How good is the new hunger data?

ADDIS ABABA (IRIN): After years of criticism, the UN’s Food and Agriculture Organization (FAO) has announced it is exploring new ways to measure “hunger”, “undernourishment” and “food insecurity” – terms used interchangeably – which will dramatically alter the number of people believed to be going hungry.

In what officials stress are “preliminary” steps, FAO is using a new set of indicators in its annual report, State of Food Insecurity in the World, prepared jointly with the World Food Programme and the International Fund for Agricultural Development. The report revises the number of hungry down to 870 million people, saying the number used after the 2007-2008 food price spike – one billion – was inaccurate because of a lack of updated country data and faulty methodology.

Over 12 percent of the global population is hungry, says the new report. Most of them – 852 million – live in developing countries, where the prevalence of undernourishment is now estimated at 14.9 percent of the population. These new figures place countries closer to achieving the Millennium Development Goal to halve hunger by 2015.

“The new estimates suggest that the increase in hunger during 2007-2010 was less severe than previously thought,” the report says. The effect of high prices during the 2008-2009 economic crisis “was less pronounced than was assumed at the time, while many governments succeeded in cushioning the shocks and protecting the most vulnerable from the effects of the price spike.”



New data, methodology

Despite questions about how FAO, in 2009, arrived at the one billion hungry figure, the number appeared in several significant reports.

FAO began the process of adopting new standards two years ago, and presented the new methodology to the National Academy of Sciences in Washington and at the International Scientific Symposium on Food and Nutrition Security in Rome in 2011. The new variables are comprehensive, taking into account the level of price volatility within a country.

The authors of the report told IRIN via email that the new figures reflect more and better data as well as major improvements in their methodology. The new data include fresh population estimates and household-level surveys of food consumption in 44 countries.

“We now recognise that not all the food produced or imported by a country actually reaches households.”

That has helped improve the agency’s statistical model, which is now sophisticated enough to cover ‘skewness’, or asymmetry, in the distribution of food in any population, making it more accurate.

The new methodology also takes into account food lost during distribution. “We now recognise that not all the food produced or imported by a country actually reaches households,” the authors said.

André Briend, a well-known nutritionist, noted, “FAO had been criticised for years for its approach based on national data with a lot of errors and approximations.” Briend said he is “not in a position to judge how the method they used was implemented in the field and how reliable it is, but if properly implemented, this should be more reliable than their previous estimates.”

Shortcomings

The new figures rely on national data, which often are not readily available or of uncertain quality. The report’s authors acknowledge this is a limitation.

“We only have 57 surveys for 44 countries, though they include most of the more relevant ones in terms of the number of undernourished (the only notable exception is China). To give an idea, the countries for which we have used survey data represent roughly 60 percent of the total number of undernourished.

“One message of this report is that we need country collaboration in providing us access to the detailed microdata (information on household spending, dwelling characteristics, etc.) of the surveys they conduct, but also that we need other types of data,” the authors said.

“We are already presenting a set of additional indicators and proposing to start a new type of survey to collect annually data on a global basis on people’s experience with food insecurity.”

This will be an ambitious project, experts say. Relying on detailed questionnaires or the “new ‘experience’-based indicators that are being proposed - in principle this is a good idea because it taps into some of the unseen effects of hunger,” said Purnima Menon, a nutrition expert with the International Food Policy Research Institute, in an email. “At the same time, years of work often go into developing experiential measures of food insecurity!”



Definitions are key

At the heart of the matter is how “hunger” or “undernourishment” are defined – usually regarded as the inability to access enough food to be able to conduct a healthy and active life.

A key issue was to define what “enough” food is, said a technical note released along with the report.

The FAO method is based on assessing energy requirements. People are considered undernourished “if the level of his or her habitual dietary energy intake is below the minimum level nutritionists deem appropriate,” the technical report states, taking into consideration age, sex and lifestyle.

The other issue was how long a person must be undernourished before their condition is considered chronic; the FAO indicator has settled on a year.

Still, the new methodology does not capture the short-term effects of food price surges or other economic shocks. FAO says it is working to develop a wider set of indicators to capture a better sense of the quality of food people have access to as well as other dimensions of food security.