expected 30 years of full quality life. This intervention would be worth 15 DALYs (30 DALYs at full quality of life – 15 DALYs with disease D). Its cost-effectiveness is $6000/15 DALYs = $400 per DALY.

Ultimately, what the DALY allows us to do is compare the cost-effectiveness of seemingly incommensurable healthcare interventions. How can we compare medical treatments for cancer to medical treatments for malaria? Use the DALY. So far, so good. Where’s the controversy?

Well, according to some of the people I’ve spoken with (not necessarily in the past week), the DALY is ableist, because it assumes that people who live disease- and disability-free lives are generally better off than people who live with disease or disability.

Plenty of people with disabilities, particularly congenital ones like blindness, are happy to be blind. They wouldn’t have it any other way. Even if someone were to come up to them and offer them a 100% guaranteed ‘cure’, free of charge, they would turn it down. In fact, there are plenty of people who weren’t born blind but acquired the condition later in life who say if they could go back in time they would still choose to become blind rather than retain their sight. So how can you say that people with sight are generally living a higher quality life?

In responding to this line of objection, I want to begin by noting an odd asymmetry. Imagine we develop two medical magic potions. The red potion permanently cures blindness – no side effects, guaranteed to work. The blue potion permanently induces blindness – no side effects, guaranteed to work. Further imagine that Big Pharmaceutical mass-produces both these magic potions and in an uncharacteristically altruistic move, offers them for free to everyone on the world, shipping the potions right to their front door.

How many blind people would drink the red potion, supposing they believed it would really work? My guess is nearly all of them. Greater than 99%. How many sighted people would drink the blue potion, supposing they believed it would really work? My guess is almost none of them, except perhaps for a few with severe psychological disorders or who were experiencing a severe momentary lapse in judgment (exhibit A: Oedipus).

There are a couple of things we can get out of this thought experiment. First, given the choice, almost everyone would prefer to be disease-free. This is already strong evidence that disease is overall bad.

Second, most people who say that a life with disease and disability is as good as or better than a disability- and disease-free life already have diseases and disabilities which are difficult if not impossible to cure. I’d be willing to bet large sums of money that a significant portion of those who make this claim would change their tune if we were to invent a painless and freely available cure that everyone knew was guaranteed to work. The primary motivation for saying otherwise, I suspect, is a sort of psychological defense mechanism. Well, I’m stuck with disease D indefinitely, and I don’t want to be miserable about it forever, so maybe I can try to trick myself into believing that my life wouldn’t be any better without it.

Still, we have to allow for the possibility that some people really do prefer to have a disease or disability, or at the very least are indifferent to them. Even if we were to invent a side-effect free “cure”, teach them all of the relevant medical science behind it and show them the results of a double-blind randomized control trial so they knew beyond the shadow of a doubt that it worked, and offer it to them free of charge, they still wouldn’t take it. Fine.

What proportion of the population thinks like this? Obviously it depends on the disease or disability. Let’s consider blindness, for example. What portion of the population would be genuinely indifferent or even happier with a life in which they will never see a Monet or Renoir? With a life in which they are at increased risk from an accident-related fatality? With a life in which they are ineligible for a vast array of employment opportunities? With a life in which they are largely dependent on the support of others, even a financial burden to their loved ones (as is particularly true in the developing world where the blind are led around not by guide dogs but by their family members pulling them along using a stick or string)? With a life in which they are unable to avail themselves of so many modern technologies which rely on the visual modality? With a life in which they will never see their spouse smile? With a life in which they will never see the face of their newborn child?

My upper bound estimate is somewhere between 1 in 10,000 and 1 in 100,000 – about 0.01% of the population. Now for my final question. Would it be ethical for us to ignore the DALY – a metric which has been tremendously useful in guiding efforts by healthcare organizations in reducing disease on a global scale for decades – because it fails to reflect the preferences of 0.01% of the population? Or can we recognize that this metric, with all its imperfections, is an effective means of measuring the suffering caused by disease?

We are in triage every second of every day

Wonderful post by my friend Holly

Holly Elmore
Spoilers ahead— listen to the episode first if you don’t want to hear rough summaries first.

I quite liked the above episode of RadioLab. The topic is triage, the practice of assigning priority to different patients in emergency medicine. By extension, triage means to ration scarce resources. The episode treats triage as a rare phenomenon– in fact, it suggests that medical triage protocols were not taken very seriously in the US until after Hurricane Katrina– but triage is not a rare phenomenon at all. We are engaging in triage with every decision we make.

The stories in “Playing God” are gripping, particularly the story of a New Orleans hospital thrown into hell in a matter of days after losing power during Hurricane Katrina. Sheri Fink discusses the events she reported in her book, Five Days at Memorial. The close-up details are difficult to stomach. After evacuating the…

View original post 983 more words

The Root Cause of Poverty

“How should we address the root cause of extreme poverty?” is a question I’m usually asked whenever I talk about global poverty, healthcare in the developing world, or effective altruism (which, as you can imagine, is pretty often).

Usually this is asked as if the querent and I have some mutual understanding about what the root cause of extreme poverty is, which is perplexing to me, although whenever this question is put to me by a liberal arts student I can pretty much guess what they have in mind: capitalism. Yes, the root cause of extreme poverty is capitalism, and the way to fix extreme poverty is to, well, get rid of capitalism (somehow) and replace it with, well, something other than capitalism.

Sometimes the “root cause” isn’t capitalism – it’s corruption, or Western imperialism, or lack of education. All of these answers are, in my view, wrong. They’re not wrong because they’ve misidentified the root cause of extreme poverty: They’re wrong because there is no such thing as a single root cause of extreme poverty. There isn’t even such a thing as a small set of root causes (plural) which are responsible for extreme poverty.

Extreme poverty is caused by a variety of complex and interacting factors. Furthermore, which factors are operant in causing extreme poverty vary between different regions of the world. Each country and even each city faces unique cultural, environmental, and economic challenges when it comes to raising living standards for the poor.

Consider treatments for neglected tropical diseases (NTDs), for instance. NTDs are a class of 17 common tropical parasites such as schistosomiasis, lymphatic filariasis, and hookworm which infect over 1 billion people and are responsible for some 500,000 deaths annually. Despite the devastation NTDs cause, they are relatively easy to prevent and treat. A couple of pills which cost $1.25 to produce and administer can immunize against a broad spectrum of NTDs for an entire year. So why haven’t we fixed this problem?

One village in South Sudan might welcome pharmacological treatments for neglected tropical diseases, but suffer from lack of availability (a supply-side problem). Perhaps this supply-side problem is caused by lack of funding for NGOs and governments which distribute the drug treatments (avarice or apathy on the part of the developed world). Or perhaps private donors and governments in the developed world have given sufficient funds to cover this village, but somewhere along the line the money was stolen by a bureaucrat in the South Sudanese government (institutional corruption).

Another village in India might reject these pharmacological treatments even if they were free (a demand-side problem). Perhaps this demand-side problem is caused by aversion to the risk of trying out new medicines (the limits of bounded rationality operating under incomplete information) compounded by the delayed nature of the benefits (a cognitive bias called hyperbolic temporal discounting). Or it could be a perfectly rational fear brought on by a history of forced sterilizations, such as those in India from the 1970s onward (an ongoing cultural problem caused by a historical institutional problem).

The point I’m trying to illustrate is that opining about “the root cause” of extreme poverty isn’t just a waste of time, but a distraction from a more granular, nuanced, and accurate picture of global poverty – a picture that actually allows us to gain traction in the fight to raise living standards among the world’s poorest populations. So from now on, whenever someone asks me how we should address the root cause of extreme poverty, I plan to respond by saying, We shouldn’t – and that’s what’s going to allow us to fix the problem.

Opportunity Cost Amnesia

Here’s a conversation that would never happen:

Stock broker: “Alright, you’ve said you’re ready to invest $10,000. I can recommend a portfolio for you that’s going to maximize yield and make you lots of money. A large part of this portfolio is going to be made up of stock A with an APY of 10%.”

Investor: “Cool, but actually I was hoping to put most of my money in stock B.”

Stock broker: “But stock B has an APY of only 2%.”

Investor: “Why can’t I invest in both?”

Stock broker: “Well, you can I guess, but you’ve only got $10,000, so if you put most of your money in stock B that means you’re going to be putting that much less in stock A.”

Investor: “Look, why are we comparing stocks A and B? You can’t compare different stocks. If I want to invest in stock B then the only thing we should be discussing is whether or not stock B is going to make me money.”

Stock broker: “Sure, obviously stock B is going to make you money, but you could make even more money by investing that money in stock A instead of stock B!”

Investor: “I don’t think you can compare stocks like that. That’s just not how it works.”

Here’s a conversation that happens all the time, at least to effective altruists:

Effective altruist: “Alright, you’ve said you’re ready to donate $10,000. I can recommend some charities for you to donate to that will save lots of lives. A large part of this donation is going to go to charity A which will use your donation to save 10 lives.”

Interlocutor: “Cool, but actually I was hoping to donate most of my money to charity B.”

Effective altruist: “But charity B could only save 2 lives with the same amount of money.”

Interlocutor: “Why can’t I donate to both?”

Effective altruist: “Well, you can I guess, but you’ve only got $10,000, so if you donate most of that to charity B that means you’re going to be donating that much less to charity A.”

Interlocutor: “Look, why are we comparing charities A and B? You can’t compare different charities. If I want to donate to charity B then the only thing we should be discussing is whether or not charity B is going to save lives.”

Effective altruist: “Sure, obviously charity B is going to save lives, but you could save even more lives by donating that money to charity A instead of charity B!”

Interlocutor: “I don’t think you can compare charities like that. That’s just not how it works.”

There’s a basic concept in economics called an opportunity cost. The opportunity cost is the cost of the ‘lost opportunity’ you would have had by using your time and money differently.

If you choose to invest in stock B instead of stock A, then the direct benefit you gain is 2% APY. But at the same time, you would have gained 10% APY by investing in stock A, which means you’ve lost 10% in opportunity cost. So instead of thinking, Cool, I’m making 2% APY, you should be thinking Damn, I’ve lost a net of 8% APY (10%-2%=8%) that I would’ve gotten by investing in stock A!

This is screamingly obvious when we’re thinking about decisions that affect us. Every meal we eat at an okay restaurant is a meal we’re not eating at a good restaurant. Every hour we spend watching TV is an hour we’re not exercising or sleeping or working. Every dollar we invest in stock B is a dollar we’re not investing in stock A.

Because we think about opportunity costs when it comes to decisions that affect us, we try to make the best decisions given our limited resources. We eat at restaurants that get us the best value for our money, we try to make time for important things like family and self improvement by cutting back on less important activities like TV, we make sure we’re investing in stocks that make us as much money as possible.

But when it comes to decisions that affect others, opportunity costs aren’t a factor. We have a sort of opportunity cost amnesia, where the concept of opportunity costs goes straight out the window.

When we donate to ineffective charity B instead of effective charity A, our reaction shouldn’t be, Cool, I saved 2 lives, it should be, Damn, I’ve let 8 people (10-2=8) die unnecessarily!

Every dollar we donate to charity B is a dollar that’s not going to charity A, and every hour we spend volunteering for organization X is an hour we’re not volunteering for organization Y. Sure, we can donate to a mix of ineffective and effective charities. But we need to recognize that when we explicitly choose to donate to ineffective charities or volunteer for ineffective organizations, we are letting people and non-human animals suffer and die unnecessarily via opportunity cost.

Opportunity cost may not be the only factor in our altruistic decision-making process, but it can’t be ignored either. We need to ask ourselves first, How many lives could I have saved by spending my time and money differently? And second Is whatever reason I have for donating to an ineffective charity worth that many lives? Very often, I believe than answer to the latter question will be an unequivocal and resounding No.