12 March 2013

ATNI: A new metric of how pro-nutrition the big food and beverage companies actually are

I am a fan of indices.  Admittedly there is not much research to show whether they do or do not change anything.  But in the absence of such evidence, my priors are that they do.

So I really welcome the Access to Nutrition Index from GAIN, supported by the Bill and Melinda Gates Foundation and the Wellcome Trust, and released today.  The research was done by MSCI Environmental, Societal and Governance research, a company that guides investors who place a special emphasis on these kinds of dimensions.

ATNI ranks 25 major food and beverage (non-alcoholic) companies on a range of indicator categories (weights of each category give in brackets):

  • Governance (12.5%) Corporate strategy, governance and management
  • Products (25%) Formulation of appropriate products
  • Accessibility (20%) Delivery of affordable, available 
  • Marketing (20%) Responsible marketing policies, 
compliance and spending
  • Lifestyles (2.5%) Support for healthy diets and active lifestyles

  • Labeling (15%) Informative labeling and appropriate use of health and nutrition claims

  • Engagement (5%) Engagement with policymakers and other stakeholders


1. On a 10 point scale the majority of companies scored below 3.  The top 3 companies (Danone, Unilever and Nestle) scored between 6 and 7.  Clearly room for massive improvement.

2.  The companies that are breast milk substitute manufacturers are designated by a red flag in the rankings. Three of  5 these companies score in the top 8. The report commends Danone and Nestle for their relatively high overall mark, but it says their "reported lack of compliance with the International Code of Marketing of Breast Milk remains a significant concern".

3.  Companies’ practices often do not measure up to their commitments. Broadly, companies’ scores on nutrition strategy and governance were higher than their scores on product formulation, accessibility, and marketing. And within each of these areas, their level of implementation lagged behind their stated commitments.

4.   Many companies are not very transparent about their nutrition practices. This lack of disclosure limits the ability to understand the full scope of companies’ nutrition-related efforts.


1. The methodology seems sound (but only a preliminary view--I have only had the report for a day).   The methodology process has tried to be inclusive.  The scoring has to rely on a lot of self reported data, but tries hard to get independent verification. I like the weighting which gives greater emphasis to what companies actually do, rather than what they say. I also like the fact that companies who do not cooperate or are not transparent with their data receive lower scores.  

2. Some would say that it is difficult to reconcile breast milk substitute (BMS) companies such as Danone getting high overall scores given violations of the Code.  I also think this is incongruous--one could argue that violations at this stage are so consequential that they negate whatever is done thereafter.  I would give this much higher weight in later versions although I realise the technical difficulties given that essentially there is a mixing of apples and oranges.  Perhaps there should be a separate ranking for BMS firms. 

3. ATNI constructs 2 sub indices: one on overweight and obesity and one on undernutrition.  While the rankings of companies does not change much, the scores for  undernutrition are much lower.  The report argues that this is because companies' undernutrition efforts are more philanthropic and not yet incorporated into core business. This accords with my own non-systematic experience and remains a massive challenge.

4. What features of the companies drive rankings?  Size? Nationality? Market share? Corporate governance?  History?  We only have 25 observations so statistical analysis is challenging, but some correlation work might throw up some interesting hypotheses to test in future case-studies.   

5.  Are there conflicts of interest?  GAIN works with many of these companies as do MSCI.  I am not too concerned about this, because the process seems transparent and the funding is from two third parties who would suffer greatly if they were seen to fund anything other than a process that aimed to be unbiased.  Many organisations also receive funding from governments to rank government performance (e.g. CGD, ODI and IDS) and so this is a wider challenge. I should note that I am a member of GAIN's Partnership Council, so I also have a potential conflict of interest. 

6. Finally, will this index change anything? Do any indices change anything?  The people who generate these indices tend to say they are highlighting best practice, and not naming and shaming.  Personally I think there is nothing wrong with doing both.  IDS is developing the Hunger and Nutrition Commitment Index (HANCI) which assesses the commitments of high burden countries to hunger and malnutrition reduction.  We are thinking hard about how to evaluate the indicator in terms of change that it might facilitate and we will be using regression methods and country case studies to do this assessment. 

So while not perfect, and while it will no doubt improve in later versions, I welcome ATNI and congratulate GAIN, Gates, Wellcome and MSCI for doing something that is not easy--technically or politically.   


Robert Peccoud said...

Definitely appropriate in these times of easternblochorsemeatinourdailylasagna panic. Nothing wrong with indices of course as long as they remain information useful to the citizens and their beloved decision makers. The trick is to ensure their usefulness, that they will be taken into account in the policy building processes. The example of the development aid index is alas not very encouraging.

klaus von grebmer said...

Great and insightful piece. Thank you very much. I agree very much with your point 2 in the reactions part.

Unknown said...

Lawrence, This is a very useful analysis, and I appreciate the insights regarding practice versus policy. In that regard, I think one of the most interesting components of the ATNI is the one on labeling. Countries scored lowest on this practice of any of the subcomponents. Given the importance of information to long run market performance, this is discouraging, and hopefully will also attract some notice.
Laurian Unnevehr

bars in Singapore said...

This is a great inspiring article.I am pretty much pleased with your good work.You put really very helpful information. Keep it up. Keep blogging. Looking to reading your next post.

Unknown said...

Your Blog Is Very Good.It providing useful information for all. Am also using similiar products.