09 October 2012

SOFI 2012: Hunger Numbers Rise in Sub-Saharan Africa

The State of Food Insecurity in the World report (known as SOFI) came out today, ahead of next week's World Food Day. 

The report from FAO, WFP and IFAD has the following headlines:

  • Hunger numbers are stalling at the global level (867m in 2007-9 and 868m in 2010-12)
  • Hunger numbers for Sub-Saharan Africa are rising, and at an increasing rate:
    • > increasing at about 2m a year between 1991 and 2000
    • > about 1m a year between 2000 and 2005
    • > about 3.5m a year between 2005 and  2008
    • > about 6m a year between 2008 and 2011
  • These are alarming increases in level and speed of increase.
  • FAO has made 5 changes to methodology and has recalculated all the hunger data back to 1990.  The 1bn number in the 2010 SOFI report has been quietly buried.
  • The biggest change in the numbers is generated by better accounting for losses of food in the retail system.  Accounting for this loss increases hunger numbers substantially, especially in 1990, but not in 2010-12. This means that hunger numbers have declined faster than previously thought and that the achievement of the MDG target, based on this indicator, is closer than previously thought.
  • The odd thing is that for the "developing world" the hunger estimates were 870m before the 2007-8 food price spikes (i.e. in 2005-7) and are lower after the food price spikes (853m in 2009 and 852m thereafter). 
  • This suggests the food price spikes did the most damage in sub-Saharan Africa, but this is also a little odd given that this is where the translation of world to national prices is weaker, on average, due to lower levels of global integration.

So, at first blush, the report is a confusing mix of good and bad news.

P.S. Congratulations to FAO for improving the methodology.

8 comments:

Richard Jolly said...

Thanks for this which directed me to the Q and A notes of SOFI. I'm afraid I am left with the confusion of results which you underline. FAO may have adjusted its methodology a bit but I can't help thinking that their basic model is deeply flawed. You know much more about all this but can one trust the numbers to give a reliable picture of hunger, let alone changes in hunger? Stats on under weight or stunting seem more meaningful and reliable by almost an order of magnitude. And FAO' s conclusion that growth is a key determinant of reductions in hunger seems to me to play to the donor and Bretton Woods gallery, even though they add an 'of course this must include growth of agriculture'. But growth of production and income of small scale farmers seems to be greatly underplayed and also growth of income of poor people in general. Richard

Michael Lipton said...

FAO hs not announced any abandonment of the Alexandratos method for estimating numbers of hungry. So presumably variations in "FAO hungry" are merely disguised variations in mean food consumption estimated from "food balance sheets" (and thus depending for Africa on largely worthless numbers for food output), adjusted by income [not calorie intake] distribution data that don't even use full household-survey information.

The "World Hunger Report" is based on these valueless data. Several people confronted the WHR presenters at the London release last year about this, and were met with complete agreement from IFPRI, but no commitment to change anything.

Direct WHO resurveys of weight-height-age and <Z-squared results are incomparably better (though local rather than national), but of course measure medium-term effects of food deprivation (but also of disease environment and, to a modest extent, work requirements and delayed impact of fetal food deficiencies).

We have no usable measures of levels or trends in current hunger, and not the least idea of ongoing changes year-to-year in sub-Saharan Africa! We do know that there is no medium-term improvement, because WHO site and population numbers for which repeat-survey undernutrition has risen from the previous to the most recent survey are about the same as numbers for which it has fallen, and so is mean shortfall. Incidentally, together with (not too unreliable) food trade data, this also shows that SSA's alleged rise in per-person calorie production - and probably, therefore, staples output - is illusory.

Judith Appleton MBE said...

Still no calculation of food lost to diarrhoeas, a major factor in the proxy under-3/5 nutstat figures we use, and consequently a practical area to target for water work, as well as for food safety and nutrition which are both in FAO. However must be grateful for small mercies, even methodogical.

Lawrence Haddad said...

Given that the FAO numbers are based on food supply figures, FAO have, it seems to me, minimised the gap between what they figures measure and what we want them to measure (i.e. hunger). This was long overdue.

But it is not really what we want to measure--hunger. How to do that?

Stunting is an indicator that can be collected well, is comparable across time and space. But it measures much more than hunger.

The best thing to do would be to fund/incentivise every country to do a food consumption survey every 5 years and then supplement that in the intervals with lighter touch data on an annual basis.

FAO are exploring the latter option with Gallup.

Unknown said...

Increased hunger means less people have the means to buy food, which, in turn, points to an unstable economy. These countries might want to have some accountants using tencia accounting software go over their cash flows to be sure what's going on before it's too late.

Jason Clement said...

I read that food production has been hit due to the civil conflict and continuous drought within the area. The private buildings who owned centralized dehumidifier australia have reported increased usage and consumption of electricity.

Unknown said...

Most perth business accountants would claim that there is a domino effect to the rise and fall of businesses. It can create a hole that can cause other sectors of the economy to spiral downwards thus the increase of hunger numbers.

Travis Dekker said...

I think that the UNICEF and World Nutrition council needs to boost their intricate hunger reduction plans for these Sub-Saharan nations. Maybe the implementation of business and accounting software will help mitigate profit loss based on food allocation and acquisition.