Photo by Mitul Grover.

Extreme Poverty as a Choice

The world is wonderful. The world is terrible. Let’s choose again.

Ernesta Orlovaitė


The world is richer, healthier, and more educated than it has ever been. In the last 150 years, we broke out of the Malthusian trap; eradicated smallpox; invented sliced bread, the automobile, the Internet, and the smartphone; and put a man on the Moon. And yet the awe-inspiring scientific, technological, and social progress has not benefited everyone equally.

Today, 737 million people are extremely poor.

821 million are undernourished.

2.2 billion do not have access to safe drinking water.

1 billion cannot read or write.

The unprecedented inequality is a choice. Extreme poverty is a choice. We, the rich, healthy, educated citizens of high-income economies, have decided that the rest of the world should stay poor. The less they live on, the more we can consume.

Don’t get me wrong — there are wonderful people and organisations out there who are working hard to address these challenges. But instead of leaving extreme poverty to the passionate few, we should all care.

This is a story about global poverty: the breathtaking progress, the heart-breaking inequality, the deliberate decision not to address it. It’s also a story about kindness: the things we say, the things we do, and the things we could do. But most importantly, it’s a story about cash transfers: the inner workings, the impact, the objections, and the promise they hold for the millions who are struggling.

If you only remember one thing after reading the article, let it be this: extreme poverty can be eradicated today. Here’s one way to do that.

The State of the World

The world is wonderful

The world is now richer than it has ever been. For millions of years, humanity was stuck in the Malthusian trap:

  • Technological advances would lead to improved standards of living.
  • Improved standards of living would lead to population growth.
  • Population growth would lead to decreased standards of living.

Most economists believe that mankind broke out of the population trap with the Industrial Revolution. In 1870, just as the Second Industrial Revolution was slowly picking up steam, global GDP per capita was int.-$1,261¹ ($1,435 in today’s prices). After millennia of stagnation, income had finally begun to grow steadily at around 2% a year. Less than 100 years later, output per capita had doubled twice, even as the world’s population increased from 1.3 billion to 3.3 billion.

The decisive change came in the 1950s; the global economy exploded, averaging an unprecedented 5% a year since then. Sure, there have been hiccups, but even the 2007–2008 financial crisis has only left a tiny dent in the ever steepening curve. The takeaway is clear: not only are humans now wealthier than they have ever been, but our incomes are also growing more rapidly than ever before.

Figure 1

The world is also healthier than ever before. At the beginning of the 19th century, the average global life expectancy was 28.5 years. By 1870, little had changed; an average human could still not expect to live for more than 30 years. But with economic development, the Second Industrial Revolution also brought rapid health improvements. By the 1950s, global life expectancy had increased to 50 years. And yet a global divide had opened. Life expectancies in Europe, North America, Australia, Japan, and Argentina had reached 60 (and in a few places even 70) years, while the rest of the world was still stuck at the 30-year mark. Since then, however, the divide has narrowed significantly. A baby born today can expect to live a whopping 72.6 years, longer than a baby born in 1950 in Norway, the country with then-highest life expectancy in the world.

In fact, one of the key drivers of the ever-increasing life expectancy is falling infant mortality. In 1800, 43.3% of children would not survive their first five years. Today, global infant mortality is 2.9%. Fewer women are dying during childbirth. Fewer people are suffering from communicable diseases, such as diarrhoea and various infections. In other words, humans not only live longer than ever before — the lives they lead are also healthier.

Figure 2

Our age is the Golden Age of education. In 1870, 19% of the global population could read. Today, 86% of humanity is literate, and 86% have at least basic education. People are also learning more than ever before. While in 1950 much of the world would spend fewer than 3 years studying, today vast majority of students stay in school at least 6 years, and often 9 to 12 years.

Figure 3

All this means that we are probably happier than we have ever been. The Human Development Index (HDI) measures human well-being along three dimensions: a decent standard of living, a long and healthy life, and knowledge. All three have been improving rapidly, and so has the HDI. But does a higher HDI mean that humans are happier? Now that, it turns out, is not an easy question to answer. While several metrics track self-reported happiness around the world, none have been around long enough to allow us to make strong conclusions.

The 2020 World Happiness Report suggests that happiness can be partially explained by GDP per capita, healthy life expectancy at birth, availability of social support, freedom to make life choices, generosity, and perceptions of corruption. On the one hand, we are now indeed richer, healthier, and freer than ever before. On the other hand, we are much more likely to live alone, while corruption and generosity trends are rather murky. In other words, we have quite a few reasons to feel happy, but humans are notoriously hard to satisfy, so the jury is still out on this one.

Figure 4

The world is terrible

It seems that the world is doing pretty well, right?

Not so fast. On average, life has indeed improved immensely over the past 150 years. But people are not averages. Your suffering and my happiness do not average to an alright life.

Take, for example, Equatorial Guinea, a tiny coastal country in Central Africa with a population of 1.2 million. At the height of its economic performance in 2008, the country was raking in int.-$40,368 per person per year, outperforming Japan, France, the UK, and numerous other rich countries. Its output has since dropped significantly, and yet Equatorial Guinea, one of sub-Saharan Africa’s largest oil producers, is still among the richest nations in Africa. It turns out, it is also one of the most unequal. While data is scarce, back in 2006 76.8% of the country’s population lived below the national poverty line of $2/day; by 2011, the rate appears to have fallen to 43.7%. Oil money hasn’t reached most Equatorial Guineans, instead lining the pockets of the country’s dictator Teodoro Obiang Nguema Mbasogo, who, according to The Financial Times, in 2016 “won his fifth seven-year term with his smallest share of the vote yet: 94 per cent”.

The Economist reports on the lifestyle of Teodoro Nguema Obiang Mangue (known as Teodorín), Vice President of Equatorial Guinea and son of the President.

So is the world more like Equatorial Guinea, spending its resource wealth on private jets, yachts, and Bugattis for the select few, or more like Norway, investing in healthcare, education, and long-term sustainability to benefit everyone?

One way to assess income inequality within countries is the Gini coefficient. The coefficient measures the extent to which a country’s income distribution deviates from a completely equal one. A Gini coefficient of 0 thus indicates perfect equality, while 100 means perfect inequality. A single glimpse at the map below shows that some countries distribute income relatively fairly, while in others the chosen few reap most of the benefits. In fact, not only inequality varies across countries — it is also distributed unequally itself, with Europe’s societies most egalitarian, and much of Africa extremely unequal.

Figure 5

Countries are also very unequal when it comes to the size of the pie they can distribute. Luxembourg, one of the richest countries in the world, has a GDP per capita that’s 150 times greater than that of Burundi. Just as states in Africa are more likely to be unequal, so they are also on average significantly poorer than most Western nations.

In practice, this means that while humanity has seen impressive economic progress, millions of people are still stuck in poverty. How many exactly depends on the definition. The World Bank defines extreme poverty as living on less than int.-$1.90/day. Extreme poverty is characterised by severe deprivation of basic human needs; in other words, below the destitution line, survival might well be impossible. Some economists argue that the current international poverty line is much too low. Peter Edward of Newcastle University, for example, has proposed the “ethical poverty line” of int.-$7.40.

Sidenote: int.-$

I’ve now mentioned int.-$ (i.e., the 2011 international dollars) in passing a few times. When we discussed the world’s GDP in 1870, I converted int.-$1,263 to today’s dollars and moved on. Int.-$1.90 is different. I will keep referring to it, so before we go any further, let’s make sure we all know what it means. And to do that, let’s consider burgers.

Every year, The Economist releases its Big Mac index. A few months ago in the US, you could buy a Big Mac for $5.67. In Switzerland, an identical hamburger costs CHF 6.50, or $6.70. To get the most bang for your buck, you would have to travel to South Africa: Big Macs there sell for ZAR 31, or $2.15. How is this related to poverty? Well, if you had $1.90, you would be better off living in South Africa than Switzerland, since burgers — and many other goods — are cheaper in South Africa. To make comparisons across countries (and time) possible, the international poverty line is thus adjusted for purchasing power parity (PPP). In other words, int.-$1.90 takes price differences into account, and should thus be interpreted as the amount of money equivalent to the consumption that $1.90 could buy in the US in 2011. To convert int.-$ to currencies rather more real, I’ll be using the World Bank’s PPP conversion factors and consumer price indices.

Back to poverty lines now. No matter which threshold you choose, one thing is clear: hundreds of millions of people around the world today struggle with the most basic necessities. 737 million people, or 10% of the world’s population, live on less than int.-$1.90/day. More than a quarter live on less than int.-$3.2/day. 46% of the world, a whopping 3.4 billion people, live on less than int.-$5.5/day.

Figure 6

These income disparities are also reflected in health outcomes. An “average” baby born in Western Europe and a few other countries today will celebrate her 80th birthday. In much of Africa, life expectancy is around 60 years. A baby born in the Central African Republic will die at the age of 53, more than 30 years earlier than a baby born in Japan. A baby born in Europe, Northern America, and a few Asian countries has a less than 1% chance of dying before she reaches five, while the Central African Republic, Chad, Guinea, Nigeria, Sierra Leone, and Somalia all have child mortality rates above 10%. In 2018, 5.3 million children died before their 5th birthday. That’s 15,000 young lives lost every single day.

Figure 7

Countries with the highest infant mortality rates also have the highest maternal mortality rates. In Sierra Leone, where 10.5% of children die before they reach five, and each woman gives birth on average 4.3 times, the lifetime risk of maternal death is almost 5%. Being a woman of childbearing age in Sierra Leone is riskier than serving in the US Armed Forces during the Vietnam war. More than twice² as risky.

Unequal health outcomes are not limited to mothers and babies. If you were born in the Central African Republic, for example, you would be much more likely to both die earlier than most other humans, and to suffer from a disability during your lifetime. While the leading cause of death and disability globally is cardiovascular diseases, they primarily affect people older than 70. Children, however, are much more likely to die from causes that are often easily preventable, such as lower respiratory and diarrhoeal diseases. In general, the global burden of disease falls disproportionately on sub-Saharan Africa, and, to a lesser extent, Asia.

Poverty and disease feed on each other. They also lead to other problems that might affect one’s future earnings, the ability to withstand economic shocks, and overall well-being. Here are just a few challenges that millions of people around the world (and especially in sub-Saharan Africa) face:

  • 820 million people are undernourished; a third of children in sub-Saharan Africa are stunted, indicating severe malnutrition. And while famine deaths have plummeted in the last decade, the UN has warned that the world is now on the brink of a “hunger pandemic”.
  • Around 1 billion people cannot read and write. While most European, Asian, and North and American countries have literacy rates close to 100%, Africa and South Asia are lagging severely behind. Chad’s literacy rate is a measly 22%; in twelve other African countries and Afghanistan, less than half of the population can read and write. These are also the countries with the lowest formal education enrolment rates.
  • Millions of children around the world work, primarily as unpaid family labourers. Some find time to attend school as well, while others will never receive an education. In 2013, for example, 5% of Bangladeshi children aged 7 to 14 had a job, working on average 36 hours/week, more than adults in the Netherlands (30.4 in 2019). In Cameroon, 62% of children were employed, averaging 21.1 hours per week.
  • Poor households often have limited access to sanitation and clean water, leading to more disease. 2.2 billion people around the world don’t have access to safe drinking water, 2 billion lack even the most basic sanitation facilities. The poor are also much more likely to die from indoor air pollution. For example, around 9% of all deaths in Madagascar could be prevented by switching to clean fuels for cooking.
Figure 8

In short, while humanity has experienced unprecedented improvements in well-being, billions of people are still struggling to make ends meet. And it gets worse.

The world is even more terrible

Let’s go back to the averages for a moment. In 2016, global GDP per capita was int.-$14,574. This roughly corresponds to the median standard of living in Costa Rica, an upper-middle-income economy. And yet in practice, 737 million people live on less than int.-$1.90/day. This means that the wealth that the world creates every year is distributed unequally. How unequally? Here’s one way of looking at it.

To truly understand the world’s income distribution, we would need to know how much everyone earns (or spends) in a year. If we then ordered the world’s 7.6 billion people according to their incomes, we could get a feeling of how unequal the world truly is just by looking at the shape of the curve. In fact, this is exactly what I did.

Figure 9 shows the annual incomes of every single person on Earth, from the extremely poor to the extremely rich. A few things stand out. First, a huge chunk of the global population earns very little. So little, in fact, that the livelihoods of the poorest few billion seem to be hovering around zero. Incomes begin to grow perceptibly at the three-billion mark. And just as we’re about to run out of people, earnings suddenly skyrocket. Yes, that’s us, the rich world.

Figure 9

Sidenote: Figure 9

Let’s pause here. Figure 9 is profound. So let me take a moment to explain how it came about. PovcalNet, a database maintained by the World Bank, provides income and consumption data for 164 economies; this includes not just average and median household spending, but also income distribution by population decile. For example, in South Africa, one of the world’s most unequal countries, the bottom 10% of the population earn less than 0.9% of the country’s income, while the top 10% take in almost 51%. Using this data, I calculated average incomes for every country-decile combination. For example, the 1st decile in South Africa has a mean annual income of int.-$347. Finally, I ordered all country-decile combinations by their average incomes. First on the list are the 1st decile earners in Haiti, averaging int.-$70/year (that’s ten times less than the extreme poverty threshold). The 10th decile earners from Luxembourg complete the list with int.-$80,283/year (corresponding to $91,246, or €79,567 today).

A few things to keep in mind as you consider the implications of the analysis:

  • This is either household consumption or household income data, not GDP per capita. According to Hellebrandt and Mauro (2015), only around 39% of GDP per capital translates into actual income to individuals, so consumption/income is much more useful when pondering poverty questions (in fact, the World Bank uses this data to track poverty levels around the world).
  • The underlying data provided by the World Bank is far from perfect. Drawn from national household surveys, it is not strictly comparable and might not work well “for other purposes, including tracing out the entire distribution of income”. Nevertheless, it is the best data source available.
  • Finally, while deciles are vastly more useful than country-wide averages, they are nevertheless only rough approximations. Haiti is relatively small, and thus each decile there contains around 1 million people. Even then, hundreds of thousands in the 1st decile must be living on less than the average int.-$70/year. China’s decile, on the other hand, corresponds to 138 million people. In such a case, very little of the intra-country income variation can be captured.

In general, while none of these challenges should affect the overall conclusions, take the exact numbers provided here with a grain of salt. And, if all this sounds interesting, check out the underlying analysis.

Let’s get back to the data now. Figure 9 suggests the world’s annual income is indeed distributed very unequally. Just how unequally? Here’s one striking fact: the poorest 10% of the people together earn less than 1% of the world’s income. The richest 10% consumer almost half of it.

In fact, the rich are so rich that they could eradicate extreme poverty tomorrow. Yes, you read that right. Tomorrow every single one of the 737 million extremely poor people could wake up with enough money in their pockets to afford shelter, sanitation, and food.

To do that, we could introduce a 100% progressive tax across the world on annual incomes above int.-$71,800. If we then distributed the resulting int.-$171 billion (194 billion in today’s dollars)³ to the world’s poorest, every single human on the planet would have at least int.-$1.90/day. Not a lot, and yet life-changing. And in fact, we could do even more. If we redistributed all income above int.-$52,000/year to the poorest, everyone would live on no less than int.-$3.20/day, the World Bank’s minimum set for lower-middle-income economies.

And if int.-$52,000/year doesn’t sound like a lot, blame PPP, for that exact amount happens to be the median income in Luxembourg, the richest country on Earth.

Alright, enough of int.-$, let’s convert all this into today’s dollars. Here’s what I’m saying:

Let’s redistribute 100% of American annual after-tax incomes above $81,605.

Let’s redistribute 100% of British annual after-tax incomes above £57,123.

Let’s redistribute 100% of Swiss annual after-tax incomes above CHF 97,110.

Let’s redistribute 100% of German annual after-tax incomes above €62,020.

Let’s redistribute 100% of Japanese annual after-tax incomes above ¥8,272,536.

Let’s redistribute 100% of Lithuanian annual after-tax incomes above €37,375.

If we did that, extreme poverty would be no more, and the world would look almost imperceptibly, and yet also radically less unequal.

Figure 10

Of course, a 100% tax rate is unrealistic. It could be considered unfair, and there’s a good chance high earners would just start earning less. After all, if most of your income is taken away, it might be difficult to find the motivation to keep pushing. But that’s not the point I’m making here. This is: lifting 737 million people out of extreme poverty is not only doable — it’s also quite easy. Instead of a 100% progressive tax rate, we could have a 92% progressive tax rate. Introduce a wealth tax. Redirect some of the military spending. Reconsider religious donations. In other words, we can handle $194 billion a year — it cost more to put Neil Armstrong and Buzz Aldrin on the Moon.

So here’s the thing. We haven’t eradicated extreme poverty not because we don’t know how to do it, nor because we lack the resources to do it. We haven’t eradicated extreme poverty because we have chosen not to.

The Great Dissonance

When asked about the world’s greatest challenges, people unfailingly bring up poverty. Some are worried about the economy at home, others brood over basic human needs abroad; even millennials, concerned with climate change and the destruction of nature, agree that poverty and inequality are as important. And yet extreme poverty not only exists — with coronavirus ravaging poor economies, 265 million more people could be pushed to the brink of starvation by the end of 2020. Why, then, are we failing to help those most in need?

Talk is cheap

One obvious answer is that talk is cheap. In 2005, President George W. Bush urged rich countries to increase their spending on development aid to 0.7% of gross national income (GNI), a level agreed upon at the Monterrey Conference in 2002. That was not the first — nor the last — time President Bush “stood on the world stage and promised to sharply increase development assistance to poor nations”. The US has indeed propped up its foreign aid over the past 15 years. To 0.18%.

And it’s not only countries, institutions, or politicians who promise, then don’t deliver. Humans in general are notorious for their tendency to procrastinate if they can get away with it. One widely discussed study published in Psychological Science in 2009 concluded that by simply announcing our goals to others we are immediately rewarded by a sense of accomplishment, and thus end up being less likely to actually achieve the goal. A related phenomenon is known as the “warm-glow”. Altruism makes people feel good; so good that it might lead to people giving to inefficient causes, or to giving less than would be rational.

The cultures that (don’t) give

Some cultures might be more conducive to giving than others. In Islam, for example, alms-giving (known as zakat) is one of the five pillars, or religious obligations, that must be observed by everyone. Muslims are required to donate 2.5% of their wealth to those in need; in some countries, the tax is collected by the state. It is estimated that the global volume of zakat amounts to tens, and possibly even hundreds of billions of dollars every year. Perhaps not surprisingly, Indonesia topped the World Giving Index in 2019.

On the other side of the spectrum are the post-Soviet republics. The USSR all but killed the tradition of alms-giving in the occupied states. “The mythology of the all-sufficient state left no room for philanthropy”, wrote the New York Times in 1988. The US, while a charitable giving mammoth in absolute terms, does not contribute its fair share either. Indeed, President Donal Trump’s slogan “America first” and his disgusting attitude towards developing countries does not bode well for the world’s poor.

That said, President Trump only brings up poor countries when talking about immigration, terrorism, and viruses. Neither he nor most other leaders of rich capitalist societies consider global poverty a topic worthy of national attention. Global poverty is simply not on the political agenda. Except, of course, when the deficit question comes up.

Making momentous mistakes

Now here’s another challenge: even when we do give, we don’t always do a good job at it. When billions of dollars and millions of livelihoods are involved, mistakes can be costly. William Easterly, a renowned development economist, has written a 400-page book on the failings of the Western aid, and he’s not alone. He is also right: many aid projects have indeed ended in abysmal failure. Extensive investments have been made into programs with unknown or non-existent effects; implementations have been botched; and, on an occasion, more harm than good has been done.

The World Bank, for example, helped finance an oil pipeline in Chad in the early 2000s, expecting investments in infrastructure, education, and poverty reduction. Instead, the country’s ruling elite used the money to buy weapons. Water pumps in the form of merry-go-rounds turned out to be expensive, slow, difficult to operate, and demeaning. The International Monetary Fund continues insisting on structural reforms that a growing number of economists consider outright harmful. Aid from the World Bank often leads to a marked increase in the country’s deposits in foreign financial havens. The list could go on forever.

My favourite example, however, was described by last year’s Nobel Prize winner Abhijit Banerjee in Making Aid Work. The Gyandoot program in Madhya Pradesh, India provided internet-enabled kiosks for accessing government services in rural areas. Most kiosks didn’t work due to poor connectivity or… lack of electricity. The original evaluation published by the World Bank concluded with a hopeful if misguided “Following the success of the initiative…”

Giving from the bottom of the heart

Such questionable impact evaluations are not limited to large development institutions. In fact, individual donors are probably even more predisposed to giving with their hearts rather than minds. For example, would you rather donate to a charity that distributes books to African kids, or to one that treats parasitic worm infections? Books are much more fun; it’s something we can all understand and relate to. And so in 2019 Books For Africa’s income was $39 million. SCI Foundation, on the other hand, only received $19.5 million. The trouble is, SCI Foundation’s impact is proven, its approach highly cost-effective. When it comes to improving lives, deworming is pretty much impossible to beat. Books might make donors happy, but they appear to make no difference to the recipient’s educational outcomes.

And yet while the above example is real, it is not exactly representative. Most individual donations in rich countries don’t go overseas; instead, they go to religious institutions (a whopping $124.52 billion in the US in 2018). I don’t know about you, but I am pretty damn sure the way to heaven is saving lives and lifting fellow humans out of poverty, not the Sunday morning cash handout. (A fascinating side note: while the Catholic Church is not required to disclose its financials, making the calculation of its true wealth all but impossible, in the US alone it was estimated to have annual spending of $170 billion back in 2012.)

But when it works…

So, is all our aid spending inadequate and misguided? Far from it. While we have made mistakes — and probably too many of them — there have also been wonderful achievements. With financial and scientific backing from the US, India underwent the Green Revolution in the 1960s, adopting high yielding variety seeds, irrigation, fertilisers, and other modern agricultural technologies, leading to increased food security, higher-quality diets, and a reduction in poverty. In 1980, the World Health Organization (WHO) declared the world smallpox-free, two decades after Viktor Zhdanov, a Ukrainian virologist, initiated an organised global effort. The WHO is now eyeing malaria, the disease that kills more than 400,000 people every year, disproportionally affecting some of the poorest countries on Earth.

We have also recognised that while the huge leaps, such as smallpox eradication, matter immensely, we can also do a lot of good by sometimes focusing on smaller, more tractable problems. Last year’s Nobel Prize in Economics went to the three economists who introduced the use of randomised controlled trials to development. Instead of hoping that an intervention will work, scientists, policymakers, and donors can test it to make sure it does. For example, Abhijit Banerjee, Esther Duflo and co-authors found that microfinance, despite its worldwide fame, had no effect on health, education, or women’s empowerment. Michael Kremer and co-authors showed that providing textbooks did not raise average test scores (yes, Books for Africa, I’m looking at you). Providing deworming drugs to schools didn’t either, but it did reduce school absenteeism by a quarter.

Similar ideas have been gaining popularity in individual giving. When I donate, I don’t have to make a decision based on the attractiveness of the charity’s website anymore. Instead, I can use GiveWell. GiveWell is a charity evaluator with a strong focus on evidence-backed impact. Every year, it spends thousands of hours reviewing candidate organisations and recommends those that have proven to be well managed, exceptionally effective, and highly cost-efficient. For example, in the past 6 months, I donated around $15,000 to the Against Malaria Foundation (AMF). I know that this enabled AMF to pay for 7,138 anti-malaria nets that will be distributed in Togo and Papua New Guinea, protecting 12,848 people. Because its operations are well-understood, well-documented, and robustly evaluated, I can estimate that this translates to averting the deaths of 9 children⁴.

To sum it all up, we could eradicate extreme poverty if we truly wanted to. Instead, we end up talking too much, doing too little, and making embarrassing mistakes. Nevertheless, we have also seen awe-inspiring progress: increasing agricultural productivity, eradicating debilitating diseases, and saving lives. We should do more of that. But we should also be more thoughtful about doing it.

Taking a Second Look at Aid

Aid is often broken down by donor type and sector. For example, in 2018 the US government disbursed $46 billion of foreign aid, double the $22.9 billion donated by individuals, foundations, and corporations. More than 30% of the $46 billion was spent on security efforts, much of that while engaging in military conflicts in the Middle East.

For a more global perspective, we could look at aid through the lens of donor countries (e.g., Turkey’s foreign aid amounted to 1.15% of its GNI, or almost $9 billion in 2019) and recipient countries (e.g., Syria received almost $10 billion in aid in 2018).

A very different way to look at aid, but one that offers interesting insights, is the scale of the intended purpose. Let’s consider four levels of scale: global, national, regional, and individual.

Solving global problems

Some challenges are inherently global, often transcending not just country borders, but continental boundaries as well. Nevertheless, many of them disproportionally affect developing countries. Take, for example, HIV/AIDS.

First identified in 1981 in the US, HIV has killed 32 million people around the world. In the 1980s, HIV was a death sentence. In sub-Saharan Africa, it is still very much a pandemic. For example, in South Africa, the virus accounts for more than 25% of all deaths. Elsewhere in the world, however, HIV has turned into a chronic disease that can be managed.

By the end of the 1980s, a complex public-private partnership had led to the creation of the first antiretroviral treatment (ART). It took another decade to reach an agreement that would make the drug available to poor countries at the cost of production ($350 per patient per year) rather than the market price ($10,000 per patient per year). ARTs are now used by 62% of those infected in low- and middle-income countries. Before the invention of ART, the average life expectancy following an AIDS diagnosis was one year. With ART, someone with HIV can expect a near-normal lifespan. It is estimated that by 2016 ART had averted more than 10 million deaths.

Figure 11

HIV is a good example of an immense challenge that affects developing countries disproportionally, and one that these countries could not have even begun to tackle without outside support. Indeed, the US alone spends more than $30 billion on HIV every year.

But HIV is just one problem. Many other illnesses, such as malaria, polio, or neglected tropical diseases require global cooperation to discover treatments, reduce their costs, and find efficient delivery approaches. And these challenges extend beyond health; we need to invest in high-yield seed research, cheap clean energy, climate change adaptability, and more. And while all these problems are global, we also need to make sure the solutions are local. After all, it turns out that ART doesn’t work well in undernourished patients. Back in 2000, that would have disqualified more than a quarter of the population in sub-Saharan Africa.

Building national infrastructure

Some challenges are decidedly national. Consider electricity, a seemingly basic necessity that 860 million people in the world still lack. Electricity plays a crucial role in economic development, and yet electrifying a country is a monumental undertaking (it took Afghanistan, one of the fastest economies ever to electrify, 10 years to increase coverage from ~25% to almost 100%). Problems such as these can only be addressed by national governments. The market won’t step in to provide electricity to remote populations at a loss, and no single region is likely to have the funds necessary to build a local grid.

In Common Wealth, a prominent development economist Jeffrey Sachs suggests that countries need different capabilities at different stages of development. Subsistence economies, those on the lowest rung of the development ladder, should focus on agriculture, roads, power, health, and primary education. Commercial economies should promote business by developing ports, telecommunications, and secondary education. By the time a country reaches the last rung, it has long since graduated from being an aid recipient and should invest in tertiary education, science, and quality of life.

Now while measuring the number of lives saved through ART is rather straightforward, one frequent challenge of large-scale projects is impact evaluation. What is the value of the national electricity grid? (I won’t go into details, but in short, electrification leads to improved living standards, while its effect on economic growth is disputed.) It thus seems reasonable for governments to focus on large projects that:

  1. Are expected to bring value over and above the investment costs,
  2. Have little risk associated with them (so there’s little chance of losing a lot of money),
  3. Will provide value soonest (so the country can start reaping the benefits right away), and
  4. Have fewest dependencies, especially early in the country’s development cycle.

In short, focus on the sure building blocks of a well-functioning society at scale.

Nailing local implementation

While the national government is busy building the basic infrastructure, the crucial job of figuring out the details is often left to regional governments. For example, the central government might allocate funding for building schools and hiring teachers. But how can we make sure these teachers actually show up to work?

Projects at this level can often benefit from robust evaluations through randomised controlled trials: the scope is narrow, the problem well understood, the goal clear and widely agreed-upon.

And, by the way, it turns out that:

  1. When teachers do show up to work, pupil test scores increase significantly, and
  2. Instructing teachers to have a student take a picture of the teacher and other students twice a day is a cheap and effective way to significantly reduce teacher absenteeism.

Living the best individual life

At the end of the day, however, global, national, and local investments only matter if they benefit individuals, adequately and fairly. Yet even in the best-run, richest countries, things go south once in a while. A pandemic hits. Someone gets sick. Loses a job. In the developing world, such personal disasters are both more terrifying and more wide-spread. The power grid, while a wonderful invention, won’t help, at least not in the short turn. One of the most valuable services that a government can provide is thus social security. Paid by everyone, available to everyone (though some countries would argue that’s much too generous).

The world is divided when it comes to social security. For example, Europe, North America, much of South America, and even quite a few countries in Asia have nearly universal health insurance coverage. Much of Africa has close to none. If you were to get sick in Angola (where health insurance coverage is 0%), you would have to pay for the treatment yourself. And yet with monthly median consumption equivalent to $70 in today’s prices, you would have to think twice before pursuing medical help.

Figure 12

While annual healthcare expenditure per capita is rising across the world, it will still be a while before most countries can offer their inhabitants affordable health insurance, unemployment protection, and adequate retirement benefits. In the meantime, donors can step in to help individuals in need, just like they help address global, national, and local challenges. Indeed, there are already ways to do just that. Microfinance, while not a silver bullet, does provide financing to people who would not be eligible otherwise, allowing them to start businesses or cover unexpected medical expenses. Grameen bank, established by the 2006 Nobel Peace Prize winner Muhammad Yunus, disbursed almost $3 billion in 2018. That’s $3 billion loaned to people who would otherwise have to borrow from loan sharks or forego the often vital expenditures.

An even better (but also more expensive) form of aid is cash transfers. In fact, many social safety net programs in sub-Saharan Africa are fully or partially funded through foreign aid, and even individual donors can help in situations where national safety nets are lacking. For example, the largest cash transfer nonprofit GiveDirectly has delivered $160 million to 170,000 poor families in Africa since 2009. With COVID-19 destroying livelihoods, GiveDirectly is now expanding faster than ever.

By the time you’re done reading this 17,000-word monstrosity, I hope I will have convinced you that:

  1. Cash transfers are one of the best ways to lift people out of poverty.
  2. We should drastically increase the amount of foreign aid dedicated to cash transfers.
  3. We should build a global welfare system based on cash transfers.

But before I do that, let me first summarise my stance on aid. Because I am definitely not saying that we should drop everything and just do cash transfers. Indeed, all four levels of aid are needed if we were to eradicate extreme poverty. Each is responsible for a particular set of problems that cannot be easily solved at a different level. Each also interplays with all others in complex, mysterious ways.

But I am saying that we should focus much more on individuals. Because if we want to eradicate — or even significantly alleviate — poverty in the next decade, global, national, and locals solutions won’t be enough. Addressing agricultural efficiency challenges will take years. Building the irrigation infrastructure will take years. Rolling out local fertiliser programs will take years. Waiting for the benefits of these investments to trickle down to the world’s poorest (if the benefits do trickle down) will take years.

So here’s an idea. Instead of waiting for years until all pieces of the puzzle finally snap into place, let’s flip the problem of poverty on its head. Let’s eradicate poverty now, then spend years building the long-term infrastructure.

Cash Transfers: An Overview

At the outset, cash transfers are simple: a benefactor sends money to an eligible individual. That’s it. That’s a cash transfer. In practice, pretty much everything about a cash transfer can be customised, and the following four dimensions in particular are worth a deeper look.

First, cash transfers can be conditional or unconditional. Recipients only receive conditional transfers if they fulfil specific requirements (e.g., school attendance), whereas unconditional transfers are distributed to selected participants without any strings attached. Second, cash transfers can be one-off or recurring. Whether one type or the other is more appropriate depends on the goals of the transfer program. Third, the amount transferred can either be sufficient or insufficient to cover a person’s basic needs; in the domain of universal basic income, this is referred to as either full or partial basic income. Fourth, cash transfers can be universal (available to everyone) or targeted (i.e., based on some definition of need).

I’ll discuss each of the four dimensions in more detail to consider:

  • When the various options should be used,
  • What the challenges associated with these options are, and
  • How each of the options might affect program outcomes.

But before we start, let me tell you about Malawi.

Supporting 1.2 million poor Malawians

Malawi, a landlocked country in Southern Africa, is one of the continent’s poorest nations: 70% of its 18 million inhabitants live below the international poverty line. In 2006, its government piloted Social Cash Transfer Programme (SCTP), a donor-supported unconditional cash transfer program. Today the program covers the entire country, assisting nearly 300,000 ultra-poor and labour-constrained households and an estimated 1.2 million individuals. Transfer amounts vary with the size of the household and the number of children in school; a single-member household currently receives MWK 2,600 ($3.50) per month. While the transfers corresponded to only 18% of the baseline consumption of beneficiary households in 2013, the Malawi government has also used the program to distribute additional money during times of hardship, such as the recent flooding and drought in Nsanje, Neno and Phalombe districts.

SCTP has been studied extensively, and has been shown to lead to a wide range of positive outcomes for recipients. First and foremost, the transfers are associated with a significant increase in household expenditures (Miller et al., 2011). Recipient households spend the additional income on more and more diverse food, and on long-term investments, such as chickens, goats, and cattle (Covarrubias et al., 2012). Children miss fewer school days and work less outside the household (though the gains are partially offset by a slight increase in time spent on chores at home) (Miller and Tsoka, 2012). Similarly, adults spend significantly less time in ganyu (informal labour). Bastagli et al. (2015) argue that this is a positive development, as ganyu is a survival strategy, with jobs never guaranteed, and rates far from sufficient to invest in sustainable livelihood development.

The silver bullet

Cash transfer programs are probably one of the most-researched development interventions. They are also extremely diverse. While Malawi distributes monthly allowances to its poorest households with no strings attached, Indonesia targets poor households with children or pregnant women and insists on numerous conditions that cover healthcare and education. In fact, most social security systems throughout the world are based on cash transfers. To keep the discussion focused, it is thus important to remember the goal of this investigation: to explore whether cash transfers can eradicate extreme poverty around the world before sustainable long-term solutions are in introduced.

With that, let’s review the evidence.

Bastagli et al. (2015) evaluated a whopping 165 studies on cash transfers published between 2000 and 2015. The conclusion is unambiguous: cash transfers lead to significant reductions in poverty. This manifests itself in increased overall household expenditure, as well as increased expenditure on food. There is also some evidence that cash transfers encourage savings and investment in productive assets.

In addition to direct effects on consumption, cash transfers are often associated with other highly positive effects. For example, cash transfer programs often increase short-term school attendance of both boys and girls. Evidence on learning and cognitive development is rather more mixed, likely because these outcomes depend on the quality of education received, not just school attendance rates.

Cash transfers also lead to an increased usage of healthcare services, but only when these transfers are conditional. While this effect might be unwelcome in rich countries, it is generally assumed that in poor countries such visits lead to improved health and are thus encouraged. Unfortunately, no data on health outcomes is available. Studies of anthropometric (i.e., body-measurement-related) outcomes, however, find no significant improvements, even though there is strong evidence for both increased food expenditure and improved dietary diversity. In some cases, this could be explained by limited local supply: Langendorf et al. (2014) found that in Niger the receipt of nutritional supplements in addition to cash transfers led to a 50% drop in the incidence of acute malnutrition in children compared to receiving the cash transfer alone.

Finally, there is compelling evidence that cash transfers lead to female empowerment. Women in recipient households tend to get involved in household decision-making more often, marry later, have fewer sexual partners, use contraception more often, and have fewer children. They are also less likely to be sexually or physically abused by their partners.

To summarise, cash transfers reduce poverty, improve diets, boost school attendance, and lead to female empowerment. In other words, they rock.

No strings attached

Conditional cash transfers are employed when donors have goals beyond poverty alleviation. Bastagli et al. (2015) conclude that conditions may lead to higher impact in areas targeted by the conditions. For example, increased usage of healthcare services has only been observed when transfers were conditional on such visits. Often, however, no or only small differences between conditional and unconditional transfers are found. For example, while some studies have shown that conditional transfers are associated with slightly better educational outcomes, differences are not always significant.

Now while both conditional and unconditional cash transfers lead to significant improvements in recipient well-being, conditional transfers have significant drawbacks. First, they impose constraints on recipient decision-making. This is not only paternalistic (I’ll come back to that in the next section), but also inefficient. Demanding that recipients spend their money on healthcare when they’d much rather invest in a cow means preventing program participants from choosing the best possible outcome given their circumstances.

Second, they limit the pool of recipients, often in ways that are highly unfair (Lawson et al., 2017). If there are no schools nearby, preventing a family from enrolling their children in primary education, that does not mean they are less deserving of the transfer. In all likelihood, they need it more than others.

Third, conditional cash transfers might not be the most cost-effective solution to problems other than poverty, especially in poorer countries. Ladhani and Sitter (2018) point out that education and healthcare often need supply-side rather than (or in addition to) demand-side solutions. In the previous example, instead of paying a family to send their children to school, the government might well be better off spending the money on building a school where one doesn’t exist. In the same vein, demanding that transfer recipients use services that are of low quality might be not only ineffective but also damaging.

Finally, monitoring compliance with conditions introduces huge overheads. White et al. (2013) conclude that monitoring might account for a quarter of a program’s administrative costs. However, the true costs of compliance are often much higher, as studies rarely take into account substantial time commitments required from both transfer recipients and non-program staff (e.g., teachers). For example, Lawson et al. (2017) point to a program in Columbia where recipients spent an estimated 45% of the transfer collecting compliance documentation.

In short, conditional transfers might have a place in social security systems (though I am beginning to seriously doubt that). But conditional transfers are not the best tool when the goal is poverty reduction; that’s the purview of unconditional cash transfers.

The never-ending story

The choice between one-off and recurring transfers is context-specific as well. Indeed, there are cases where one-off transfers are not only acceptable but also preferred. For example, in the aftermath of a disaster, recipients can benefit from a single transfer to help them address their immediate needs, be it food, shelter, or medical services. More generally, Bastagli et al. (2015) suggest that frequent transfers favour consumption smoothing and spending on smaller assets, while lump-sum payments might be associated with larger investments. Now poverty is not a temporary humanitarian crisis. Nor is it a state of being one cow away from wealth. Instead, it is a long-term condition caused by a variety of factors, vast majority of them not controlled by those in need. Without a doubt, recurring payments are thus preferred.

Recurring transfers have several advantages over one-off payments. First, they can help smooth consumption when things go south while leaving recipients the flexibility to save or invest when incomes grow. Second, they offer a sense of security; a drought or a flood is still as terrifying, but at least the aftermath will not lead to starvation. Third, recurring payments are more appropriate in a wider set of recipient contexts. One-off payments are necessarily large and thus require that the recipient invests the transfer, yet good investment opportunities are not always available (Farrington and Slater, 2009).

In short, while a single transfer can help alleviate poverty, poverty eradication requires recurring payments.

A glass half full

As could be expected, larger transfers are associated with larger expenditures, greater poverty reduction, and increased savings and investments. On the other hand, larger transfers appear to have no effect on educational outcomes, and might also be associated with some reduction in employment.

Bastagli et al. (2015) suggest that mixed evidence on education might arise due to external factors, such as the available supply and quality of schooling. The evidence for reduced employment is similarly context-dependent. In the particular programs evaluated, that context was of family members working less to take care of children or the elderly. In such cases, reduced employment could be seen as a positive outcome that alleviates the overall work burden of a family with dependents.

The choice is thus again suggested by the goal. If we were to eradicate extreme poverty, the size of the cash transfer should match the recipient’s poverty gap. In practice, of course, the size of the transfer is much more likely to be decided by the available resources than the extent of the need.

The right tool for the right job

Now I believe in universal basic income as a tool to eradicate poverty, reduce inequality, improve well-being, encourage creativity, manage job automation, and more. And yet in the context of global poverty, the universality of cash transfers has a few glaring problems.

First, it is probably not feasible. Eradicating extreme poverty will be quite a feat. Eradicating extreme poverty while also paying everyone else is just not going to work — at least not in the near future.

Second, given limited resources and the extent of global poverty, it would be unfair to distribute the available money equally to everyone. We would be taking away from the poorest members of society while adding little value to those who are already wealthy.

And then there’s politics. White et al. (2013) point out, correctly if cynically, that certain targeting methods are much more likely to generate broad support. Universal is not one of them.

In other words, targeting is a necessity. It does not, however, come cheap.

First, there are the administrative costs of identifying potential recipients. On the one hand, a complicated poverty index might be difficult to compute and keep up to date, especially if it requires multiple inputs. On the other hand, while some targeting mechanisms are much simpler, they are also more likely to exclude some of the poor. At the end of the day, this is the question of balancing fairness and costs — Grosh et al. (2008) show that targeting accounted for 24% to 88% of administrative costs (or 0.6% to 6.3% of total costs) across eight evaluated programs.

Second, targeting often imposes costs on recipients. Participants might be required to collect documents, fill an application form, attend an interview, or even pay a small application fee. Just like with compliance monitoring, these costs can accumulate quickly.

A curious (also often baffling, ridiculous, and inappropriate) approach to targeting that is especially hard on recipients is self-targeting. A self-targeted program is open to many but is designed to primarily attract the poor. The mechanisms that discourage the non-poor from participating might include substantial costs (e.g., manual labour), unattractive benefits (e.g., low-quality food), and social stigma.

For example, the stigma associated with the US food stamp program is widely felt⁵, even as 1 in 9 Americans benefited from the program in 2019. Similarly, these past few months have seen heated debates about the arbitrary roadblocks enacted to deter people from applying for unemployment benefits. Indeed, it seems that self-targeting is rarely appropriate, and should definitely not be used in cases where reaching as many poor people as possible is the ultimate goal.

Third, targeting will necessarily leave out some people who desperately need help. And the problem will likely get worse as the identification criteria are simplified. Having reviewed 85 targeting mechanisms, Coady et al. (2004) conclude that while a median program transfers 25% more money to the target group than would be the case with a universal allocation, in 21 cases the employed targeting mechanism was worse than random. And to add insult to the injury, poorer governments appear to be doing a worse job targeting the poor than their richer counterparts.

In short, given limited resources, targeting is a must, and yet it comes with numerous challenges, including high costs to both governments and recipients, and a high propensity to exclude those most in need.

Alright, but won’t they spend it all on cigarettes?

Ah, I’m glad you asked. Cash transfers might indeed seem too good to be true. Less poverty, better nutrition, improved school attendance, and more. There must be a downside to all this, right?

Cash Transfers: The Objections

There are no silver bullets nor free lunches. Even though some interventions are indeed more cost-effective than others, all come with a price tag. But before I get into objections, let me remind you that all these conclusions, while based on a review of robust, experimental evaluations of existing programs, might still fail to apply in specific situations.

Evidence limitations

For example, much of the evidence comes from Latin America, while extreme poverty eradication will necessarily focus on sub-Saharan Africa. Similarly, conditional cash transfers have been studied more widely than unconditional ones. Most importantly, however, all evaluations so far have focused on near-term effects. For example, it is clear that people who receive transfers eat more nutritious food. It is not clear, however, whether they also end up healthier. They might not because health, like education, is affected by a range of factors. Or they might, but no study has yet been run long enough to capture these long-term effects.

In short, just like people are not averages, so cash transfer programs aren’t either. While there is clear evidence in favour of cash transfers in general, program design, implementation choices, and context matter a great deal. After all, Albania’s Ndihma Ekonomike managed not only to spend 6.3% of total program costs on targeting, but also led to a significant drop in recipient employment, a decrease in monthly household expenditures, and an increase in poverty (Dabalen et al., 2008).

With that in mind, let’s go over some of the most prevalent cash transfer objections.

Will cash transfers lead to…?

Dependency? Unemployment? Waste? All these questions have been asked before. And indeed, it is important that we ask them. This is the only way we can be sure that all these programs truly work. But there is a dark side to this.

In most cases, poverty is not self-made. The most important factor that determines whether somebody is rich or poor is their place of birth. Even in the developed world. Take Chicago, one of the most racially and ethnically segregated cities in the US. The life expectancy of a newborn in Streeterville is 90 years. In Englewood, a mere 45-min metro journey away, a baby born on the same day is unlikely to celebrate her 60th birthday. In other words, poor people are not lazy, foolish, or otherwise flawed. Rather, they lack the opportunities that the wealthy take for granted.

Figure 13

So while asking questions is important, it is also important to be fair and thoughtful while doing so. If you were so poor your children had to go to bed hungry every night, would you spend a cash transfer on cigarettes? No? Then why would you think that someone else would?

And here’s another thing. Just like it’s not alright to see others as inferior human beings, it’s also not alright to impose one’s will on them. How could a wealthy, healthy, employed Westerner know better how to survive in sub-Saharan Africa than a person who has spent their entire life there? We deceive ourselves when we think we can make good decisions for the poor. The poor know better, and cash transfers (rather than cow transfers, corn transfers, or cheese transfers) give them the dignity and the resources to make those decisions.

With that, let’s begin the grilling.

… waste?

A common argument against cash transfers is that recipients will spend them on bads (e.g., tobacco) or luxury items. In the words of a senior government official in Nicaragua (Handa et al., 2018):

Husbands were waiting for wives to return in order to take the money and spend it on alcohol.

Now we’ve already seen ample evidence that cash transfer recipients spend significantly more money on food, while also managing to invest in livestock and durable goods. Clearly, there’d be little left for “unworthy” purchases. Indeed, according to a review of eight studies by Handa et al. (2018), spending on temptation goods does not change after cash transfers are introduced. In fact, an older survey by Evans and Popova (2014) evaluated 44 estimates of alcohol and tobacco spending, and only found 2 that indicated a statistically significant (but minor) increase, while 12 pointed to a significant decrease. All others found no perceptible change in either direction.

Myth busted.

… increased unemployment?

There is a widespread perception that cash transfers lead to dependency and thus reduced employment. From a fascinating survey of elite attitudes towards poverty and cash transfers in Malawi (Kalebe-Nyamongo and Marquette, 2014):

This is not sustainable and bleeds laziness. When someone gives you money you have not planned for, you can’t think of what you want to do with that money.

This belief is one of the few that are actually based on economic theory. Cash transfers provide additional income, thus reducing incentives for recipients to work; that’s the income effect. Cash transfers might also change the effective wage rate; that’s the price effect. In theory, it is possible to construct a cash transfer program that would discourage participants from working at all. As we know, however, microeconomic theory and reality seldom have much in common.

Indeed, the vast majority of studies reviewed by Bastagli et al. (2015) found either no change in adult employment, or a marked increase in employment among transfer recipients. In a few cases where employment did go down, it was the result of old-age pensions, allowing the elderly to finally retire. At the same time, cash transfers appear to reduce child employment — but hopefully, that’s a good sign.

Handa et al. (2018) have also looked at the types of work cash transfer recipients do. First, they similarly conclude that cash transfers do not discourage recipients from labour force participation. Second, they note that transfers provide participants with some flexibility in allocating their productive time. For example, recipients, when possible, switch from casual day labour (unstable and often back-breaking) to work on their own farms.

In short, there is no evidence that cash transfers lead to reduced adult employment; in fact, in some cases, participants end up working more.

Myth busted. But before I tackle the next one, let me ask you two questions.

First, given that cash transfer recipients are often elderly, disabled, or mothers with young children, would it be such a bad outcome if employment did go down?

Second, would you yourself not want to sometimes work less? You probably would; after all, we invented the limited workday, the lunch break, and the vacation allowance so that we could, once in a while, take a deserved break. In fact, while data is scarce, it seems that the rich world spends significantly less time at work than the developing countries. So it’s alright to cut people some slack.

Figure 14

… increased fertility?

Some policymakers believe that targeting families with young children will lead to increased fertility. Because, you know, cash transfers make rearing children profitable.

While this sounds ridiculous, high fertility is a huge challenge in poor countries. Exactly because bringing up children costs money, every additional child will push a family further into poverty.

Handa et al. (2018) have reviewed four studies, concluding that cash transfers do not lead to increases in fertility in recipient households. On the contrary, in certain cases, cash transfers were associated with a delay in the first pregnancy and an increase in the spacing between births. In fact, Hindin et al. (2016) conclude that when it comes to reducing unintended pregnancy rates, cash transfer programs work better than other reviewed interventions (that said, reliable evidence is rather sparse and so it is possible that more effective interventions have simply not been studied robustly).

In conclusion, cash transfers do not lead to higher fertility, and in some cases might actually reduce it. Or, as a female participant of a Turkish cash transfer program remarked (Grosh et al., 2008):

Allah, were there women that got pregnant just for this money, really? Ha ha ha!

Again, myth busted.

… increased inflation?

An unintended side effect of cash transfers could be inflation in small, isolated communities. Indeed, if cash transfers in an area are not followed by increases in supply, prices will rise, both reducing the value of transfers to recipients, and crowding out non-recipients.

Handa et al. (2018) review evidence on both national and local economies and conclude that cash transfers do not lead to inflation. Unchanged prices are explained by three factors:

  1. Cash transfers coverage is relatively low (~20% of households in evaluated programs),
  2. Cash transfers amount to a very small proportion of the overall economy, and
  3. Markets, even in rural areas, are reasonably well connected, so increased demand can be easily met.

However, the inflationary effects of cash transfers have been observed in the past. Peppiatt et al. (2001) document several such cases in the 1980s. For example, a cash transfer program administered by UNICEF during the 1983–1985 famine in Ethiopia led to the doubling of maize prices in some regions. Similarly, Filmer et al. (2018) found that cash transfers have led to a 6–8% increase in the prices of protein-rich perishable food in Philippine villages. This was further associated with an 11 percentage point increase in the stunting of non-beneficiary children. On the other hand, numerous other countries have experienced no changes in prices under similar circumstances.

The key takeaway is that inflation is caused by limited supply rather than cash transfers per se. Most local markets are well connected, often spanning country borders, and would thus respond to increased demand by increasing supply. In the most remote areas, however, inflation is something to keep in mind.

Fine, but aren’t there better options?

Not only cash transfers appear to be effective in reducing poverty, they do so without any obvious side effects. Money is not spent on temptation goods, unemployment does not shoot up, fertility does not skyrocket, and inflation does not ravage local economies. But before we can say that cash transfers are the best tool we have for eradicating extreme poverty, we need to look at the alternatives.

Cash Transfers: The Alternatives

The painkiller and the cast

Imagine you’ve just broken your wrist. Your family rushes you to the nearest hospital while you’re doing your best not to swear and scream. The doctor greets you with painkillers and a splint to prevent movement. An X-ray evaluates the damage. Luckily for you, the broken bone is in position, so you get a cast, more painkillers, and a hearty goodbye. You can go home.

Now the painkillers aren’t going to heal the fracture. They’re there to address your immediate symptoms and help you manage the ordeal while the treatment (six weeks in a cast) takes its course.

The same is true for poverty. Long-term solutions address the underlying causes of poverty. Short-term solutions provide a lifeline to the poor now. While cash transfers in the developing world are gaining popularity, the spending on the long term is still disproportionate. Imagine your doctor telling you that you should forget painkillers and focus on your long-term well-being.

Cash transfers cannot replace clean water, schools, or roads as the solution to extreme poverty. These building blocks of a developing society are crucial. Instead, cash transfers should be seen as a short-term remedy to poverty. A painkiller that provides housing and food. A painkiller that saves lives, gives back the dignity of being human, and offers opportunities to build a better future. It should not be an either-or question. Cash transfers should be an important part of the development narrative. The part that comes first, because millions are destitute now.

I will thus not compare providing cash transfers to building roads. Instead, I will look at other interventions that can address poverty in the short term. It turns out, there aren’t that many.

Short-term remedies to extreme poverty

Sulaiman et al. (2016) have reviewed 48 poverty reduction interventions, focusing on lump-sum unconditional cash transfers, livelihood development, and graduation programs. Lump-sum unconditional cash transfer programs provide one-off grants. Livelihood development programs help poor people build skills. Graduation programs combine the two approaches, offering the poor various services (e.g., saving, mentoring), while also providing a productive asset or a grant.

Livelihood development programs seem to fare worst. Many do not target the extremely poor, and those that do perform worse than those working with more affluent participants.

Graduation programs perform much better, both by focusing specifically on the very poor, and by leading to marked improvements in participant livelihoods. In a review of six randomised controlled trials, Banerjee et al. (2015) found significant increases in spending on consumption, accumulation of savings and assets, food security, mental health, and even income three years after a productive asset was provided. In another promising study, Bandiera et al. (2016) found positive effects extending up to seven years after the transfer of an asset (e.g., livestock) in rural Bangladesh.

Nevertheless, cash transfers appear to outperform even that. Sulaiman et al. (2016) attempted to estimate the costs and perform a cost-benefit comparison of the three intervention types. While the analysis should be taken with a bucket of salt, they conclude that cash transfers lead to the highest impact on household consumption for every dollar spent, outperforming graduation programs by 45%.

A conceptually different approach to poverty alleviation is microfinance. While it had been lauded as a game-changer, Banerjee et al. (2015) (and numerous other studies) have recently concluded that microcredit exhibits “a consistent pattern of modestly positive, but not transformative, effects”. No reduction in poverty, no improvements in living standards.

In short, the evidence is clear: cash transfers consistently outperform other short-term poverty reduction approaches.

The logical argument

If you think about it, it should not be at all surprising that unconditional cash transfers are the best tool for reducing poverty. We have defined poverty as living on less than int.-$1.90/day. Cash transfers thus attack the problem directly — 100% of every dollar that reaches the recipient can be used to fill the immediate consumption gap. Not so in other interventions. Instead of money for food, a livelihood development program would offer you training. (Try focusing on a lecture when your stomach is growling.) Instead of money for cataract surgery, a graduation program will give you a goat. And instead of money for fertiliser, a conditional cash transfer program will finance regular doctor visits instead.

So here it is: if you want to eliminate short-term poverty in the most cost-effective, straightforward, and fair way, nothing can beat unconditional cash transfers.

The European Commission agrees:

Multi-purpose assistance should be considered alongside other delivery modalities from the outset — we need to always ask the question “Why not cash?”

And so does the former UN Secretary-General Ban Ki-moon:

Where markets and operational contexts permit, cash-based programming should be the preferred and default method of support.

From the short term to the long term

Of course, structural eradication of extreme poverty is still very much a goal. And that goal cannot be achieved by cash transfers alone (Ladhani and Sitter, 2018):

Given the multi‐dimensional and complex nature of poverty, economic measures in themselves are insufficient to overcome poverty.

Which begs the question: how should the funds be allocated between cash transfers and long-term solutions?

On the one hand, this question is very difficult to answer — there is very little evidence on the long-term effects of cash transfers in developing countries, and there is no quantifiable evidence whatsoever on most large-scale infrastructure investments.

Bastagli et al. (2015) conclude that while cash transfers have significant first- and (often) second-order effects, leading to immediate expenditure increases and subsequent behavioural changes, there is little evidence for third-order, or long-term effects. For example, while conditional cash transfers lead to increased school attendance, test scores change little.

On the other hand, longer-running programs do lead to higher household expenditure levels. Longer transfer durations are also associated with improved health behaviours, increased contraceptive use, decreased fertility, and a lower likelihood of early marriage. Evidence for other effects is mixed, including, again, educational outcomes.

Reynolds et al. (2017) define long-term effects rather more broadly. For example, in addition to considering body measurement changes in children (outcomes that are likely to require longer time horizons to manifest), they also look at short-term immunisation rates, infant mortality, and HIV and STI prevalence. While these outcomes do not guarantee improvements in long-term health, they make such improvements significantly more likely.

Figure 15

For now, this will have to do. In a decade or so, we can expect results from a truly long-term study of unconditional cash transfers: in 2017, GiveDirectly began a 12-year experiment in Kenya to investigate the effects of universal basic income. The initial results are expected sometime this year.

Balancing the short term and the long term

On the other hand, the answer to balancing the short term and the long term is surprisingly simple: spend as much on cash transfers as is required to lift everyone from extreme poverty immediately. I know this might sound crazy, but hear me out.

First, a basic standard of living is a human right:

Everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.

Second, there is some evidence that unconditional cash transfers lead to long-term improvements in welfare. A few papers have looked into long-term effects, countless have investigated short-term effects. Theory tells us that some of these short-term effects can be expected to translate into long-term effects. In other words, there is a good chance that giving cash to poor people will not only keep them afloat but also enable them to invest in themselves and start generating more income in the future. And there’s no evidence to the contrary.

Third, while the evidence for long-term outcomes of cash transfers is scarce, most other poverty reduction programs have not been robustly evaluated at all. It doesn’t mean that they’re now off the table — but they might merit a second look. In some cases, cash transfers might turn out to be a better investment.

Fourth, cash transfers are good not just for recipients. While a few studies have pointed to increased inflation in remote and poorly connected areas, in general, a richer member of a community means a richer community. While households will save some money, the rest will be either consumed or invested. In other words, recipients not only improve their own lives but also pass on their income to others — the handyman who installs metal roofs, the farmer who grows chickens, the artisan who makes school uniforms. Egger et al. (2019) have found that unconditional cash transfers in Kenya had a local fiscal multiplier of 2.6. In other words, $1 from a cash transfer generated a $2.6 increase in the local GDP.

Fifth, cash transfers are not that expensive. No, really, let me show you.

Cash Transfers: The Costs

A quick primer on cash transfer program costs

Broadly, total cash transfer program costs can be broken down into administrative (set-up, roll-out, operational, monitoring and evaluation) and transfer costs.

Set-up costs are paid at the beginning of the program and will vary significantly by country (e.g., due to the differences in available infrastructure). Program complexity (e.g., targeting methodology) will also affect set-up costs.

Roll-out costs will be incurred at launch, every time the program is expanded, and whenever retargeting is performed. Again, the chosen targeting methodology will be the key driver of costs. White et al. (2013) estimate that set-up and roll-out costs of the Progresa program in Mexico accounted for 71% of administrative costs in its first year, falling to 15% in the fourth year.

Operational costs are long-term program costs that will become the dominant component of administrative costs once the program matures. While unconditional cash transfers avoid significant compliance monitoring costs, payment delivery in some countries might prove quite expensive.

Monitoring and evaluation is concerned with both ensuring the smooth running of the program, and learning from the program to inform future decisions. Evaluations are generally expensive, while monitoring costs will depend on the quality of existing reporting systems.

Finally, transfer costs refer to the money that is sent to program participants. Having reviewed tens of cash transfer programs, Grosh et al. (2008) conclude that administrative costs of a well-executed cash transfer program should not exceed 12 to 15% of the total costs, and might be significantly lower for more generous programs.

Figure 16

This, of course, only tells us that cash transfer programs can be run very efficiently (for comparison, food-related programs have overheads closer to 25% of the total program costs). But the real question is, how much money does a cash transfer program cost?

The cost of safety nets around the world

Since transfers make up the bulk of cash transfer program costs, the flippant answer is: however much you decide to spend on addressing poverty. Let’s see if we can dissect that response.

In the developed world, social welfare makes up for a significant portion of government spending. In 2018, public social spending accounted for 20.1% of GDP in the OECD member states. At one extreme was Mexico’s, with a mere 7.5%. At the other — France with 31.2%. The US hovered in the middle with 18.7%.

Cash benefits amounted to over half of welfare spending, or 11.1% of GDP in the OECD countries. Around 2.7% of GDP was spent on social safety nets, defined by The World Bank as “noncontributory interventions designed to help individuals and households cope with chronic poverty, destitution, and vulnerability”. In low-income countries, on the other hand, safety nets only amounted to 1.5% of GDP, though spending varied significantly between countries. For example, South Sudan spent 10% of GDP on two donor-financed programs, while Somalia’s safety nets amounted to less than 0.2% of GDP.

Figure 17

Now given that most countries do not invest enough in safety nets to eradicate poverty, knowing current spending is not terribly useful. Instead, I will take Tanzania as an example, and will try to figure out how much eliminating extreme poverty would cost there. (I chose Tanzania due to its similarity to sub-Saharan Africa in terms of GDP per capita and the poverty gap.)

The cost of eradicating extreme poverty in Tanzania

Tanzania, famous for the Serengeti National Park, is home to 56.3 million people, 49% of them extremely poor. The country’s poverty gap at int.-$1.90/day is 15.7%, close to Sub-Saharan Africa’s 16.6%. Its GDP per capita is $1,105. In 2016, Tanzania’s government spent 2.35% of GDP (or 12% of total government expenditure) on social protection.

To eradicate extreme poverty in the country, the government would need to close the poverty gap. With perfect targeting and no administrative costs, this would require almost $7 billion, or 11.2% of GDP. Assuming a 15% administrative cost rate, the funding need grows to $8 billion, or 12.9% of GDP (you can see the analysis here).

Unfortunately, perfect targeting is a unicorn. Coady et al. (2004) estimate that an average targeting mechanism transfers only 25% more to the target group than a random allocation of recipients would. At the same time, differences between programs are huge — and the worst offenders are subsidy programs. One of the most successful cash transfer approaches in their sample, a program in Yemen, allocated 68% of the available funds to the poorest 10%, 75% to the poorest 20%, and 86% to the poorest 40%. The least successful performed just as well as flipping a coin would have.

This means that in practice the only way to reach all poor people is to send cash transfers to everyone. If Tanzania were to do that, its social safety net would amount to 82% of its GDP — or to more than 400% of its annual government spending. Not exactly sustainable.

Instead, let’s say we decided to do our best targeting the extremely poor, only transferring so much as to lift them above the poverty line. If we did no better than a flip of a coin, we could expect extreme poverty in Tanzania to go down from 49% to 25%. At the price of 12.9% of the country’s GDP, 14 million people would now have enough to survive.

If, on the other hand, we did as well as Yemen, distributing 86% of the funding to the poorest 40% (and, say 96% to the poorest 50%), we’d only be left with a 2% poverty rate. And that, it turns out, not only achieves but also beats the first Sustainable Development Goal (SDG 1).

The reality is probably somewhere in between. If our targeting mechanism identified 75% of the extremely poor, for example, Tanzania’s poverty rate would fall to an unprecedented 12%. That’s not zero, nor would it achieve SDG 1 right away, and yet for tens of millions of people, it would be life-changing.

The bottom line

So, can extreme poverty in Tanzania be eradicated through cash transfers?

No, if the Tanzanian government had to shoulder the costs alone.

No, even if the government liquidated some of its highly ineffective programs (e.g., spending on subsidies, a terrible mechanism for poverty alleviation, accounted for 0.42% of GDP — significant, but not nearly enough).

Yes, if instead of taking an all-or-nothing approach, Tanzania started small and expanded. While the goal is poverty eradication, some support is better than no support. National safety nets grow as economies mature. It’s alright to begin with a limited pool of participants. It’s alright to start with smaller transfers. In fact, if the program was launched to 20% of potential participants, and covered 15% of the poverty gap for each, it could be funded by redirecting spending from subsidies. If aimed at the poorest of the poor, such transfers would, on average, increase their consumption by 34%. In many developing countries, cash transfer programs cover even less than that (e.g., 15% in Kyrgyzstan, 18% in Malawi), and yet change lives.

And yes, extreme poverty in Tanzania could be eradicated if the program was partially funded by international donors.

Eradicating Poverty: The Funding

Remember, I estimated that it would cost $194 billion/year to eradicate extreme poverty everywhere in the world. And yet eradicating poverty in Tanzania alone seems pretty damn hard. So how do we approach this global challenge?

Well, I have a few ideas.

Global leadership, with the World Bank at the helm

The mission of the World Bank is “to end extreme poverty and promote shared prosperity in a sustainable way”. Its experience, size, and access to funds mean that it is uniquely positioned to lead the global response to acute poverty.

First, it should include short-term poverty alleviation in its strategy. Today, the World Bank is guided by three pillars: sustainable economic growth, investments in human capital, and resilience to global shocks. According to its former president Jim Yong Kim, this means:

  1. Doubling down on infrastructure investments,
  2. Investing in “skilled, healthy, productive workforce” (presumably by focusing on education and healthcare), and
  3. Preparing for looming threats: population displacement, climate change, and pandemics.

All of these are important long-term solutions. But by focusing solely on investments into the future, the World Bank is making today’s poor pay the full price for tomorrow’s prosperity. This is not kind, fair, or even strategic. To productively use all this infrastructure, people need to be able to survive first. And that survival should be at the core of the World Bank’s strategy.

Second, the World Bank should organise its members into action. With COVID-19 ravaging the world, prominent development economists are now calling for concerted action. Here’s Esther Duflo:

You want to think of it as a new Marshall Plan to help the countries in the south to bounce back from this crisis in the same way that Europe was able to bounce back from the World War Two crisis.

Even Dambisa Moyo, the author the famous book against aid, has pleaded for immediate action:

A decade ago I gained notoriety as a critic of large-scale foreign-aid programmes that flow from Western countries to developing economies. […] Yet today it is clear that Africa urgently needs a substantial aid injection or it will be destroyed by the coronavirus.

Moyo is addressing the US in her appeal. Yet unless November brings a change in American politics, it’s more likely the country will cut whatever aid remains, not offer more support — or leadership. The World Bank, however, is the ideal organisation for the job. It knows how to get countries to work together, and it can achieve what the US couldn’t: include all potential donors in the plan. Yes, that means working with, rather than against China.

Third, the World Bank should lead by example. In 2019, its lending for social protection was $4.3 billion, or 9.5% of its annual budget. Now while the amount is impressive, it is unclear how much went into cash transfers. It’s also unclear why $8.4 billion, or 18.7% of its budget, would go to “public administration” programs. Of course, it’s impossible to say how effective all these investments are, but I am willing to hazard a guess: spending 18.7% of the budget on social protection rather than public administration would have had a much more notable effect on global poverty.

If all this were to happen, the results could be unprecedented. Several times in our shared past we showed that can overcome our differences and solve the world’s most challenging problems. In 1980, we swept smallpox from the face of the Earth. In 1987, we adopted the Montreal Protocol, with 98% of ozone-depleting substance use now phased out. Eradicating extreme poverty could become our next big hit.

National commitments, fulfilled

The World Bank can lead the fight against poverty, but so can individual countries.

First, it’s about time all rich economies fulfilled their Monterrey Consensus commitments. Increasing spending on aid to 0.7% of GNI in high-income economies would expand annual development funding by $214 billion. And that, of course, is more than enough to eradicate extreme poverty today.

Figure 18

Second, some rich economies could well do more than just the bare minimum. Four countries today spend around 1% of their GNI on development assistance: Turkey, Luxembourg, Norway, and Sweden. If all high-income countries increased their development assistance spending to 1% of GNI, official development aid would more than triple, reaching $544 billion. If Turkey, a country whose GDP per capita was under $10,000 in 2018, can do it, couldn’t others?

Third, it might be worth taking another look at how national development aid is spent. I bet some programs that are being financed right now are obviously less effective than cash transfers. For example, the New York Times recently reported that the EU has spent hundreds of millions of euros on infrastructure projects in Eritrea. The problem? Many workers on these projects are forced conscripts (a system the UN says is “tantamount to enslavement”). In other words, a reshuffle of spending, even without increases in said spending, can make a difference.

Doing all this would change the world. It would also be very difficult — COVID-19 is threatening the global economy with a long and painful recession, and everyone’s now focusing on their own problems. When the discussion turns global, the spotlight is often on the devastation in the travel industry. But the people on the ground, those whose livelihoods weren’t secure before the pandemic, will suffer much more than the airlines and the hotels. Now might seem like a really bad time to talk about safety nets in the Global South. In fact, the time is ideal. The extremely poor need assistance now.

Charities obsessed with problems, not solutions

Thousands of charitable organisations out there are doing amazing work. Thousands more aren’t nearly as effective. Just as governments should have a good look at their aid budgets and consider redirecting some of the spending to cash transfers, so could large foundations. For example, Rockefeller Philanthropy Advisors, one of the wealthiest charitable foundations in the world, directs millions of dollars to arts and culture. Is that really “thoughtful, effective philanthropy”? Peter Singer, a moral philosopher and an avid defender of utilitarianism, responds bluntly: “I don’t think so”.

It’s even harder to be objective when the organisation under scrutiny is built around a single solution. Remember Books for Africa? While there is no evidence that they are making a difference (and it could be argued that they are making things worse by competing for donations with truly effective organisations), they can’t exactly turn around and start treating intestinal worms instead. Very few charities have been built to truly address a problem. Most came about as inspiring solutions based on hope and hunch. They are unlikely to change. They believe they are making a difference, even when from the outside it’s clear they aren’t.

But while pivoting to a new approach in an established charity is difficult, there is space for new effective charities. The cash transfer scene seems particularly bare. In fact, GiveDirectly is the only cash transfer nonprofit I know. Its top charity status, evidence-based approach, and extreme transparency mean that in most cases donating to it would be more efficient than starting a new organisation. However, GiveDirectly only works in seven African countries. The transfers it provides are large, but one-off. And while it is growing fast, it cannot take on the challenge of global poverty single-handedly. I thus believe that the world could do with quite a few more cash transfer charities.

To summarise, large foundations can make a difference by reviewing their strategies and redirecting some of their spending to safety nets. And while smaller charities are unlikely to change their approaches, there is space for new organisations that would work together to expand access to cash transfers around the world.

Your money and your voice

It would be easy to conclude that individual actions don’t matter. It would also be wrong. Every one of us can do an immense amount of good.

The most obvious way to make a difference is to give. $1,000 donated to GiveDirectly would fund one cash transfer, covering around one year’s worth of expenditures for the average household. $1,000 is a considerable amount of money. But look at it this way: you could spend it on the iPhone 11 Pro (it’s got three rear cameras), or you could support a family of five for a year. This is how Lily, a 45-year old Kenyan mother of six, spent a third of her transfer:

My first priority after I got the money, I opted to clear all the school fees for the whole year so that my children would be allowed to concentrate with their studies.

And here’s Betty, 40, also from Kenya:

When I received the message that early morning, I checked its content and when I saw that I had received my first transfers I felt so happy. I knew that my dream of owning a cow would come true.

If I had to choose between a new iPhone and a cow, a cow would win every time.

The Economist recently reported that most American households give away around 1.5% of their income. 1.5% is not enough. For example, all Muslims donate 2.5% of their annual wealth above a minimum amount known as nisab. The effective altruism community has encouraged its members to pledge at least 10% of their incomes to charity. Peter Singer’s The Life You Can Save stands somewhere in between, proposing a progressive giving rate.

Giving more doesn’t have to be painful. It is estimated that American households waste 31.9% of their food. That’s $1,866 per year that could go to charity instead. An average American also throws away 37 kg (81 lb) of clothing every year. The next time you want to buy that new pair of jeans? Don’t. Donate to charity instead.

And while you’re scrutinising your spending, review your donation strategy as well. For example, donations to religious causes are unlikely to be effective. In fact, they might not even go where you’d expect them to go. They might finance luxury property purchases in London, plug the Vatican’s budget hole, or disappear in “off the books” Swiss bank accounts, all in the name of helping the suffering. Why not send the $124.52 billion Americans donate to religious causes every year directly to the poor instead?

But money is not the only asset you have. You also have a voice. Global poverty is rarely seen on the national political agenda. And yet without a national discourse, Australia, Canada, France, Japan or the US are unlikely to honour the 0.7% commitment. Without a national discourse, increasing development funding might even be politically risky. It’s up to the voters to make not honouring the commitment politically risky instead.

And now, let me put it all together into one coherent vision.

Eradicating Poverty: The Vision

Extreme poverty could be eradicated tomorrow.*

The beauty of this challenge is its simplicity. We know exactly how to lift every single person on Earth above the international poverty line:

daily cash transfer = int.-$1.90 — current daily consumption

The magic of this challenge is its versatility. The solution could come from anywhere. Whether the World Bank leads a concerted global action, or every one of us donates a bit more, the outcome will be the same: fewer people living in destitution.

To have the best shot at eradicating poverty, however, I say we attack it from several directions simultaneously. Here’s how I’d go about that.

* This, of course, is a bit of an overstatement. The logistics of delivering cash transfers alone would be quite an undertaking. However, such challenges have been solved efficiently in the past. A national cash transfer program cannot be designed and implemented overnight, but it can be in a year. And while that is not exactly “tomorrow”, in terms of the current pace of poverty reduction, it almost is.

The guiding principles

It’s about fundamental human rights. There is no intrinsic value in eradicating poverty. The value comes from lessening human suffering, bringing happiness, giving back the right to dignity.

It’s not all or nothing. Poverty is not a binary variable. Poverty can be absolute or non-existent, but more often it is somewhere in between. You don’t need to eradicate poverty in one go.

Global challenge, national solutions. While global poverty is, well, global, it is made up of national poverty. Don’t try and solve poverty in the world. Solve poverty in every country instead. Poverty is context-specific, solutions should be too.

Not all suffering is created equal. All humans are of equal value. However, some suffer more than others. Use the poverty gap to assign a numeric value to human suffering.

Effectiveness is an obligation. Choose the most effective way to address poverty. If you don’t, you will be failing those who are suffering but could have been assisted had you made a different decision.

The vision

A decentralised global safety net system, available to everyone who needs it, supported by everyone who can.

The system has three pillars:

  1. The recipients are the poor individuals selected to receive cash transfers.
  2. The safety nets are a set of national cash transfer programs that benefit the recipients. A national safety net is funded by the local government and the funding system.
  3. The funding system is a network of transnational organisations, donor countries, charitable foundations, and individuals who provide partial financing for national safety nets.

The recipients

The ultimate goal of the system is to identify everyone who lives on less than int.-$1.90/day and to provide them with cash transfers to make up the difference.

However, neither sufficient funding nor the necessary infrastructure will be available on Day 1 to lift everyone out of poverty. Prioritisation thus matters immensely. Some countries are poorer than others. Some already have targeting, delivery, and monitoring mechanisms in place. The prioritisation exercise should take all this into account to make sure we’re making the most out of every available dollar.

The safety net

The safety net will necessarily be country-specific. However, it should comply with a set of basic requirements, such as cash transfers being unconditional and recurring. With so many people requiring assistance, safety nets also need to be efficient. One way to ensure this is certification. First, a cash transfer program would be evaluated in terms of its impact on recipients and its cost-effectiveness. Second, ongoing monitoring would ensure that the program is implemented and runs as planned, and that the benefits are thus reaching the intended recipients.

Now while many countries have existing cash transfer programs ready to be expanded, in some cases such safety nets will have to be built from scratch. The preferred approach is for the country’s government to introduce and manage such a program. The World Bank could offer technical assistance, and academic partnerships might finance an evaluation of a pilot.

In rare cases, however, a government might not be willing or able to take on a cash transfer program. Take Yemen (population: 28 million people), the leader of the Fragile States Index. With a civil war that began in 2015, poverty in the already poorest country in the Middle East and North Africa skyrocketed to around 75%. While cash transfers would help alleviate poverty, Yemen’s government is currently dealing with bigger problems.

Figure 19

One approach is to wait while the government is ready. Indeed, while the Yemenis are in dire need of humanitarian assistance, introducing a large-scale, efficient cash transfer system in the war-torn country would be nigh impossible. Waiting, I suspect, is the most feasible option in conflict situations.

Another approach is for a nonprofit to (temporarily) fill the space. In a way, this is what GiveDirectly is doing in the Democratic Republic of the Congo and elsewhere in Africa, though at a much smaller scale. In the long run, safety nets should be managed by governments. However, poverty is afflicting millions of people right now, making temporary, non-governmental solutions suboptimal, but necessary.

Finally, one component of the safety net deserves particular attention: delivery. GiveDirectly, for example, only works in countries that have reasonably well developed electronic payment systems. Delivering cash physically would increase both administrative costs and the risk of corruption, while introducing electronic payments where none existed before would be prohibitively expensive. In the short run, readily available payment systems, whether electronic or physical, will have to do. In the long run, however, a mobile banking service could be a worthy investment.

The funding system

The feasibility of global poverty eradication ultimately depends on funding. The goal of the funding system is thus to cover the difference between what a national government can provide itself, and what is needed to eradicate poverty in the country.

This funding gap would be paid for by transnational organisations, wealthy countries, charitable foundations, and individual donations. Donors would pool their assistance in funds of two types. Capital from the global fund could be directed to any recipient country. Money from national funds, on the other hand, would have to be spent on the safety nets of the specified recipient countries. The funds would be managed by the World Bank.

Financing the global fund is preferred, as it would allow for the greatest flexibility, and thus greatest efficiency. For example, if funds were earmarked for Tanzania, but the country’s distribution mechanism was limiting the extent of transfers while Malawi was ready to go, not funding poverty reduction in Malawi would be inefficient. However, donors cannot always provide unconditional aid. In such cases, making grants to specific national funds is acceptable. If the global fund is sufficiently large, efficiency losses will be minimal.

Finally, countries and large organisations would transfer their assistance to these funds directly. However, the funding system should not be concerned with individual fundraising. Instead…

Charity fundraising

Individual donations should be collected by new or existing charities. While basic guidelines (e.g., transparency expectations) could be provided, it would be up to the charities themselves to decide how exactly to raise funds — including whether to target global or national funds. For example, the Ghanaian diaspora could set up a charity to collect donations from Ghanaian expatriates around the world, with the specific purpose of fundraising for the national fund of Ghana.

National donor registers

An idea I’m rather more excited about is the national donor register. But first, let’s talk about organ transplants.

Many high-income countries operate under what’s known as the presumed consent framework for organ donations. In short, every adult citizen is considered to be a willing organ donor unless they explicitly opt out. In practice, presumed content is rather more complicated than that and is unlikely to lead to an increase in organ donations in the short term. However, a mere introduction of presumed consent changes the conversation from organ donation as the stuff of saints to organ donation as the new normal. Here’s how we could use this framework to fight poverty.

The government of a wealthy country would introduce the presumed consent framework for global poverty eradication. Under this framework, every individual taxpayer would be considered to be a willing donor. The government would also introduce a very small income tax. (In some cases, introducing a tax, no matter how small, might be counterproductive, so repurposing an existing tax might work better.) Everyone would automatically start paying the tax, with the proceeds directed to the global fund for poverty eradication.

Unlike most other levies, however, this one would be customisable. Taxpayers could:

  1. Redirect their tax to a particular national fund,
  2. Voluntarily increase the tax, or
  3. Opt out of the tax.

While such a framework would require a considerable investment (and a thoughtful introduction), it has a number of potential advantages over the status quo. First, it would make global poverty an issue of national and personal interest. The introduction of the framework alone would likely lead to more people becoming aware of extreme poverty, and to more people choosing to care about it. It might also change the widely-held belief that local problems are more important or more worthy.

Second, even if most people didn’t care either way due to the tax rate being very low, they would be contributing to poverty eradication nevertheless. For those taxpayers, the tax amount might seem inconsequential; for the extremely poor, it would be life-changing.

Third, it would introduce a very convenient way to donate. While many charities support recurring payments, no other system can automatically take into account changes in one’s income. With this approach, however, the taxpayer wouldn’t need to worry about donating too much when the times are lean, nor would they have to remember to up the amount when their earnings increase.

But most importantly, it would take us a long way towards eradicating extreme poverty forever. For example, a 1% global poverty tax, if levied on taxable income in the US in 2017, would have brought in almost $16 billion. Not enough to eradicate extreme poverty everywhere, but a significant amount nevertheless. With a few more rich countries in the mix, individual donations could quickly become a major source of financing the fight against poverty.

The global safety net

So here it is, a way to eradicate extreme poverty that involves all stakeholders, yet does not depend on any one of them to function. One donor and one recipient are enough to get this started. With hundreds of donors and dozens of recipients, extreme poverty might well become a relic of the past.

Let’s Choose Again

The world is wonderful. We live longer. We feel better. We know more. We experience more. We dream, then we do.

The world is terrible. We invented algorithmic trading, but 20 million of Madagascans live in extreme poverty. We cloned a sheep, but 39 million Tanzanians fall asleep hungry every night. We put a man on the Moon, but 40 million Ugandans have no access to safe drinking water. We discovered the Higgs boson, but 16 million Nigeriens cannot read and write.

The world is even more terrible. We could eradicate extreme poverty. Yet we have chosen not to.

Let’s choose again. And this time, let’s choose to wipe extreme poverty from the face of the Earth:

daily cash transfer = int.-$1.90 — current daily consumption

¹ Here and elsewhere, int.-$ refers to constant 2011 international dollars. Monetary values in int.-$ are thus adjusted for inflation and price differences between countries.

² With 58,220 recorded deaths and 2,594,000 Americans serving in South Vietnam between 1955 and 1975, the US Armed Forces fatal casualty rate in was 2.2%.

³ Using the global poverty gap is a more accurate way to calculate the amount needed every year to eradicate poverty. With a poverty gap of 3.1%, a global population of 7.593 billion, and a 13.6% increase in the US consumer price index since 2011, the annual amount needed to eradicate poverty is 3.1% * 7.593 billion * (100% + 13.6%) * int.-$1.90/day * 365 days = $185.5 billion.

⁴ The calculations are necessarily very rough. However, GiveWell estimates that it costs AMF around $1,690 to achieve an outcome that is comparable to that of averting the death of a child under 5.

⁵ That said, while the stigma might have been an intended side effect back when the program was first introduced, the government is now working hard to overcome it.



Ernesta Orlovaitė

Bookworm (but I sometimes go on real adventures) · Obsessive thinker · Inconsistent writer · “You live and learn. At any rate, you live.” — Douglas Adams