Voluntary initiatives by the food and beverage industry have been proposed by policy makers and industry stakeholders as one strategy for improving population-level nutrition( 1 – Reference Kraak and Story 5 ). From a public health perspective, voluntary industry initiatives may be attractive for several reasons. They may be more achievable than government measures, which can be hampered by pressure from interest groups, political gridlock and bureaucratic inertia, and may allow to achieve public health objectives quicker, more efficiently and less intrusively than governmental regulation( Reference Yach 6 ). From an industry perspective, a number of motives for engaging in such initiatives may exist, including: contributing to socially desirable ends; creating publicity for the brand and goodwill among stakeholders; framing the public debate to align it with company goals; creating tactical disagreements among potential critics; and preventing binding government regulation and fiscal measures, among others( Reference Koplan and Brownell 3 – Reference Kraak and Story 5 , Reference Nixon, Mejia and Cheyne 7 , Reference Kraak, Swinburn and Lawrence 8 ).
Against this background, several large European food retailing chains have announced voluntary nutrition initiatives( 9 – 11 ). Food retailing chains play an important role in population-level nutrition. In most high-income countries they are the most important source of foods and beverages consumed by the population, and can influence consumption through decisions on stocking, pricing and promotions( Reference Taillie, Ng and Popkin 12 – Reference Sacks, Robinson and Cameron 14 ). Moreover, with their bargaining power over manufacturers and their range of own-brand products they can impact the upstream food production system( Reference Taillie, Ng and Popkin 12 , Reference Sacks, Robinson and Cameron 14 ).
For the present case study, we selected a major European food retailer with a relatively comprehensive publicly available nutrition strategy and used the approach recommended by the private-sector module of the INFORMAS (International Network for Food and Obesity Research, Monitoring and Action Support) food environment monitoring framework to analyse its scope and content( 15 – 17 ).
Methods
Analytical framework
A number of frameworks have been proposed for the analysis of food industry initiatives in the field of nutrition and health( Reference Kraak and Story 5 , Reference Kraak, Swinburn and Lawrence 8 , Reference Dibb 18 – Reference Sacks and Vanderlee 23 ). We used the INFORMAS food environment monitoring framework( Reference Sacks, Swinburn and Kraak 24 ). The INFORMAS framework has been developed by an international expert group based on reviews of the available evidence and is, to our knowledge, the most comprehensive and widely accepted framework of its kind( Reference Swinburn, Sacks and Vandevijvere 25 , 26 ). It comprises twelve modules and three steps of data collection and analysis, namely minimal (step 1), expanded (step 2) and optimal (step 3)( Reference Sacks, Swinburn and Kraak 24 , Reference Swinburn, Sacks and Vandevijvere 25 ). Here, we applied the approach recommended by the module on private-sector policies and practices, and conducted a step 1 analysis which comprises the collection and analysis of an organization’s publicly available policies and commitments sourced from organization websites, reports, the media and the organization directly( Reference Sacks, Swinburn and Kraak 24 ). Following the framework, we focused on policies regarding product composition with respect to salt, added sugar, saturated fat, trans-fatty acids and energy density, as well as on policies regarding marketing and promotion, product availability, portion and package sizes, pricing, and nutrition information and education( Reference Sacks, Swinburn and Kraak 24 ). Further details are provided in the online supplementary material, Supplemental File 1.
Case selection
Our case selection was informed by two criteria: company size, namely being among the ten largest European food retailers by annual sales in 2016; and existence of a publicly available nutrition strategy with quantitative targets covering the whole own-brand product range. We chose the first criterion because, in the context of our analysis, company size by sales can be considered a workable proxy for the company’s potential impact on population-level nutrition. The rationale for the second criterion was that we were seeking to identify a company with relatively more comprehensive publicly available nutrition commitments and policies rather than a typical case representative of the food retailing industry in general. To select a case we did a scoping assessment of the companies meeting the first criterion, namely the Schwarz Group, Tesco, Carrefour, Rewe, Metro, Edeka, Aldi, E.Leclerc, ITM and Auchan( 17 ). Methods and results of this assessment are reported in detail in the online supplementary material, Supplemental File 2. In short, we reviewed company websites, contacted companies by email and collated statements relating to topics covered by our analytical framework. We found that most companies are involved in nutrition education activities, often in partnership with non-profit organizations. Moreover, some companies have committed themselves to non-specific reduction targets. However, based on the information reported on the company websites or provided to us by the companies, we found that Lidl is the only retailer among those we have assessed which has issued a comprehensive nutrition pledge including quantitative targets with timelines for its whole own-brand product range.
Geographic scope
To determine the geographic scope of our analysis we undertook an assessment of Lidl’s international operations. Methods and results of this assessment are presented in detail in the online supplementary material, Supplemental File 1. In short, we reviewed the websites of Lidl International, as well as of Lidl’s national operations in the USA, Germany, France, Great Britain, Spain, Ireland, Northern Ireland, Austria, Switzerland, Luxemburg and Hong Kong, and collated statements made by Lidl regarding the issues covered by our analytical framework. Among the countries assessed, we found that Germany is the only one for which Lidl has outlined a comprehensive nutrition strategy including quantitative targets. In our main analysis we therefore focused on Germany, where Lidl generates about one-third of its revenue( 15 ).
Identification and retrieval of primary sources
We used Lidl’s ‘Position Paper Healthy Nutrition’, a forty-page document outlining Lidl’s pledge, published on 25 January 2017 by Lidl Germany, as our key source of primary data( 9 ). We identified further sources of primary data by searching the websites of Lidl and organizations mentioned on Lidl’s website, including charitable and advocacy organizations supported by Lidl, and by conducting Internet searches and retrieving documents cited in Lidl’s position paper. Moreover, we contacted Lidl asking for any information it might be able to share with us regarding its nutrition strategy. We included all documents discussing Lidl’s pledge or other activities by Lidl regarding issues covered by our analytical framework. The full list of documents retrieved and analysed is provided in the online supplementary material, Supplemental File 1.
Data analysis and synthesis
We used qualitative content analysis methods to extract and synthesize relevant information from our primary sources, using a mixed deductive–inductive approach( Reference Hsieh and Shannon 27 ). We used the qualitative content analysis software MAXQDA (Verbi GmbH, Berlin, Germany) for coding. In a first step, two authors (P.v.P. and J.M.S.) independently coded all primary sources, inductively creating and applying codes. These codes were then compared, combined and systematized in a structured code tree consistent with the categories used by the INFORMAS framework. In a second step, two authors (P.v.P. and J.M.S.) used these codes to independently code all primary sources. The coded data were then extracted, compared and synthesized narratively and with tables by the same two authors. A third author (T.L.H.) checked the results for correctness, plausibility and clarity. Further details are provided in the online supplementary material, Supplemental File 1.
Results
Scope and content of Lidl’s nutrient and food group targets
Lidl’s pledges regarding the nutrient content of its own-brand products sold in Germany are outlined in Table 1, and its pledges regarding marketing and promotion, product availability, pricing, portion and package sizes, and nutrition information and education are shown in Table 2. In short, Lidl promises to improve the nutritional profile of its own-brand products – constituting about 75 % of its overall sales in Germany( 28 ) – by reducing the content of sugar, salt, saturated fat and trans-fatty acids, and to reduce their energy density. With regard to monitoring and evaluation, Lidl states that it will ‘continually compare [its] declared goals with actual achievements’. For its sugar and salt reduction targets this is described in greater detail: to measure its progress towards these targets, it will calculate annually the average, sales-weighted content of added sugar and added salt in its own-brand products, and compare results with those of the previous year( 9 ). Concerning transparency Lidl states that revisions to its targets and its continuous comparisons of goals with achievements will be ‘made transparent to the public’( 9 ). No details are provided on how, when and by whom this will be done( 9 ).
Table 1 Pledges on key nutrient and food groups and on energy density of own-brand products made by Lidl
Table 2 Pledges made by Lidl on promotion and marketing, product availability, pricing, and nutrition information and education
Discussion
Lidl’s nutrient and food group targets
Strengths and limitations regarding Lidl’s nutrient and food group targets are summarized in Table 3. In short, key strengths are that the pledge covers a range of relevant nutrients, food groups and intervention areas; that it includes a number of quantitative targets with clear timelines; and that it applies to the complete range of Lidl’s own-brand products sold in Germany. Key weaknesses are the omission of some important intervention areas, such as pricing and labelling, as well as the vagueness of most targets and the lack of independent monitoring and evaluation. The latter shortcomings are of particular concern. A recent review of accountability frameworks to promote healthy food environments through voluntary partnerships between government and the food industry identified a number of criteria that should guide such initiatives( Reference Kraak, Swinburn and Lawrence 8 ). These criteria include clearly defined benchmarks, as well as independent monitoring and evaluation and public disclosure of relevant information – criteria on which Lidl’s pledge falls short.
Table 3 Strengths and limitations regarding the scope, quantification of targets and transparency of Lidl’s pledge
A number of further issues are worth mentioning. First, the pledge contains extensive references to national and international policy documents as well as nutrition guidelines and recommendations by expert institutions and scientific societies( 9 ). Moreover, it explicitly states that Lidl will refrain from fortifying foods and beverages with micronutrients, except for a few cases in which this is generally considered justified, such as the fortification of soya milk with calcium( 9 ). This is remarkable since the food industry has in the past been criticized for using fortification as a strategy to create an appearance of healthfulness in products which are not healthy( Reference Nestle 29 ). Moreover, the pledge does not mention Lidl’s healthy checkout initiative, which was first introduced in its UK stores in 2014, and later in several other countries including Austria, Switzerland and Luxemburg( Reference Lidl 30 , 31 ). This inconsistency might be due to the limited attention this issue has received so far among advocacy groups in Germany as compared with the UK, where a major national advocacy campaign preceded Lidl’s initiative( 32 ).
Moreover, Lidl’s position paper devotes considerable space to issues not covered by our analytical framework, particularly regarding chemical residues from food production and processing and food additives. It specifies numerous quantitative targets regarding them, many of which go beyond the safety thresholds set by German and international food safety authorities( 9 ). The focus on these issues might be due to the well-documented phenomenon that the German public is more worried about chemical residues in food than about energy-dense, nutrient-poor dietary patterns rich in ultra-processed foods and beverages( 33 ).
Comparison with other recent evaluations of food retailing nutrition initiatives
A recent study using a revised and expanded version of the INFORMAS framework examined the nutrition policies and commitments of the four largest Australian supermarket chains( Reference Sacks, Robinson and Cameron 14 ). It found large variations in the transparency, comprehensiveness and specificity of the companies’ policies and commitments. Overall, its findings are in line with the results of our study. Three out of four companies discussed nutrition and health as part of their corporate reporting, and two reported efforts to reduce levels of salt, sugar and saturated fat in their own-brand products, as does Lidl. By contrast, none of the four companies had published formal policies to limit the marketing and promotion of unhealthy foods and beverages, which is also a weakness of Lidl’s pledge. Going beyond Lidl’s approach in Germany, the two largest Australian food retailers have committed to implementing a rating score nutrition labelling system and one of them has committed to introducing healthy checkouts( Reference Sacks, Robinson and Cameron 14 ).
Strengths and weaknesses of the present analysis
Our analysis has various strengths. We use a systematic, evidence- and consensus-based analytical framework, as well as established methods of qualitative content analysis. All steps in the main analysis were carefully executed, with two researchers independently assessing the same documents and the whole research team interpreting overall findings. However, our study also has limitations. Within the INFORMAS framework, we conduct only a step 1 analysis based on information that is publicly available. Unlike the Australian study mentioned above we were unable to use the revised and refined framework, which was developed by INFORMAS specifically for the analysis of nutrition policies and commitments of supermarkets, as this was published after we had already begun our analysis( Reference Sacks, Robinson and Cameron 14 , Reference Sacks and Vanderlee 23 ). While we contacted Lidl’s press office and customer service, we were not successful in sourcing additional information from Lidl. We were also unable to clarify a number of ambiguous statements contained in Lidl’s position paper; in response to our enquiry Lidl stated that it was unable to comment on our questions. Moreover, we purposefully selected a company with relatively more comprehensive publicly available nutrition policies and commitments, implying that the findings of our case study are not likely to be representative of food retailers in general.
Conclusions
Lidl is, to our knowledge, the first major European food retailer committing itself to a comprehensive nutrition strategy which includes at least some quantitative targets with clear timelines covering its whole own-product range. Lidl’s pledge covers a broad range of relevant nutrient and food groups and intervention areas. Furthermore, given Lidl’s position as Europe’s largest food retailer, even minor changes to the nutrient profile of Lidl’s sales would likely be relevant on a population level. These strengths, however, contrast with several limitations. The lack of clarity in the definition of several of its key targets, the absence of information regarding baseline assessments for all but a few sub-targets, and the lack of independent monitoring and evaluation cast doubt on Lidl’s confidence in its ability to achieve meaningful changes and will make any rigorous assessment of its achievements difficult. Moreover, important areas of concern with regard to diet and health, such as energy density, dietary fibre, red and processed meat, as well as pricing and labelling, are not covered systematically.
A number of options for strengthening Lidl’s healthy nutrition strategy exist. It is a general management principle that objectives should be SMARTER, namely specific, measurable, achievable, relevant, time-bound, and (independently) evaluated and reviewed( Reference Yemm 34 ). The relevance of these principles for voluntary industry initiatives in the field of nutrition and health has been highlighted by a number of reviews of such initiatives and relevant frameworks( Reference Kraak and Story 5 , Reference Kraak, Swinburn and Lawrence 8 , Reference Sharma, Teret and Brownell 19 ). If Lidl is serious about its pledge, it should therefore specify quantitative targets, baselines and timelines for all main objectives, using a sufficiently ambitious and comprehensive benchmark as point of reference; commit to use the available range of interventions more consistently; strive to achieve a healthier composition of all products sold, including products not part of its own-brand range; broaden the geographic scope of its pledge; ensure systematic monitoring, evaluation and review, preferably through an independent research institute; set up a mechanism for staying up to date on research and societal developments, for example through establishing a scientific advisory board; and increase transparency, for example through comprehensive progress reports published on a regular basis. Other food retailers, which have not yet developed comprehensive healthy nutrition strategies, should learn from the strengths and limitations of Lidl’s pledge, and account for these from the start.
Acknowledgements
Acknowledgements: The authors would like to thank Gary Sacks (Deakin University) for advice on the use of the INFORMAS framework. All errors are our own. Financial support: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Conflict of interest: H.H. is chairman of the German Diabetes Foundation and scientific advisor and member of the non-executive board of diabetesDE, a diabetes patient organization which has received funding from Lidl. He has not been involved in any communication between diabetesDE and Lidl, and has no knowledge of the content of the confidential partnership agreement between the two organizations. Moreover, he is coordinator of the enable Competence Cluster of Nutrition Research funded by the German Ministry of Education and Research (BMBF, FK 01EA1409E) and has received honoraria for membership on scientific advisory boards of Nestlé and Danone, providing advice regarding the nutritional content of breakfast meals and dairy products. R.L. is member of the board and plenipotentiary of the German Diabetes Foundation and P.v.P. has received travel funds from the German Diabetes Association, which are both member organizations of diabetesDE. All other authors have no competing interests to declare. The opinions expressed in this article are the authors’ own and do not necessarily reflect the views of the organizations mentioned above. Authorship: P.v.P. conceived the study and drafted the manuscript. P.v.P., J.M.S. and T.L.H. searched for, retrieved and analysed the primary data, and conducted the systematic literature search. E.A.R., H.H. and R.L. contributed to the design of the study and provided methodological and content-related advice. All authors discussed the results and implications and commented on the manuscript at various stages. All authors have read the final draft and agreed to its publication. Ethics of human subject participation: Not applicable. Availability of data and material: All sources of primary data used are referenced in the online supplementary material, Supplemental Files 1 and 2, and the full data sets and coding files are available from the authors upon request.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1368980018002744
