Oxfam partners with Coca-Cola to study the company's poverty footprint in Zambia and El Salvador

Oxfam America in March released a report analyzing the poverty footprint of beverage giant Coca-Cola and multinational bottling company SABMiller in Zambia and El Salvador.  The report was jointly authored by the three organizations.

Marion Nestle gave Oxfam a hard time about this report: "I can only guess that Coca-Cola’s grant to Oxfam must have been substantial."  In a comment on Marion's blog, Chris Jochnick from Oxfam explained that Coca-Cola had contributed $400,000 to the research project, and -- separately from this research project -- had given Oxfam $2.5 million in 2008-2010 for humanitarian work.

Altogether, I feel the Oxfam project contributed to the companies' public relations messaging, overstated the companies' beneficial contribution to local economies, under-emphasized the health concerns about their impact, and did not adequately preserve Oxfam's own independence in the cooperative analysis.

Oxfam America is truly my favorite humanitarian assistance organization -- because of sensible economic and policy commentary combined with good works on the ground -- so I hope my blog post on this particular report gets a thoughtful reading from Oxfam staff. In particular, I have no complaint about Oxfam's vision for the private sector role in economic development. Yet, I did not like this report.

Poverty footprint

First, the report appeared to credit the companies with contributing more than $100 million in economic activity to the local economy, generating millions of dollars in tax revenue for local governments and creating many thousands of jobs.  However, after reading the report closely and asking Oxfam staff some questions about it, I think readers should be careful not to think of those dollars and jobs as a real impact of Coca-Cola's presence.

The report itself has a bold statement of its analytic goals: 

Oxfam is developing the Poverty Footprint Methodology as a means to understand the full range of impacts multinational corporations have on poor communities, and to provide a platform for engagement around those impacts.

The report's most important quantitative results imply the companies have a large and beneficial macroeconomic impact:

An examination of the Coca-Cola/SABMiller value chain’s macroeconomic impacts reveals that its Gross Value Added (GVA) in 2008 was approximately $21 million in Zambia and $83 million in El Salvador.  In addition, the Coca-Cola/SABMiller value chain supported an estimate of more than 3,741 formal and informal jobs in Zambia and 4,244 formal jobs in El Salvador.

However, these numbers are not a correct estimate of the companies' "footprint" or impact on local economic activity, tax collections, and jobs.  If Coca-Cola did not exist, or were not allowed into Zambia and El Salvador, two things would be different from the current situation: (a) other beverage companies, including local companies, would sell more product, and (b) other beverages, including traditional beverages and water, would provide a larger share of the consumer's hydration needs.

If other beverage companies took up the slack, much of the economic activity and tax payments and job creation would have happened anyway.  It would be interesting to know how much profit Coca-Cola takes out of the local economy and returns to international shareholders in the United States and Europe.  At times, the Oxfam report appeared to be addressing the issue, but it didn't really.  Buried deep in the report, footnote 22 on p. 84 acknowledged: "The Coca-Cola Company’s profit information was not shared with the research team." 

I asked Oxfam if the analysis compared the situation with Coca-Cola to a situation without Coca-Cola, which is the relevant comparison for assessing "impact."  Helen Dasilva of Oxfam replied, "The objective was not to compare an economy with the system to an economy without."  The result is to give the companies credit for big dollar impacts that overstate their real contribution to job creation and the economy.  This is an analytic approach that one commonly sees when a county or State or industry boasts about the importance of its local economic activity, but this is not an approach that an independent analysis should take in assessing a multinational company's impact in a developing country.

Sugar-sweetened beverages and obesity

Second, the report included no critical discussion of expanded consumption of sugar sweetened beverages, displacement of traditional foods and beverages in the diet, and rising rates of overweight and obesity in developing countries.

When I asked Oxfam about this, Dasilva responded:

The focus of this project was not to study or address the issues surrounding obesity, nor did we conduct an analysis of the impact of Coca-Cola products on overall nutrition or health. That was a result of our limited bandwidth.

Dasilva agreed to forward some of my questions to Coca-Cola, which would not answer specific questions about the growth of sales of sugar-sweetened beverages in Zambia and El Salvador.  In particular, because a large fraction of the population in developing countries is children, I asked about growth in sales to children.  Coca-Cola's answer was clever:

We have a global Responsible Marketing Policy that covers all our beverages, and we do not market any products directly to children under 12. This means we will not buy advertising directly targeted at audiences that are more than 35% children under 12. Our policy applies to television, radio, and print, and, where data is available, to the Internet and mobile phones. Because of this policy we do not track sales to children under age 12 as it is against our global policy to directly target this age group with any marketing for our beverages.

I think of the advertising policy as a secondary issue. The real question is how much full-sugar Coke is the company selling to children. I don't believe a policy about advertising is sufficient reason to dodge a question that was not about advertising, but rather was about sales.  This is a tough question that Oxfam should have asked Coke but didn't.

Oxfam's independence from Coca-Cola messaging

Third, because of the joint authorship, it is impossible to tell what parts of the report are Coca-Cola writing, and what parts are Oxfam writing.

I asked Oxfam if this joint authorship caused the organization to make compromises in the language it would have used in a report that it authored independently.  Dasilva responded:

Bringing two significant multinationals and a global development organization together to agree on language in any report will be challenging. This report was no different and the result isn’t perfect. While it is safe to assume our varied cultures, missions and ways of working led to differences of opinion, it would be tough to pinpoint specific language differences given how many comments from all sides went into the final document.

I appreciate Dasilva's frank answer.  An Oxfam-authored report would have been quite different from this jointly authored report.  I look forward to reading the Oxfam-authored version some day.