A Primer on Poverty and Economic Growth

Both China and India have seen their economies grow at breakneck speeds over the last decades. At the same time, the number of poor people residing in these two countries has been reduced substantially, although somewhat less sourcespectacularly in India compared to China.

Other indicators of wellbeing point in the same direction. Life expectancy in India is now 65 years as opposed to 42 in 1960; in China the numbers are 73 and 42 respectively. The number of Indian children dying before they reach the age of 5 is now about 60 for every 1000, compared to almost 200 in 1970. In China: about 20 for every 1000 now versus more than 100 in 1970.

It’s not outlandish therefore to assume that there’s a link between good economic growth and large reductions in poverty rates. Given the fact that some other potential causes of poverty reduction – such as foreign aid and good governance – have not been prominent in those two countries (to put it mildly), the growth hypothesis is even more persuasive.

Here are number of correlations for a larger set of countries. Here‘s the example of Africa, where growth and poverty are almost absolute mirror images.

However, does this mean that economic growth is a sufficient condition of poverty reduction? I don’t think so. A growing level of GDP per capita in an economy means that the average person is better off, financially as well as on some other dimensions given the strong correlations between GDP and other indicators of wellbeing. But improvements for the average person can go hand in hand with stagnating or even worsening poverty rates. A lot depends how the proceeds of high growth are distributed among the citizens of a country; distributed either

  • intentionally by government policy: through taxation and welfare policies that can be more generous in a richer economy
  • or automatically by some form of trickle down effect: more aggregate national income or production means more jobs, better paid jobs etc., at least in theory.

Trickle down economics has been somewhat discredited of late, so perhaps the causal effect of growth on poverty reduction must pass through government redistribution, at least in part. The anti-poverty efforts of India’s successive governments are well-known. Less well-known is the fact that in China as well governments have implemented a strong although probably insufficient social security system. (Insufficient because inequality has risen dramatically in China, and somewhat in India. Greater inequality dampens the poverty reducing effect of growth).

The crucial role of government led redistribution has been confirmed by this paper in which Lane Kenworthy compared growth and income data for 17 developed countries. Specifically, he looked at the ways in which the incomes of people in low to middle income groups benefit from economic growth. The nature of government transfer systems is the reason why the effect of growth on the incomes of the poor is not the same in all countries.

[W]hen households on low incomes got better off, it was due most often to a rise in net government transfers. Where net transfers increased, incomes tended to increase in concert with economic growth. Norway, the UK, Sweden, Finland, and Denmark illustrate this pattern. Where net transfers were stagnant, income trends were decoupled from growth of the economy. We observe this in the United States, Canada, and Switzerland. This is an important finding. It means that, as a general rule, growth has not trickled down to low income households through wages or employment. And it means that, when government transfers haven’t grown, wages and employment haven’t stepped in to take their place. (source)

So we can conclude that economic growth, although important, is probably not a sufficient condition for poverty reduction, at least not in all cases.

A final remark: it’s interesting to note that poverty reduction is one of the drivers of growth. So the causation goes both ways, as is often the case in correlations.

Policies that are effective in increasing the incomes of the poor–such as investments in primary education, rural infrastructure, health, and nutrition–are also policies that enhance the productive capacity of the economy in aggregate. (source)

Some other interesting studies on the topic are cited here.

The Causes of Poverty (79): Poverty Traps

Many among us will experience short spans of poverty at some stages in our lives. I lose my job or my unemployment benefits, I have a catastrophic but transitory health problem, an extreme weather event destroys my crop, or an economic crisis forces me to declare bankruptcy. As a result, I have to live off my savings or my parents and friends will have to lend me money. Still, in time I find another job; my health improves as I benefit from cheap healthcare (perhaps provided or subsidized by the government); the weather returns to normal and I can resume my profitable farming activity; or I can start a new business under the protection of bankruptcy laws that don’t burden me with debt.

However, I may also be what’s called a “structurally” poor person, meaning that I’m poor for most if not the whole of my life. Perhaps I was even born into poverty. The reason may be that I find myself in a “poverty trap”, a self-reinforcing mechanism which causes poverty to persist. In other words, I’m poor because I’m poor. And because I’m poor I’ll always be poor. I’ll die without ever having had an “adequate” standard of living, all the while passing on my poverty to my descendants.

Here are some examples of poverty traps:

  • I have a job, but the wage is low. As with many low wage jobs, I have almost no control over my work schedule. That means I can’t take on a second job and I can’t send my kids to child care. I have to spend time, money, effort and other people’s good will to take care of my kids. My job is physically hard and so I tend to have some health problems. My life is relatively expensive and it’s hard to find a better job. My salary doesn’t really cover my spending needs, hence I’m poor.
  • I can’t afford to pay the security deposit for a rental apartment, so I’m stuck in an expensive motel or I have to live with my parents who can barely afford their own survival. I also don’t have a refrigerator or a microwave, so I have to buy more expensive food. I have to wash my clothes by hand because… you guessed it. This takes a lot of time, time that I can’t spend on wage labor.
  • I don’t have tap water or heating because those aren’t things that people have where I’m from. I use wood for fuel like everyone else. The result is deforestation, soil degradation, lower crop yields and yet more poverty. My children have to help me – which is why I have a lot of them – to the detriment of their education. My kids will probably inherit my poverty because of this.
  • A lot of the things I’m forced to do because I’m poor are illegal. The lights of my car broke down, and I got a fine. I should have made the financial sacrifice and get them replaced, but I gambled on not being caught. I couldn’t pay the fine and my car was repossessed. Now I have to take public transport but can’t pay for that either. So I often get a fine for that as well. I know some homeless people who get a fine just for being homeless.
  • My calorie intake is too low to give me the strength to work. The quality of work I’m able to offer is inadequate for obtaining the food I require, and the food I do get isn’t enough to allow me to deliver quality work. My productivity is low, my earnings are low, and ultimately I can’t even keep a job or work the farm. My low calorie intake levels lead to health problems. My inadequate housing makes those problems even worse. My ill health, caused by my poverty, makes my poverty worse. I’m more likely to catch a disease, and also less likely to recover from it.
  • Like many poor people I have a low credit rating, making it difficult to get credit. The credit I do get is very expensive, which sort of defeats the point. Now, I do need the credit because I don’t have any savings. People say that I exhibit a high discount rate, that I’m too present-oriented and that I’m unable to delay gratification. Instead of borrowing money at high interest rates as a means to satisfy my unrealistic consumption desires, I should moderate myself and save for the future. But I’m present-oriented because I live in an environment in which I can’t trust people. Better to consume what I have than to save it and lose it later.
  • People also say that my issues with gratification extend to my sex life. I was indeed a teenage mother, and my education suffered as a result. This in turn affected my job prospects and my income. But this wasn’t just stupidity on my part. Being a mother gave meaning to my life. Other meaningful options just didn’t seem realistic.

So, there you have it. I think a lot of these stories are very real, and the problems that poor people face are often self-reinforcing. Of course, I don’t want to deny human agency. There are people who, even in the face of the worst possible circumstances, can fight their way out of poverty traps. So “trap” may be too strong a word. Individual responsibility still plays a role. Yet, let’s not forget that a poverty trap is sometimes intergenerational, as I’ve said before. Some children are born into a trap, and you can’t insist on responsibility and agency when we’re talking about children. A child growing up in a poor family may suffer in its early development. Undernourishment for instance can have a lasting impact on learning ability and earnings as an adult. Children of the poor are perhaps even more affected than the parents because the latter need a minimum calorie intake to work. They have to eat first. If they choose not to eat first, they will only make the poverty of the household worse.

Just to be clear: I’m not talking about an entire economy or country being stuck in a poverty trap. If you were expecting a post about that, I’m sorry I’ve wasted your time. I’m not wading into the treacherous debate about the necessity of large foreign aid injections to break the cycle of poor nations that can’t save enough to finance investment necessary to growth.

This post seems to be going on forever, so I’ll limit myself to a description of the problem. The solution – how to get out of poverty traps – is a topic for another day.

More posts in this series are here.

The Causes of Poverty (78): High Discount Rates and Lack of Delayed Gratification

You talk to conservatives about the reasons why poor people are poor, and chances are that the discussion turns to lack of self-control, high discount rates and inability to delay gratification. “High discount rates” means that things in the present or near future are viewed as having a higher payoff than things in the distant future. If you have a high discount rate, you focus on immediate gratification. This in turn shows up in low savings rates, high debt, obesity, teen pregnancy, drug use, high drop out rates, low school attendance and other vices supposedly common among the poor.

Some even argue that differences between people in the apparent levels of self-control, discount rates or time preferences – which is all the same thing – appear at a very early age and are therefore probably innate. The famous marshmallow test will then get a favorable citation: you give kids a marshmallow and tell them they can either eat it now or, if they wait a few minutes, have two marshmallows. Kids who wait do better later in life.

However, recent studies have suggested that the marshmallow test does not, in fact, reveal innate (in)ability. The environment in which tests such as these take place determines to a large extent the levels of self-control revealed through them. Whether or not people are capable of delayed gratification depends not on their abilities but on their assessment of the reliability of the world around them. When the world is not worthy of trust, the best course of action is often to live for today.

This attitude towards the world and the future is probably internalized from a young age onward, which makes it hard to change. What it takes is to offer young children a reliable environment allowing them to develop levels of trust which will in turn yield low discount rates and the ability to delay gratification later in life. But in order to do that we’ll need to reduce parental poverty. Claims about lack of self-control as a cause of poverty then have things completely backward. Rather than a cause we’re dealing with an effect of poverty.

More posts in this series are here.

The Causes of Poverty (77): The Lottery of Birth and the Country You Live In

Charles Kenny explains to what extent the country you live in affects your livelihood:

[P]overty in Africa and Asia isn’t the result of something about individual Kenyans and Pakistanis, it is instead something about Kenya and Pakistan. Individuals the world over have the same drives and capacities, but the societies and places in which they live present radically different opportunities to turn that drive into wealth, health, and well-being.

That’s clear from evidence compiled by Princeton economist Orley Ashenfelter for the National Bureau of Economic Research. He looks at the wages earned by staff working at McDonald’s franchises around the world and compares what they earn to the cost of a Big Mac in that same franchise. The Big Mac is a standard product, and the way it is made worldwide is highly standardized. The skill level involved in making it (such as it is) is the same everywhere. And yet McDonald’s employees worldwide earn dramatically different amounts in terms of Big Macs per hour.

In the United States, a McDonald’s employee earns an average of $7.22 an hour, and a Big Mac costs an average of $3.04. So the employee earns 2.4 Big Macs per hour. In India, an employee earns $.46 an hour. The average Indian Big Mac (made of chicken, which is cheaper than beef) costs only $1.29. Still, the employee earns only one-third of a Big Mac for each hour worked. Same job, same skills—and yet Indian workers at McDonald’s earn one-seventh the real hourly wage of a US employee. There’s a huge “place premium” to working in the United States rather than India.

The place premium affects more than just low-end service jobs. Economist Michael Clemens, a colleague of mine at the Center for Global Development, studied a group of Indians working in an India-based international software firm who applied for a temporary work visa to the United States to do the same work in the same firm, just on the other side of the Pacific Ocean. Some of them then won the lottery by which visas were issued, while others lost. The winning workers, who were still in the same firm and still doing the same type of job on the same projects, suddenly saw dramatic differences in their pay.

The ones who moved to the United States started earning double what their colleagues back in India were earning (adjusted for purchasing power). They were earning more not because they were different from the colleagues they left behind—selection was not based on education, talent, or drive but was entirely random. And once they returned to India, they went back to earning pretty much the same as their colleagues who had never left. They briefly earned more in the United States simply because they were in the United States rather than India. (source)

Some more numbers are here, regarding how much more workers in the U.S. make compared to identical workers in developing countries; e.g. Nigerians and Yemenis stand to gain upwards of 10 times as much from moving to the U.S.

The place premium is a strong argument in favor of reducing migration restrictions: it doesn’t seem just that people’s income is determined by the good or bad luck of having been born somewhere, and the use of force to keep people in their country of birth only aggravates the injustice. However, by the same logic we can also argue for a more generous welfare state: it’s not just your country of birth that affects your income. Your parents, social class, genetic endowment, health prospects, looks etc. are also a lottery that affects your income and good fortune.

More posts in this series are here.

The Causes of Poverty (76): Farmer vs. Hunter Thinking

Tim Harford mentions an interesting study about the origins of different ideas about justice. Farmer cultures seem to stress desert, whereas hunter cultures believe that solidarity is the more important focus of justice. Hunters tend to share because their “incomes” are volatile: some days they catch too much, other days not enough. Luck also determines farmer incomes, but to a lesser extent. Bad weather means bad luck, but it’s also bad luck for neighboring farms. A sharing culture won’t solve that kind of bad luck in the same way as it will in the case of bad luck while hunting. Another reason why a sharing culture will be less important in farmer cultures is the fact that farm crops can be stored more easily than meat in primitive societies.

A farmer mentality will therefore stress self-sufficiency over sharing, and perhaps this will fuel desert-based theories of justice even centuries after farming or hunting has ceased to be an important social role. That may have an impact on the way a society deals with poverty. If you adopt a desert-based theory of justice then you’re normally less inclined to enact policies that reduce poverty since you believe that poverty is deserved. If people deserve their poverty then they can’t claim assistance, and if assistance were to be given anyway that would be an injustice to those whose stock of means is used as a source of assistance, because they too deserve what they have.

It’s tempting to use this farmer-hunter difference to describe the different approaches to poverty in Europe and the US. There’s more opposition to the welfare state in the US, and desert-based theories of justice are more popular there. Hard work and self-sufficiency are common topics of political talk in the US, whereas words such as solidarity and equality are more often used in Europe.

And of course the US was founded as an agrarian society (Thomas Jefferson for instance was a staunch agrarian), with the South of the country remaining agrarian deep into the 19th century.

However, careful with national stereotypes. It’s not as if the whole of the US is hardhearted. It’s a matter of degree.

Steven Pinker has come up with a similar story, although he contrasts farmer and herder cultures.

More posts in this series are here.

The Causes of Poverty (75): Different Types of Colonization

Some time ago, Acemoglu, Johnson and Robinson argued that areas of the globe where early colonists did not face a high mortality risk – such as North America and Australia – are now much richer countries. Many ex-colonies in Africa and the Caribbean, on the other hand, where colonists did face high mortality rates because of tropical diseases such as malaria, are now poorer.

Why is that? AJR claim that the reason is institutional. In those poor countries, the only institutions that were created in early colonial times were extractive. If only a small elite of colonists could survive the local diseases, colonizing nations had little incentive to create durable and non-extractive institutions or provide public services like health and education to the masses of the local populations. More livable colonies could be occupied en masse by the natives of the colonizing nations. These natives required institutions and had the knowhow to create them. The societies that developed there were therefore better organized and far more equal (if you leave out the indigenous populations who were often exterminated). The early institutional built-up, the argument goes, has survived until today, and it’s commonly accepted that good institutions play a key role in development.

Make of it what you will. Perhaps it obscures more than it reveals. In the wrong hands, this argument can be used to exonerate present-day autocratic rulers. After all, it takes time to build institutions, especially in countries burdened by a long tradition of (the wrong kind of) colonialism. Path dependence can be a lousy excuse.

More posts in this series are here.

The Causes of Poverty (74): Family Structure, Ctd.

The more traditionally minded among us often blame family structure for high poverty rates. Family structure is of course a euphemism hiding several very specific moral judgments about people’s behavior, about single motherhood, divorce, paternal negligence and incarceration. Those are the things that supposedly make people poor. “Family structure” just sounds nicer and more neutral.

At first sight, this does make some sense. It is a lot harder, financially and otherwise, to raise kids on your own, and if you find yourself in this situation it’s often your own fault or the other parent’s fault. Having a kid or not is a choice given the availability of contraception and abortion. Divorce is a choice. Finding yourself in prison as a parent is a choice. And even if you’re not a parent, marriage or cohabitation is just plainly cheaper than living on your own because you can share costs. You’ll have to buy just as much food as a single person, but the cost of rent, heating, internet access, the use of a car etc. can be split. A lot of consumption goods are non-rival, and marriage and cohabitation are wonderful places for non-rival goods (the park as well, but you can be lonely there).

Given the high rate of children living with single parents it’s not a priori crazy to assume that there’s a link with poverty rates. It does seem to be the case that poverty rates among single parents are higher than average.

However, we have to be careful when assuming causation. While it can be the case that your income is lower than it would have been had you (remained) married or chosen not to be a single parent, it might just as well be true that your preexisting poverty causes you to be single.

Suppose you are a single person making $9,000 a year and therefore live in poverty. Now suppose you meet someone else making $9,000 and you are considering marrying them. If you marry, the family income goes to $18,000 and is therefore above the poverty line. On a very superficial take, this seems like it would be a real improvement. But that is only if you assume your potential spouse will necessarily remain employed. If they lose their job, you will go from supporting one person with $9,000/yr to supporting two people with $9,000/yr. On the low-end of the labor market, precarity is very common and so this is a very real risk. (source)

There’s also some literature about how teenage pregnancy results from poverty: poor teenagers often see parenthood as one of the few meaningful options that are available (work, education etc. may not be realistic options).

Another point: traditionalists who make the argument that we should promote marriage in order to reduce poverty can perhaps be somewhat dishonest about their motivations. It may be that what they really want is more marriage for its own sake and just dress it up as an anti-poverty measure because arguing outright for more marriage for its own sake is just not that convincing anymore. It’s telling that cohabitation doesn’t figure as an equivalent alternative in their arguments, even though in theory marriage and cohabitation have the same effect on poverty.

And there may be another hidden motivation. Traditionalist proponents of marriage are often situated at the right of the political spectrum, and being right-wing often also implies being opposed to the welfare state. Arguing that poverty should be solved by way of increased marriage rates is perhaps just a roundabout way of downsizing the welfare state: why should we have a welfare state if marriage can solve poverty? Some make this argument explicitly, saying that welfare destroys marriage because it allows people to survive without getting married (I can’t find a citation just now).

What I dislike about the focus on family structure is not really these possible motivations, but rather the inherent simplifications and victim blaming. There are a lot of causes of poverty, and behavior is probably not the most important one. Many single parents are doing a fine job, both financially and otherwise. Low marriage rates are common in many countries, including those where poverty rates are low. And those single parents who struggle probably do so for other reasons than family structure. It’s also true that many working married parents are poor, whereas most celebrity divorcees don’t have a trouble in the world.

A final remark: even if higher marriage rates would be an effective anti-poverty measure, how on earth do we get more people to marry? Tax cuts? A government sponsored dating service? Flower shop vouchers? It all seems so impractical, especially given the ease of other anti-poverty measures (for example…). And not just impractical but also paternalistic and lacking in respect for people’s choices.