Copenhagen Consensus Center
Home Menu

Post-2015 Consensus: Food Security and Nutrition Perspective, Horton Hoddinott

Summary of Benefit - Cost Ratios by Country for stunting target - By 2030, reduce by 40% the number of children who are stunted. 

Note - Categorized by discount rates and final working ages

  Benefits to Age 36 Benefits to Age 50 Benefits to Age 60
Country 3% 5% 3% 5% 3% 5%
DRC $7 $4 $12 $5 $15 $5
Madagascar $19 $10 $34 $13 $43 $14
Ethiopia $21 $11 $37 $14 $46 $15
Nepal $25 $13 $45 $17 $56 $18
Uganda $26 $13 $45 $17 $56 $18
Tanzania 29 $15 $51 $20 $64 $20
Burma $34 $17 $60 $23 $75 $24
Kenya $34 $17 $60 $23 $75 $24
Bangladesh $35 $18 $62 $24 $78 $25
Sudan $45 $23 $80 $31 $100 $32
Nigeria $48 $24 $85 $32 $106 $33
Yemen $56 $29 $100 $38 $124 $40
Pakistan $57 29 $101 $39 $126 $40
Vietnam 69 35 $123 $47 $154 $49
India $76 $39 134 $51 $168 $53
Philippines 86 $44 153 $58 $190 $60
Indonesia $94 $48 $166 $64 $207 $65


Nutrition has always been a key development indicator.  Inadequate nutrition is a major contributory factor for child mortality; good nutrition is important both for healthy growth and cognitive development. Cognitive development itself leads to educational success, and both are important determinants of labour productivity and hence economic growth. Balance is also important: there should be neither under- nor over-nutrition.

The Millennium Development Goals included a target of halving the proportion of underweight children between 1990 and 2015. This has served its purpose to focus attention on nutrition, but can be improved upon in two ways when setting the post-2015 agenda. First, stunting (a measure of height for age) is a better indicator of nutritional status and second, halving the number who are hungry rather than the proportion under-nourished is a stronger and more meaningful goal in the light of population growth in the affected regions.

A child growing up on a diet of starchy staples is likely to end up stunted by age two (after which catch-up is difficult), although of normal weight. The underweight goal would have suggested that all countries in South America were meeting their nutritional target by 2008 while, on the basis of stunting, five of the 13 countries would not be on track.

The most recent draft of the SDGs from the Open Working Group includes adoption of two WHO nutrition goals on stunting and wasting:

  • Reduce by 40 per cent the number of children under 5 who are stunted
  • Reduce and maintain childhood wasting to less than 5 per cent

Stunting has a number of advantages as a goal. It is a good indicator of the quality of the early life environment (including nutrition during pregnancy and early childhood, health status and quality of care), and child height at two is a good predictor of achieved adult height. Adult height is itself correlated with hourly wages and tracks economic development quite well. There are, nevertheless, a few disadvantages. In particular, height deficits cannot be overcome in a single generation, since they are to some extent determined by the mother’s own height. Also, height is a measure of long-term nutritional status and takes no account either of wasting due to famine nor obesity.

As for the quantitative target, merely extending the time horizon by five years to fit the SDG timescale seems unambitious, but on the other hand the target itself is considerably more stretching than the MDG. Based on a UNICEF estimate of 169 million stunted children in 2010, the 2030 target should be to reduce this to 101 million (40% reduction).

There are a number of ways in which height can affect future income.  There is a direct effect of taller individuals earning higher wages, particularly for manual jobs. Individuals with improved cognitive abilities (and increased schooling) also earn more. There are also potentially increased health costs for chronic disease in adulthood, associated with under-nutrition in childhood. A study in Guatemala followed up individuals two or three decades after they had taken part in a controlled trial of a nutritional supplement in childhood. Individuals who were not stunted at three years of age had a 66% higher household consumption in adulthood, which can be taken as an estimate of the returns to better nutrition (reduced by 10% to 59.4% to be a little conservative).

The nutritional intervention package used is expected to reduce stunting by 20% (elimination would require a much more wide-ranging – and costly – set of measures to improve agricultural output, empower women and invest in sanitation). This intervention has been modelled for 17 countries with a high burden of stunting (nine countries in Africa and the Middle East, five in South Asia, and three in East Asia, with a combined population exceeding 2.5 billion). The costs are calculated for a cohort of children born in 2015, who receive the interventions up until age two, who enter the labour market at age 21, and for whom the benefits are modelled until they reach age 36.  The dollar value of the benefits is based on current per capita income, projected growth rates of GNP, and the 59.4% benefit from improved nutrition. 

The benefit/cost ratios calculated range from 3.5:1 (Democratic Republic of the Congo) to 42.7 (Indonesia), for a 5% discount rate assuming a final working age of 36.  Countries which are growing faster and/or have higher incomes have higher benefit/cost ratios, because the absolute dollar value of the benefits (due to higher wages) are greater, while there is less variation in costs of the nutrition intervention.

Bangladesh is the median country, with $1,735 worth of benefits (NPV) being delivered for a cost of $97.11, a BCR of 17.9. If a discount rate of 3% is taken, this rises to 35.1. If one assumes the individual works until age 50, the BCR increases to 62.2 under a 3% discount rate. The stunting rate in the country in 2010 was 41.4%, having fallen by 35% since 1990. Assuming that the nutrition package led to a further 20% reduction and the existing downward trend continued at the same rate for another 20 years, this should be sufficient to reduce the rate of stunting by nearly 50% between 2010 and 2030. Bangladesh had 6.3 million stunted children in 2010. In 2015, the total number of children under 5 is projected to be 14.2 million, and if the proportion stunted halves to 20.7%, there would be 2.9 million stunted children in 2025, which achieves the WHO goal.

It is difficult to generalise this estimate to other countries, because Bangladesh has a higher stunting reduction trend than average for middle- and low-income countries. The biggest concern is sub-Saharan Africa, where absolute numbers of stunted children are increasing.

Overall, we can say that stunting is a better goal than underweight.  It is an excellent measure of the health, diet and care provided to children during the 1000 days from conception to age two.  Economic models suggest that the returns to investments in nutrition have high benefit:/cost ratios, and that this should be a top development priority.  A very rough estimate suggests that reducing numbers stunted by 40% by 2030 globally would be a “stretch” goal – optimistic, but possibly achievable with strong effort.