Henry Phipps

Development, Population Growth and the Mortality-Fertility Link

Many people express the concern that population growth is a barrier to development

 . Different mechanisms have been postulated as to how population growth could impede development. For instance, when parents have large families they may be less able to invest in their children, whether this be by providing them adequate nutrition, healthcare or schooling. [1] [2] [3]The effects of this lack of investment are thought to undermine development.

Sometimes these concerns are raised regarding charities that save lives through medical interventions, with some claiming that money would be better spent limiting population growth. What’s more, if limiting population growth is an objective for development, then saving lives might at first glance seem to run counter to this goal.

One popular response is to argue that saving lives does nothing to increase population growth and in fact actually decreases population growth

 . The Gates Foundation and Gap Minder have both claimed that reducing mortality causes the fertility of the population (the number of children per woman) to decrease. The long-term effect of this is will be to reduce population growth. [4][5]

There are biological reasons for supposing high levels of mortality will not constitute a barrier to population growth. The death of an infant will lead to a cessation of breastfeeding meaning that the mother is able to fall pregnant again sooner than if the child had survived. [6]Further, if a large family size is desired for cultural or economic reasons, parents may take into account high levels of mortality by having more children in order to ensure that they reach their desired family size. [7]

Reducing mortality may actually decrease population growth. We might think that parents choose to have large numbers of children because they rely on at least some of their children surviving to provide for them in old age and carry on the family lineage. If mortality is high then parents will choose to have more children to try to increase the odds that at least some of them reach adulthood. However, if mortality is lowered, parents will have more confidence that their children will reach adulthood. Therefore they will have less of an incentive to conceive more children. [8]

If the above arguments are sound, then we have a strong response to concerns about population growth. By saving lives we also reduce population growth. But is this always the case?

In this report I look at the link between mortality reduction and population growth. I find that 

we should not be confident that reducing mortality will always reduce population growth

Methods other than reducing mortality seem more effective in reducing population growth.

A Strong Correlation Between Mortality and Decreased Fertility

Since 1950 we have seen extremely strong correlations between reductions in mortality and reductions in fertility. [9] [10]A recent review states that:

“A ‘folk regression’ of the fertility trend on the mortality trend would conclude that saving one life prevents two births.” [11]

no country has ever experienced a fall in fertility without first experiencing a fall in mortality

 . Contrary to popular belief, it does not look like we are heading towards an ever increasing population. Across most of the world, we have seen dramatic reductions in fertility accompanying drops in mortality. In Europe and North America the number of children per family has fallen to between 2 and 3 children, which keeps the population at its current size. In Latin America and Asia, we have also seen rapid convergence to lower fertility rates, with many countries that were once characterised by large families now near the fertility rates of the richest countries. [12]

If family size remained at around 2 children we would see dramatic population growth over the next few decades due to decreased mortality, but ultimately by 2100 the population size would stabilise. [13]

But correlation does not mean causation. 

The drop in mortality since 1950 took place alongside many other developments such as economic growth, increasing levels of education, female empowerment, and provision and acceptance of family planning services.

 Although we might expect all of these trends to lead to decreased fertility, identifying the relative effect of each of these factors is extremely difficult. [14]

For one, it is plausible that there are complex causal relationships between each of these factors. Maybe a combination of increased education and decreasing mortality is needed to see any significant impact on fertility. Also, the desire to limit family size might struggle to translate into drastically lower fertility without widely available modern methods of contraception.

That reductions in mortality are not always the best way to reduce population growth can be seen by comparing the fertility rates of different countries. Both Bangladesh and Niger have traditionally had large family sizes. Bangladesh has now reached a fertility rate of 2.4 children per women whereas Niger has a fertility rate of over 7 children per woman. It does not look like a high level of mortality is the main issue preventing Niger from reaching lower levels of fertility. In Bangladesh child mortality fell at a steady rate from 198.4 out of 1000 in 1980, to 143.7 out of 1000 in 1990. During this time total fertility dropped from 6.4 children per woman to 4.6 children per woman. Contrast this with recent developments in Niger: over a ten year period between 1998 and 2008, Niger experienced an even more dramatic fall in child mortality than was seen in Bangladesh, from 245.2 per 1000 to 141.4 per 1000, yet total fertility only fell from 7.8 to 7.6. [15]

To cite another example, Ethiopia and Uganda have similar levels of child mortality. In 2010, under 5 mortality was 75.6 per thousand in Ethiopia and 78.4 per thousand in Uganda. Both countries had seen large drops in child mortality since 1990 when under 5 mortality stood at 205 per thousand in Ethiopia and 178.7 in Uganda. Yet total fertility in Ethiopia presently stands at 4.8 children per woman as opposed to 6.2 children per woman in Uganda. It doesn’t look like it is high levels of mortality that are preventing Uganda from reaching Ethiopian levels of fertility. [16]

A recent review of the literature relating mortality to decreased fertility and population growth by David Roodman came to the following conclusion:

I think the best interpretation of the available evidence is that the impact of life-saving interventions on fertility and population growth varies by context, above all with total fertility, and is rarely greater than 1:1 [rarely enough for the fertility decline to exceed the mortality decline in terms of effect on population size]. In places where lifetime births/woman has converging to 2 or lower, family size is largely a conscious choice, made with an ideal family size in mind, and achieved in part by access to modern contraception. In those contexts, saving one child’s life should lead parents to avert a birth they would otherwise have. The impact of mortality drops on fertility will be nearly 1:1, so population growth will hardly change.

we should be wary of supposing that further decreases in mortality in high fertility countries will lead to decreased population growth

 . Indeed, as the above extract claims, in some contexts decreasing mortality may be increasing population growth as the impact of declining fertility on population growth may be less than the contribution of more people living long enough to start families of their own. The above review claims that in certain countries, such as Niger and Mali, for every six children that do not die in infancy, only 3 or 2 fewer children will be born. Here, saving lives can be expected to increase population in the short-term. [18]

Nonetheless, it is important to remember that the causal pathways relating mortality to population growth are complex and difficult to assess. In some contexts high levels of mortality might well have been the chief barrier to decreasing fertility. Additionally, although the effect of mortality decreases on fertility might be weak in isolation, it might be that in some contexts decreasing mortality alongside other interventions may be necessary to reduce fertility and limit population growth. This belief is evidenced in another review of the literature:

“If a consensus emerges at all from the wealth of past studies on mortality and fertility declines, it would be that while a direct, independent effect of mortality change on fertility change is often found to be weak, ‘child survival and family planning programs play important complementary roles’.” [19]



It seems fair to say that 

we cannot assume that decreasing mortality is an effective means of reducing population growth. The impact of reduced mortality on population growth will depend on the specific context in which the reduction takes place

Those who are concerned with the effects of population growth on development should predominantly be concerned about Sub-Saharan Africa. Whereas other regions of the world have converged on or are about to converge on average fertility rate of just over two children per family, fertility declines in Sub-Saharan Africa have lagged behind the rest of the world. [20]This lag in declining fertility rates is projected to lead to high levels of population growth in Sub-Saharan Africa. [21]

A particular extreme example is that of Niger, where the population has tripled since the 1970s and is expected to triple again before 2030. [22] Fertility declines have begun later and been slower in Sub-Saharan Africa than in other regions. [23]In part this might be due to the fact that declines in mortality took place later than in other regions. But it appears that there are other factors which are preventing a more rapid decline in fertility.

Some of these are regional-specific factors that lead to a desire for large families, including religious beliefs, marriage practices and communal land ownership. [24]However, many of these factors - such as low levels of female education and a lack of governmental enthusiasm for family planning - have been faced in other areas. [25]

The success of some African countries in dramatically lowering fertility has been attributed to the provision of family planning programmes. This is the case in Ethiopia [26], Kenya [27(which reached a fertility rate of 4.6 in 2010), and Rwanda [29], which has seen a particularly dramatic decline in fertility from 6.1 in 2005 to 4.6 in 2010.

We have good reason to suppose that decreasing child mortality in Sub-Saharan Africa will lead to decreases in fertility. But we also have good reason to think that in many instances falls in mortality may in fact lead to population growth in the short term, as the decrease in fertility may be outweighed by the increasing number of people who survive to have families of their own.

Comparing countries in Sub-Saharan Africa to one another as well as to countries from other regions indicates that factors other than the level of mortality are the major impediment to decreasing fertility and population growth. We have good reason to believe the conclusion of a recent UN report that suggests that 

the most effective means to achieve decreases in fertility and population growth in Sub-Saharan Africa would be to invest in family planning services and female education.

Mortality, Population Growth and Aid: Some Conclusions

Decreasing mortality may have an important role to play in moving to a low fertility society. Nevertheless, we cannot assume that reducing mortality in a particular instance will lead to decreases in fertility that are significant enough to outweigh the direct contribution to population growth caused by reduced mortality.

the major barrier to decreasing fertility and population growth in Sub-Saharan Africa today is not high mortality but factors such as the low levels of female education and lack of family planning programmes

 . If reducing population growth is of overriding concern, it seems that the most effective means to do this would be to attempt to implement the above programmes rather than trying to further reduce mortality.

reducing population growth is but one concern amongst many when deciding which interventions to support in low income countries

 . Just how damaging population growth is for development is unclear and there is dispute as to whether it is particularly damaging at all. Even if population growth does significantly impede development, 

it is still plausible that donating to charities that directly save lives and reduce morbidity confers more valuable benefit than donating to charities that try to reduce population growth.

Finally and most importantly, the historical record shows us that 

we are able to massively reduce fertility and thus (in the long-term) population growth whilst also dramatically reducing mortality

 . Additionally, over the long-term it is plausible that these processes are mutually supporting. Concerns about population growth at worst may give reason to move marginal dollars from health interventions to interventions supporting other causes such as family planning, but they certainly do not give any reason to abandon the process of attempting to reduce mortality and morbidity in low income countries.

[1]Health and Economic Growth, Wile pp. 656-657 

http://www.sciencedirect.com/science/article/pii/B9780444535405000033#





[2]When Does Improving Health Raise GDP?, Ashref et al p. 8





http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3860117/





[3]Copenhagen Consensus 2012 Challenge Paper on Population Growth, Kohler p.50



[4]Gates Foundation Annual Letter 2014





http://www.gatesfoundation.org/Who-We-Are/Resources-and-Media/Annual-Letters-List/Annual-Letter-2014





[5]Will saving poor children lead to overpopulation?





https://www.youtube.com/watch?v=BkSO9pOVpRM





[6]Mortality Decline and the Demographic Response: Toward a New Agenda, Montgomery pp 5-6.





http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.175.9555&rep=rep1&type=pdf





[7]Ibid



[8]Will saving poor children lead to overpopulation?





http://blog.givewell.org/2008/08/03/infant-mortality-and-overpopulation/





[10]Macroeconomics and Health: Investing in Health for Economic Development, Sachs pp. 35-38.



[11] 

http://davidroodman.com/blog/2014/04/16/the-mortality-fertility-link/

https://www.youtube.com/watch?v=FACK2knC08E





[18]Ibid



[19]Demographic pressure, economic development, and social engineering: An assessment of fertility declines in the second half of the twentieth century. Heuveline p.389.



[20]How Exceptional is the Pattern of Fertility Decline in Sub-Saharan Africa? Bongaarts pp. 1-4.



[21]Don’t Panic









[23]How Exceptional is the Pattern of Fertility Decline in Sub-Saharan Africa? Bongaarts pp. 1-4.



[24]Ibid pp. 1-2.



[25]Ibid.



[26]Ibid p.11



[27] 





[28]The Effects of Improved Survival on Fertility: A Reassessment, Cleland pp. 84-85





http://www.jstor.org/stable/3115250?loginSuccess=true&seq=1#page_scan_tab_contents





[29]How Exceptional is the Pattern of Fertility Decline in Sub-Saharan Africa? Bongaarts p.11

How Long Will it Take Economic Growth to Eliminate Extreme Poverty?

Many charitable interventions in low-income countries have a significant positive impact, but what such interventions can achieve is frequently constrained by under-development. It is often claimed that economic growth is the only means to completely eradicate poverty. Sometimes this claim is meant to criticise the efficacy of giving to charity. In this report I accept that continued economic growth may be the only means to completely eradicate poverty, but I will argue that we should be sceptical that continued economic growth on our current model will lead to the elimination of extreme poverty over acceptable time-horizons. If we wish to eliminate extreme poverty in less than a century, the poorest people in the world will need to benefit from a much greater share of the growth in global income. One way in which we could achieve this shift in the distribution of income is for those of us living in wealthy countries to dedicate a significant portion of our income to charities that benefit those living in low income countries.

The World Bank defines extreme poverty as living on less than $1.25 a day.[1] A recent study by David Woodward published in the World Economics Review claims that if we assuming pre-crisis levels of growth (1993-2008) continue indefinitely it will take 100 years to bring the world’s population out of extreme poverty and 200 years to bring everyone to the modest income of $5 a day at 2005 purchasing power parity.[2] Global GDP would have to reach 15 times its 2010 levels by 2115 and 173 times its 2010 levels by 2222.

Closing the poverty gap by increasing the income of the 63.2% of the world’s population living beneath $5-a-day to above this line would require an increase of $4.5 trillion to their incomes. Achieving this through our current growth pattern would involve increasing global GDP by more than $11,500 trillion and reaching a GDP per capita of $1.3 million. Attempting to eliminate poverty by this means would thus entail increasing production and consumption by more than 2,000 times the income we are seeking to generate, and still taking two centuries to do this. A process Woodward describes as “at best grotesquely inefficient”.

Woodward’s claim isn’t that global income is too low to alleviate poverty but rather that it is too unequally distributed. For example the poorest 30% the world population received just 1.2% of the additional income generated by global GDP growth between 1999 and 2008, the poorest 60% received 5.0% but the richest 40% accrued 95% of additional income.[3]

Consider a more proportional scenario in which the income of the poorest decile rises at the same rate as global GDP per capita, whilst maintaining growth at the 1993-2008 rate. Everyone will be brought above the $1.25-a-day poverty line by 2068 and above the $5-a-day poverty line by 2076, this path would only necessitate increasing GDP by a factor of 3 and then 9. If we go even further and assume a pro-poor bias in global growth changing the ratio of income growth for the poorest decile to that of global GDP per capita to 1.82 (i.e. incomes of the poorest rise faster than World GDP) the results are even more impressive. Extreme poverty is eliminated by 2042, the entire population is brought over the $5-a-day line by 2076 and the scale of GDP growth required to achieve this is reduced by a factor of more than 28 than on our current projection.[4]

The idea that economic growth on the current model will not eradicate extreme poverty over acceptable time horizons is lent support by other recent literature. A recent study that looked at the relation between increases in GDP per capita and childhood undernutrition between 1970 and 1996 across 36 low and middle-income countries found a very small to non-existent association between increases in GDP per capita and undernutrition.[5] Even those that contested the findings of this study argued only that there was a marginal decrease in undernutrition associated with GDP per capita growth.[6] These findings are echoed in a recent study which looked at the link between GDP per capita growth in Indian states and found no consistent evidence for the relationship between growth and undernutrition.[7] One plausible mechanism for this effect is that many of those groups presently at risk of undernutrition receive such a small share of the rewards of economic growth that they are not able to increase the quality and quantity of the food they consume.

Another recent study claims that on current projections it will take until 2111 for sub-Saharan Africa to reach 100% lower-secondary school completion and at least 100 years for the average scores for students in the developing world to begin to match the average scores of students from developed countries in 2015.[8] Another study indicates that at current rates it will take 96 countries beyond 2040 to achieve the 2015 Millennium Development Goals relating to significant reductions in maternal mortality and providing universal access to reproductive healthcare.[9] This study also claims that it will take 23 sub-saharan African countries beyond 2040 to achieve the 2015 Millennium Goal of reducing child mortality by two thirds.[10]

Given how pessimistic Woodward’s conclusions regarding the current model of economic growth are it is worth asking whether there are any reasons to doubt his reasoning.

Woodward excludes China from his projection of current economic growth and poverty alleviation. If China were included the resulting numbers would be more optimistic.[11] China is obviously an outlier but one might hope that other countries could follow China’s enormous rates of growth and poverty reduction. However, there is good reason to suppose China’s economic rise is exceptional. Chinese growth is broadly attributable to economic reforms begun in the 1970s. Huge gains in productivity were gained by moving away from collectivised agriculture, relaxing price controls, and easing restrictions on the migration of farmers to cities. Other aspects of the economy were gradually reformed to introduce more private industry and greater competitive pressure on state owned industries.[12] [13] The model of huge economic growth through gradual liberalisation of a centrally planned command economy is clearly not directly applicable to most countries today.

Additionally, even on the very optimistic projections of global growth (an average of 4% growth in GDP per capita per annum), the time horizons for poverty elimination are still extremely unsatisfactory, with extreme poverty elimination only projected to occur by 2072.[14]

One might think that as economic growth occurs the share of this growth amongst the population will eventually become more equal. The economist Simon Kuznets hypothesised that as an economy develops market forces lead to an increase in inequality, increased investment opportunities cause the rich to become richer whilst influxes of labour into the cities keep down wages amongst the working class. Eventually processes associated with the development of industrialisation and the welfare state lead to an increase in equality. This hypothesis can be captured by an inverted U-shaped curve in which income per capita is plotted on the horizontal x axis and inequality on the Y axis.[15] However whether this hypothesis is borne out in reality is highly controversial. [16] [17] [18]

One might be concerned that the line of $1.25 a day adjusted for purchasing power parity might not be the best way to measure poverty. However other indicators do not give a more cheerful prognosis. The Multiple Poverty Indicator which measures material deprivation through questions about housing, healthcare and diet places almost 1.7 billion people in “acute poverty” as opposed to the 1 billion people who live in “extreme poverty” according to the $1.25-a-day metric of the World Bank.[19] [20] There is no reason to think continued economic growth along current lines is any more likely to impact this measure of poverty than the $1.25 dollar a day metric. In any case the number of people who need to be lifted from this form of poverty is even greater to begin with.

Finally there are reasons to think that Woodward’s projections might actually be too optimistic. One such reason cited by Woodward himself is that the growth in income for the poorest decile between 1993-2008 probably reflects a number of favourable shifts for the poor in this period such as increased debt cancellation and more focus on poverty reduction rather than structural adjustment, the adoption of the Millennium Development Goals and a doubling of per capita aid receipts between 2001-2008.[21] Extrapolating income growth rates from this period implicitly assumes a continuation of this pro-poor shift at a similar rate for the indefinite future. And there is no reason to think this will necessarily be the case.

Another issue is that when we consider the increasing carbon and environmental constraints facing the global community it is plausible that this enormous level of growth may not even be viable. On current projections we will need to increase the size of the global economy by a factor of 173 times it’s 2010 size. Even if we expect future growth to be much less carbon and resource intensive the sheer size of the economic growth required may be beyond what carbon and resource constraints allow. This is not just a question of not wanting to pursue economic growth at such a cost to global warming and the environment, but also a question of whether such growth is even possible. Global warming and resource depletion will have a highly negative impact on standards of living, particularly harming the poorest in the world. These factors might be absolute constraints on what levels of economic growth and poverty alleviation can be achieved. [22]

An assessment of the literature indicates that we should be sceptical that “business as usual” economic growth will lead to eradication of extreme poverty over anything approaching an acceptable time horizon. Environmental and carbon constraints may even make it the case that eradicating extreme poverty will be impossible on our current trajectory. To eradicate poverty over a time-period of decades rather than centuries, the world’s poorest people need to receive a much greater share of global income. It may be necessary to radically change our entire approach to economic development to achieve this, but exactly what radical changes would be required and whether such changes would be tractable is unclear. Certainly one means to attempt to achieve this is to encourage many more people living in wealthy countries to give a significant portion of their income to charities operating in low-income countries.

Incrementum ad Absurdum: Global Growth, Inequality and Poverty Eradication in a Carbon-Constrained World

Association between economic growth and early childhood undernutrition: evidence from 121 Demographic and Health Surveys from 36 low-income and middle-income countries

Association between economic growth and early childhood nutrition

Is Economic Growth Associated with Reduction in Child Undernutrition in India?

[](Why Wait 100 Years? Bridging the Gap in Global Education. 

http://www.brookings.edu/research/reports2/2015/06/100-years-gap-global-education-winthrop-mcgivney

Progress towards Millennium Development Goals 4 and 5 on maternal and child mortality: an updated systematic analysis

Henry Phipps

Articles

Development, Population Growth and the Mortality-Fertility Link

How Long Will it Take Economic Growth to Eliminate Extreme Poverty?