Source: UNIV OF CONNECTICUT submitted to NRP
ADVERTISING TO CHILDREN AND FOOD CHOICES: IMPLICATIONS FOR PUBLIC POLICY
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0220266
Grant No.
2010-65400-20440
Cumulative Award Amt.
$199,248.00
Proposal No.
2009-04121
Multistate No.
(N/A)
Project Start Date
Jan 1, 2010
Project End Date
Dec 31, 2012
Grant Year
2010
Program Code
[96160]- Agribusiness Markets and Trade
Recipient Organization
UNIV OF CONNECTICUT
438 WHITNEY RD EXTENSION UNIT 1133
STORRS,CT 06269
Performing Department
Agri & Resource Economics
Non Technical Summary
Many urge regulating television food advertising to children to address fast-rising childhood obesity. However, scientific evidence on whether advertising contributes to obesity is weak, as previous studies largely failed to separate the effects of advertising on food choices from those of TV watching (e.g., reduced physical exercise). The proposed study is be the first to tackle the identification problem head-on by directly examining how advertising to children affects their consumption of advertised foods, using a unique panel data set with detailed brand-level information on household purchases of candy products and related advertising aimed at children. The identification strategy exploits a natural experiment involving recent self-regulation programs curbing advertising to children by leading firms. The project has two objectives: (1) to empirically estimate how advertising impacts consumer preferences and choices, with an emphasis on advertising directed to children, in a structural model that distinguishes informative and persuasive advertising; and (2) to increase our understanding of how potential regulation schemes affect competition, industry structure, and consumer welfare, by considering firms' strategic responses to the interventions in a dynamic game framework. This project develops new models to enhance understanding of (1) changes in domestic consumer tastes and preferences and (2) changes in agribusiness structure and conduct and the ensuing public policy implications.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
6045010301010%
6046299301020%
6075010301015%
6076299301020%
6105010301015%
6106299301020%
Goals / Objectives
The proposed project has four objectives: The first objective is to employ data from the U.S. industry of candy, gum and confectionery products to tackle the main difficulty of the obesity debate, the identification issue, by directly examining the impact of TV food advertising on children's consumption of the advertised foods. The second objective of the study is motivated methodologically but has great policy relevance. We attempt to distinguish empirically the informative and the persuasive roles of advertising. The distinction is important because the two different roles advertising plays can have radically different implications for consumer welfare and market outcomes. The third objective is to empirically investigate the relationship between advertising channels and advertising effects on consumers of different demographic groups. The fourth objective deals with the supply side effects of advertising regulations. The research seeks to study the immediate and long-term impacts of possible interventions in TV food advertising practice on industry competition and structure, which in turn would affect consumer welfare. This project can be divided into two phases, roughly corresponding to years 1 and 2, i.e., the demand effects effects of advertising (objectives 1-3) and the supply side effects (objective 4) as well as writing up publicaitons. The demand estimates would shed light on a number of interesting questions, such as how effective advertising to different age groups are in influencing households' decisions,whether past advertising fades quickly, and whether advertising makes consumers less sensitive to changes in prices. We would then be able to tell whether the self-regulation on advertising to children by the candy industry has succeeded in moving Americans towards healthier candy products. The project examines whether certain consumers, for example,lower income consumers, are more affected by the "obesity tax" than other consumers. This project estimates the effects of a selective ban on some media channels. On the supply side, the project generates new knowledge with respect to the impact of advertsing restrictions on pricing and advertsing strategies of firms.
Project Methods
The proposed research starts with estimating a structural demand model. For demand estimation, it employs a mixed discrete choice model that allows for flexible substitution pattern and realistic distribution of consumer preferences among heterogeneous consumers. Based on the results from the structural model, the project analyzes how consumers respond to different intervention schemes and more over, how consumers in different social-demographic groups are differently impacted. Based on the results, counterfactual experiments will be conducted on how different interventions can affect consumer choices, assuming no changes in the marketing strategies of the manufacturers. Then, the project will simulate for supply side long-term responses and industry structure in a dynamic game. Finally, one can evaluate what the full impacts of alternative intervention schemes on consumer and manufacturer's welfare, as well as on the risk of obesity, taking into consideration of the results from both the demand and the supply side. The project employs the following two proprietary data sets. The data are obtained through a purchase contract between the Food Marketing and Policy Center at University of Connecticut and AC Nielson and include (a) advertising data 2006-2008 and (b) household level purchases of candy and four food categories. With the combination of the unique data sets and new methods, the propsoed research will greatly enhance the understanding of changes in consumer tastes and preferences, and it will significantly deepen the understanding of the changes in agribusiness structure and conduct. We summarize the novel contributions of the proposed research in the following three aspects. To our knowledge there is no existing research that has achieved this. Most previous research only focuses on the relationship between TV viewing and weight gains. This approach makes it difficult to separate the effects of TV advertising from those of other activities that occur during watching TV. A novel feature is the use of a natural experiment which aids in the identification process: the introduction of self-regulation in advertising to children by leading firms in 2007. Another important contribution is that the project seeks to separate and identify two different roles advertising might play in shaping consumer preferences, namely the persuasive role and the informative role. Distinguishing the two roles has important policy implications because informative advertising is usually viewed as pro-competitive while persuasive advertising is usually labelled as anti-competitive. In contrast to the abundance of theoretical literature on the two different roles of advertising, empirical studies distinguishing the two roles are scant.

Progress 01/01/10 to 12/31/12

Outputs
OUTPUTS: During the reporting period (01/01/2010 to 12/31/2012), I have conducted three research projects on public policy implications of food marketing to children. Two of the papers have been accepted for publication at top-tier agricultural economics and marketing journals. I am preparing the third paper for the submission to a journal. In carrying out the research projects I mentored and trained a Ph.D. student under my supervision. My coauthors and I have presented research supported by the NIFA grant in numerous professional conferences and invited seminars, including the prestigious invitation-only NBER summer institute workshop. Huang, R., Cohen, M., and Chen, Z. , "The Use of Voluntary Marketing Initiatives to Improve the Health Profile of Kids Cereals", International Industrial Organization Conference, George Mason School of Law, VA, March 2012; invited research seminar at New York University, Fordham University, and University of Massachusetts, 2012 Huang, R. and Kiesel, K., "Does Limited Access at School Result in Compensation at Home--The Effect of Soft Drink Bans in Schools on Purchase Patterns outside of Schools", Using Scanner Data to Answer Food Policy Questions (Sponsored by ERS, USDA), Washington D.C., May 2011; EAAE/AAEA seminar, Freising, Germany, September 2010; Agricultural & Applied Economics Association Annual Conference, Denver, CO, July 2010 Huang, R. and Yang, M., "Buy What Is Advertised on Television Evidence from Bans on Child-Directed Food Advertising", National Bureau of Economics Research Summer Institute, Cambridge, MA, July 2010 PARTICIPANTS: Rui Huang, Investigator. Graduate students. TARGET AUDIENCES: Policy makers, advocate groups concerned with childhood obesity. PROJECT MODIFICATIONS: Not relevant to this project.

Impacts
My grant-supported research has led to a change in knowledge related to food marketing and consumers' food choices. I have evaluated the actual impacts of implemented food marketing regulations and assessed potential consumer and producer responses to potential food marketing policies. One of my papers is among the first to evaluate the effects of banning soft drink from campus on children's at-home soft-drink consumption. Using the Connecticut ban as a case study, we show that banning soft drink from campus does not result in an increased soft drink consumption at home, which implies that the policy could be effective in reducing overall children's soft drink consumption as long as the campus ban itself is effective in reducing at-school consumption. The results are of high policy relevance as many states are considering similar policies in an effort to combat childhood obesity. Another paper under the grant is a first empirical study to assess the impacts of Children's Food and Beverage Initiative (CFBAI), a voluntary advertising restriction initiative, on children's food choices. CFBAI participants commit to stop advertising to children food products that are not meeting certain nutrition standards. Using household scanner data on candy purchases and data on advertising exposure, we show that the CFBAI implementations of two leading candy companies are ineffective in influencing consumers' candy purchases, because the implementation fails to effectuate real changes in children's exposure to television advertising of these two companies' products. The reason is that CFBAI only restricts advertising on programs with children audience exceeding a certain percentage threshold, and the two companies advertise on general programming which can be viewed by a large number of children but not a high percentage of children among the audience. One of the CFBAI implementation we studied, in contrast, led to real reduction in children's television advertising exposure and consequently real reduction on household purchases. Based on the results, we summarize the key conditions for a CFBAI implementation to be able to effectively reduce consumers' purchases of unhealthy products. In a third paper I look at the participation decisions of the companies to join a voluntary marketing initiative such as the CFBAI, in which they either commit to stop advertising of unhealthful products or they need to reformulate the products. This is the first paper that provides a framework to analyze these important firm strategies. Joining the marketing initiative could be image-enhancing for the firms because it showcases their corporate social responsibility (CSR) to consumers. Based on demand estimates, I simulate for firms' interlinked strategies on entering and exiting this CSR marketing initiative, pricing, advertising, and product reformulation in a dynamic framework. The results show that reformulation is a dominant steady state strategy, but there are two distinct strategies to for reaching the steady state. The market leader would continue with their advertising and the challenger would participate in the voluntary CSR advertising restriction.

Publications

  • Huang, R. and Kiesel, K. (2012) Does Limited Access to Soft Drinks in Schools Result in Compensation at Home European Review of Agricultural Economics, doi: 10.1093/erae/jbs003, (in press)
  • Huang, R. and Yang, M. (2012) Buy What Is Advertised on Television Evidence from Bans on Child-Directed Food Advertising, Journal of Public Policy and Marketing, doi :10.1509/jppm.11.114, (in press )


Progress 01/01/11 to 12/31/11

Outputs
OUTPUTS: During 2011, I engaged in research activities on four individual research projects under the grant. I worked with intramural and intermural colleagues for three of the projects, and with a Ph.D. student under my supervision for the other one project. For all the projects, we managed and analyzed data and my Ph.D. student was able to develop many skills an effective and competent researcher needs in the process. I have also engaged in disseminating the funded research. We present two of the research projects to national or international audiences in two conferences. Specifically, we presented "Does Limited Access at School Result in Compensation at Home--The Effect of Soft Drink Bans in Schools on Purchase Patterns outside of Schools" at the Conference of Using Scanner Data to Answer Food Policy Questions (Sponsored by ERS, USDA) held in Washington D.C., May 2011. We also presented "An Empirical Analysis of Advertising, Pricing, & Product Line Extension" at INFORMS Marketing Science Meeting held in Houston, TX, June 2011. We also submitted two papers to peer-referred journals, with one accepted for publication and the other invited for a revision and resubmission. PARTICIPANTS: Not relevant to this project. TARGET AUDIENCES: Researchers, food and beverage industry, policy makers PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
The research conducted under this grant aims to understand questions related to public policies reducing childhood obesity. At this point, the work disseminated has contributed to a change in knowledge. One research project entitled "Does Limited Access at School Result in Compensation at Home--The Effect of Soft Drink Bans in Schools on Purchase Patterns outside of Schools" investigates whether mandatory bans on soft drinks in school would result in compensation in soft drink consumption at home. This is an important question as two states, Connecticut and California, have implemented such bans while many other states are considering similar legislation. If there is a compensation effect, that is, children drink more soft drink at home in presence of such bans on campus, then such bans are not effective in reducing overall soft drink consumption while they hurt school revenues. Our research show that there is no empirical support for such compensation effect, therefore a soft drink ban could reduce overall soft drink consumption among children provided the school bans are effectively enforced. We are working on a follow-up study investigating whether the soft drink ban reduces on-campus soft drink consumption. We presented this paper in an international conference in 2011, and it was accepted for publication in an international journal. Therefore, we believe that this research has resulted in a change in knowledge as no previous studies were conducted on this issue. We hope that eventually the change in knowledge will lead to a change in action. In another research project entitled "Buy What Is Advertised on Television Evidence from Bans on Child-Directed Food Advertising", we examined whether Children's Food and Beverage Advertising Initiative (CFBAI), a voluntary advertising restriction had reduced advertising exposure to children and reduce household consumption of unhealthful foods as it intended to. We find mixed success of CFBAI. In some cases, the industry's response to CFBAI resulted in a dramatic reduction in advertising exposure and household purchases, while in other cases the CFBAI did not appear to change either advertising exposure or household purchases. We find that some of the loopholes in the current CFBAI guidelines are responsible for the limited success so far and we made some policy recommendations based on our findings. This paper was submitted in 2011 and it was invited for a resubmission in a national journal specifically focusing on public policy issues. We hope disseminating this work in this journal in the future will change existing knowledge on this voluntary advertising restriction. We also presented "An Empirical Analysis of Advertising, Pricing, & Product Line Extension" in a national marketing conference, which we believe has resulted in a change in knowledge.

Publications

  • Huang, R. and Kiesel, K. 2012, "Does Limited Access to Soft Drinks in Schools Result in Compensation at Home", European Review of Agricultural Economics, (pending)


Progress 01/01/10 to 12/31/10

Outputs
OUTPUTS: The objective of the project is to understand how advertising directed to children affects food preferences and choices of Hispanic children. To this purpose, we use a unique proprietary data set acquired by the Food Marketing and Policy Center at the Department of Agricultural & Resource Economics, University of Connecticut from the Nielsen Company. Since the data first arrived in early 2010, we have been dedicated to data cleaning and exploratory analysis. We presented our preliminary findings in the poster session of the Annual Meeting of Agricultural & Applied Economics Association in July, 2010. However, during this process, we encountered some data quality issues and contacted the Nielsen Company. The Nielsen Company acknowledged these issues and provided us with updated data in late November, 2010. Therefore, we are redoing our analysis with the updated data right now for the presented work. PARTICIPANTS: PI: Rui Huang Graduate student: Zhu Chen TARGET AUDIENCES: Policy makers and advocate groups concerned with childhood obesity PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
Three dominant candy makers, namely, Hershey's, Mars, and Cadbury Adams have signed on the Children's Food & Beverage Advertising Initiative (CFBAI). CFBAI is a self-regulatory program launched in 2006 that is designed to influence children's diet choices by changing the advertising mix of food and beverage products. The three firms pledged to stop advertising their products to children under 12 on television and started implementing their pledges during the middle of our data period. Using a unique data set with detailed information on household candy purchases, and brand level advertising exposure to children audience and audience above 18 via both English and Spanish TV programs, we investigate whether the implementation of their CFBAI pledges have been effective in reducing exposure faced by Hispanic children under 12 and consequently in changing households' purchase decisions. Using the age-specific advertising exposure data, we did not find any significant reductions in advertising exposure of children under 12, via either English or Spanish TV programs, for any candy products of Hershey's and Mars due to the implementation of their CFBAI pledges. Consequently, there are no changes in purchase choices of these products due to the CFBAI. Cadbury Adams pledged to eliminate all its advertising of its children-targeting bubble gum product "Bubblicious" brand and the implementation of its pledge did result in a drastic reduction in the advertising exposure of the Bubblicious brand. However, the Bubblicious brand did not advertise on Spanish TV programs prior to Cadbury Adam's CFBAI implementation. Also the brand has tiny market share and the number of Hispanic households who bought the products in our sample is too small for statistical analysis. In summary, the CFBAI implementations have failed to effectuate any significant changes in the CFBAI participants' advertising practices except for one bubble gum brand "Bubblicious", during our data period. As a result, we do not find any evidence supporting of a significant impact of the CFBAI implementation on Hispanic or non-Hispanic consumers' purchase behavior. The lack of reductions in children's exposure to candy advertising reveals an important loophole of the CFBAI's guideline. The CFBAI participants self-define the "children-directed programs" on which they are committed to stop advertising of their traditional products. For example, a participant defines the "children-directed programs" as programs with more than 50% of estimated audience being children under 12. There are thus two problems. First, the definitions are based on percentage, but not on the numbers of children audience. Thus, a participating firm does not need to adjust its advertising practice if it only advertises on programs with a large number but not a large percentage of children audience, such as "American Idol". Second, self-defining the threshold of the percentage of children audience could also be self-serving. These findings suggest that these issues have to be redressed in order for the CFBAI to generate any desired impacts on Hispanic consumers' food choices.

Publications

  • No publications reported this period