Introduction of New Food Products in China : Is There a Trend towards Healthier and Safer Products ?

The Chinese society has undergone an important transformation in the last 20 years, with changes in lifestyles due to increasing urbanization and rising income levels. The emergence of modern supermarkets, convenience stores, and hypermarkets has run alongside the changes in consumers’ lifestyles, revolutionizing the Chinese food system and the nature of its food supply. Changes in food consumption patterns have also accompanied these shifts. One of the distinguishing aspects of this modern food sector is the continuous introduction of new foods and beverages to the market, and the communication of their associated attributes through labels, in-store displays, and advertising. The purpose of this paper is to explore whether there have been any trends towards the introduction of products with greater health, nutrition, and food safety attributes to help consumers attain healthier diets making trade-offs between purchase options. Results show that there has been a rise in the number of new products in the Chinese food retail sector, particularly branded products introduced mainly by Chinese companies making food safety claims. It is clear, however, that the new food and drink products follow the consumption trends highlighted in the literature, and, therefore, there are low chances that they will positively influence the Chinese diet to an important degree, as they are mainly reinforcing the observed dietary trends.


Introduction
In recent years, China, the world's most populated country and the second biggest economy in the world, has been experiencing changes in the distribution of its population as well as in the level of its wealth, which has directly affected people's lifestyles (Liu et al. 2014;Mintel 2013;Gordon-Larsen et al. 2014).During the last 15 years, China's population has increased by 7.8% (107 million people), and in the same time, the urban population in China has expanded from less than 1.5% of the total population in 2000, to more than 56% in 2015.This growth has also been reflected in the population density (person/square km) in urban areas, which grew by 81% during this period, whilst real disposable income rose by about 60% during the same period.
The emergence of 'western style' retailers in China has run alongside these changes in the consumers' lifestyles.Hu et al. (2004), Reardon and Swinnen (2004), and Reardon and Hopkins (2006), among others, have explained the strong link between the expansion of modern supermarkets and population growth, and rapid urbanization and rising incomes.According to Zhai et al. (2014), the emergence of modern supermarkets, convenience stores, and hypermarkets has transformed the food sector and the nature of food supply in China, where the number of retailer from outside of China entering the market have increased since 1992.China's food system used to be dominated by small farms, ration stores, and fresh markets, but this has now shifted to a direct relationship between the manufacturer or processor and the supermarket, with an accompanying expansion of imported food products (Zhai et al. 2014;Popkin 2014;Si et al. 2016).Retailers have created new forms of selling that have allowed firms to introduce new, safe, and high-quality foods and beverages to the Chinese market at competitive prices (Gale and Thomas 2005;Gale et al. 2015).As part of this retail revolution, international food companies such as Kraft, Heinz, Coca-Cola, PepsiCo, and Nestlé, and even foreign retail chains such as Wal-Mart, Carrefour, and Tesco, have become established names in Chinese households and are leading to fierce competition between domestic and foreign grocery retailers (Dawson 2013;Cheng 2017).In 2015, American retail giant Wal-Mart generated more than 73.5 billion RMB pre-tax sales, owning 432 stores across China, with a store number growth rate of 5.1% (USDA 2017).In 2018, the Wal-Mart group was ranked as the third largest retailer in China (USDA 2018).
A survey conducted by Ipsos in 2013 revealed that 53% of Chinese consumers liked to try new and adventurous foods (Alick 2016;Ipsos 2014) emphasizing the potential impact of the expanding supply on consumers' food consumption habits.According to Zhou et al. (2015), Chinese consumers had adapted their eating behaviour to the new food landscape, following the expansion of modern retail in urban areas.The major shift had been the availability of packaged food and beverages (Popkin 2014;Baker and Friel 2016;Monteiro et al. 2013), with the result of an increase in snacking and consumption of processed food such as breakfast cereals, confectionaries, ready-to-eat meals, and carbonated sodas, among others (Hawkes and Ruel 2011), and a decrease in the consumption of food prepared at home (Zhai et al. 2014).In addition, the population (mainly middle class and younger people) were diversifying their diets and increasing the demand for products such as meats, dairy, edible oils, and fruits and vegetables (Gomez and Ricketts 2013).
Taking into account the fact that highly processed foods are more convenient and often more palatable, but often high in calories, their consumption has been associated with increasing rates of obesity or obesity-related non-communicable diseases, such as diabetes (Gomez and Ricketts 2013;Popkin and Kenan 2016;Fiolet et al. 2018).In fact, where in the past a large portion of China's population was under-nourished, there is now a rapid increase in the rates of obesity (Gordon-Larsen et al. 2014;Shang et al. 2012).However, note that at the same time, certain processed foods are fortified to deal with specific micronutrient deficiencies in order to help in reducing malnutrition levels (Gomez and Ricketts 2013).
Alongside the modernization of its food system, China's fast-growing economy has stimulated a change in focus from food supply (i.e., based on food security) to food safety (Lam et al. 2013).The food industry has the primary responsibility to prevent food safety problems, but considering the scale, diversity, and complexity of China's domestic food industry, which consists of roughly 450,000 food production and processing companies, of which close to 80% have fewer than 10 employees each, it is challenging to ensure their compliance with high safety standards.Moreover, small enterprises might not be able to implement the vast array of required tests, which can be costly.As a consequence, many food safety scandals have emerged in the last 15 years, mainly due to illegal chemical additives being added during food processing.Although the Chinese authorities consider food safety a national priority, adopting key international standards, the number of food scandals has not declined, and public confidence has not been restored (The Lancet 2014).
Although some research is available on health and food safety issues associated with food production and consumption in China (e.g., Zhou et al. 2015;The Lancet 2014), little is known about the introduction of new food and drink products with health and food safety attributes.This area of research is important, because, on one hand, it provides an indication of what factors, in the eyes of suppliers (i.e., retailers, reflected through their private labels, and manufacturers, through branded products), are important to consumers, and this is reflected in the claims that are made when marketing their products and highlighting particular attributes (e.g., 'containing only natural ingredients').On the other hand, as new products alter the range available to consumers, it is important to explore whether suppliers show any interest in helping consumers achieve better food consumption patterns.We used the works of Smith 2008 (sustainability) and Martinez et al. 2018 (healthy food) as a framework.To understand the rationality behind manufacturers and retailers on the introduction of healthier food, they emphasized that a food supply chain approach must be considered.Smith (2008) argued that corporate social responsibility and the introduction of standards to pass responsibility along the chain was the only way to move towards a more sustainable supply chain.The same could apply to the case of healthy products.The work from Martinez et al. (2018) with retailers showed that although suppliers were supposed to influence retail decisions, retail managers believed that they kept the ultimate decision on their business practices.However, these were influenced by their knowledge about the availability of healthy options, consumer demand, and high price.
The purpose of this paper is to explore the launching of food and drink products with health and food safety attributes in the Chinese market.Specifically, the purpose of this empirical study is to address the following three points: 1.
To identify whether, through launching new products, over time retailers and manufacturers were improving the assortment of products with health and food safety attributes available to consumers; in this regard, whether they could be considered 'agents of change' towards healthier patterns of consumption.2.
To discover which companies were leading the introduction of new products making health and food safety claims and in what categories.

3.
To explore whether foreign retailers and manufacturers played a significant role in the introduction of food and drink products with health and food safety attributes.
The structure of this paper is as follows-it begins with a brief literature review focused on the role of suppliers in shaping consumers' diets.This is followed with the empirical analysis, which provides a description of the data and statistical methods used in the paper.The next section presents and discusses the results and finally some conclusions are drawn.

The Role of Suppliers in Shaping Consumers' Grocery Purchasing Decisions
According to product diffusion literature (Talukdar et al. 2002), the potential penetration of a new developed product depends on the ability of consumers to pay for it (i.e., affordability), their willingness to pay for it (i.e., if the product provides value for money), and the physical accessibility of the product; if consumers do not have access to the product, they will never uptake it and it will not have an effect on their food choice.A key question for this paper is what is offered on retailers' shelves for companies to make trade-offs between purchase options.Dawson (2013) stated that although, through their food buying activity, consumers shaped many of the dimensions of what retailers offered (including which new products to introduce), the decisions on what was available to consumers were made by the retailer (see also Dawson et al. 2008).In other words, consumers choose from the options offered by suppliers, leaving such businesses in a powerful position to shape consumers' choice (Revoredo-Giha 2014).In this regard, the literature explained that the defining characteristic of the point-of-purchase retailers was their capacity to establish the boundaries of consumer choice (e.g., Bava et al. 2009;Dawson et al. 2008;Dawson 2013;Martinez et al. 2018).Since many purchasing decisions were made at the retail point, these boundaries directly affected those decisions (Bava et al. 2009).
Another aspect that is in the hands of suppliers is what characteristics of the products are communicated to consumers.One of the major changes in the food system is that personal communication between buyers and sellers has been replaced with communication through food labels, in-store displays, and advertising, representing important changes in the way Chinese consumers make purchasing decisions (Grunert et al. 2015).Many new retail models have emerged in the Chinese market, including 24-h unstaffed convenience stores (Fung business intelligence 2018) and businesses in the fast-growing e-commerce sector being the mobile shopping alternative extensively accepted by consumers.This is especially important when considering credence attributes (e.g., health and food safety attributes), which must be communicated on the package to inform consumers of their occurrence.In this context, consumers look for information on the label as the product delivery system and the supplier (manufacturers and retailers) decide which message to place on it (Dawson 2013).As stated by Machín et al. (2018), package information can empower consumers to identify unhealthy food products and to make more informed food choices.
The above points on retailers' decisions of what to offer and how to offer them to consumers (e.g., which product attributes to highlight) make them, and also manufacturers, the major influencers in shaping consumers' food preferences and choices (Dawson 2013;Liu et al. 2015;Martinez et al. 2018).Therefore, by exploring the trends of which new products are introduced in the market, it is possible to infer to what extent suppliers are interested in improving the food and drink identifying as healthier and safer products; this is the focus of the rest of the paper.

Data
The data analysed in this paper is taken from Mintel's Global New Products Database (GNPD), which provides information about new products launched in selected countries around the world.For new products launched in the Chinese market, the dataset contains information on 90,720 products from September 1997 to December 2016 by 22,269 manufacturing or retailing firms and considering 30,742 different brands.Figure 1 presents the evolution of the launch of products for the period of 1997-2016.Although the database starts from 1997, only data since the year 2000 were used for the trend analysis based on the suggestion of Mintel due to issues with data collection during the first three years of the dataset.In the total sample, 96.9% were branded products and the remaining 3.1% were private label ones.The launched products in the database corresponded to five groups-new products, new variety/range extension, new packaging, new formulation, and re-launched products.Of these groups, new products and new variety/range extension were the most common, representing 47% and 36%, respectively, of all the launched products.This was followed by products introduced under new packaging, comprising 14% of total launched products.(Dawson 2013).As stated by Machín et al. (2018), package information can empower consumers to identify unhealthy food products and to make more informed food choices.
The above points on retailers' decisions of what to offer and how to offer them to consumers (e.g., which product attributes to highlight) make them, and also manufacturers, the major influencers in shaping consumers' food preferences and choices (Dawson 2013;Liu et al. 2015;Martinez et al. 2018).Therefore, by exploring the trends of which new products are introduced in the market, it is possible to infer to what extent suppliers are interested in improving the food and drink identifying as healthier and safer products; this is the focus of the rest of the paper.

Data
The data analysed in this paper is taken from Mintel's Global New Products Database (GNPD), which provides information about new products launched in selected countries around the world.For new products launched in the Chinese market, the dataset contains information on 90,720 products from September 1997 to December 2016 by 22,269 manufacturing or retailing firms and considering 30,742 different brands.Figure 1 presents the evolution of the launch of products for the period of 1997-2016.Although the database starts from 1997, only data since the year 2000 were used for the trend analysis based on the suggestion of Mintel due to issues with data collection during the first three years of the dataset.In the total sample, 96.9% were branded products and the remaining 3.1% were private label ones.The launched products in the database corresponded to five groupsnew products, new variety/range extension, new packaging, new formulation, and re-launched products.Of these groups, new products and new variety/range extension were the most common, representing 47% and 36%, respectively, of all the launched products.This was followed by products introduced under new packaging, comprising 14% of total launched products.The GNPD classified products into 26 categories (which were also broken down into 71 subcategories)-snacks, bakery, sauces and seasonings, dairy, sugar and gum confectionery, processed fish, meat and egg products, juice drinks, hot beverages, meals and meal centers (i.e., ready meals), desserts and ice cream, fruit and vegetables, baby food, breakfast cereals, alcoholic beverages, chocolate confectionery, side dishes, other beverages, ready-to-drink (RTDs), water, sweet spreads, The GNPD classified products into 26 categories (which were also broken down into 71 sub-categories)-snacks, bakery, sauces and seasonings, dairy, sugar and gum confectionery, processed fish, meat and egg products, juice drinks, hot beverages, meals and meal centers (i.e., ready meals), desserts and ice cream, fruit and vegetables, baby food, breakfast cereals, alcoholic beverages, chocolate confectionery, side dishes, other beverages, ready-to-drink (RTDs), water, sweet spreads, carbonated soft drinks, soup, sports and energy drinks, sweeteners and sugar, and savory spreads.Table 1 presents the frequency distribution of launched products by category.It is of particular importance for this paper that the dataset also provides information on the products' positioning claims (i.e., the product attributes that suppliers want to communicate to consumers).The quality of the claims that manufacturers put on the label must follow the regulation of the country where the product will be marketed.In the case of China, all prepackaged food nutrition information on nutrition labelling must follow the standards of the Food Nutrition Labelling Act (FNLA) which was first introduced in 2011, and revised in 2013 and 2015.FNLA states that 'Labels, instructions and packaging of food and food additives may not contain false or exaggerated information, nor may they make statements about disease prevention and treatment functions'.The identification of the claim has been done through the description of the product label and is a variable in the original database.
A total of 74 different claims were found in the dataset, which were divided into six groups-convenience (e.g., microwaveable), demographic (if destined for a particular demographic), health and nutrition (e.g., low in calories), safety (e.g., no additives/preservatives) sustainable (e.g., environmentally friendly packaging) and other (e.g., premium, cobranded, and novel) 1 .It is important to note, as shown in Figure 2, that a large percentage of the new products introduced in the market do not carry any claim (57% as of 2016).However, this percentage has shown a decreasing trend since 2004 when it was 73%.The increasing proportion of products introduced with claims can be associated with the transformation of the Chinese food sector and the need to convey more information about the products to consumers.The classification of claims was done as follows: for the case of health claims following European Commission Nutrition and Health Claims' definition (claims relating to the growth, development and functions of the body; referring to psychological and behavioural functions; on slimming or weight-control; reducing a risk factor in the development of a disease; and referring to children's development), and for the safety claims, we considered those claims that differentiated the product with the intention to precaution a safety.

Methods
The methods used in the paper comprised descriptive statistics (i.e., frequency distributions and cross tabulations) and trend analyses.With regards to the trend analysis, since the number of launched products is increasing in all the studied categories, in order to explore in which categories the growth of products with health, nutrition, and food safety attributes has been more important, the information was transformed to location quotients (Mayer and Pleeter 1975).Since not all the product categories are equally active in the launching of new products, location quotient allowed for a comparison of the growth of a specific group of products in a category (e.g., those making health claims) with the growth of those products considering all categories.2, that a large percentage of the new products introduced in the market do not carry any claim (57% as of 2016).However, this percentage has shown a decreasing trend since 2004 when it was 73%.The increasing proportion of products introduced with claims can be associated with the transformation of the Chinese food sector and the need to convey more information about the products to consumers.

Methods
The methods used in the paper comprised descriptive statistics (i.e., frequency distributions and cross tabulations) and trend analyses.With regards to the trend analysis, since the number of launched products is increasing in all the studied categories, in order to explore in which categories the growth of products with health, nutrition, and food safety attributes has been more important, the information was transformed to location quotients (Mayer and Pleeter 1975).Since not all the product categories are equally active in the launching of new products, location quotient allowed for a comparison of the growth of a specific group of products in a category (e.g., those making health claims) with the growth of those products considering all categories.
A location quotient is a way of measuring the relative contribution of one specific category to the whole for a given outcome.Let  and  denote the number of products launched with a particular attribute (e.g., health) in category  and the total number of products launched in category , respectively.Similarly, let  and  be the number of launched products with a particular attribute considering all the categories and the total number of launched products, respectively.Then, the location quotient for the category  is defined as in Equation ( 1) The classification of claims was done as follows: for the case of health claims following European Commission Nutrition and Health Claims' definition (claims relating to the growth, development and functions of the body; referring to psychological and behavioural functions; on slimming or weight-control; reducing a risk factor in the development of a disease; and referring to children's development), and for the safety claims, we considered those claims that differentiated the product with the intention to precaution a safety.A location quotient is a way of measuring the relative contribution of one specific category to the whole for a given outcome.Let x i and n i denote the number of products launched with a particular attribute (e.g., health) in category i and the total number of products launched in category i, respectively.Similarly, let x and n be the number of launched products with a particular attribute considering all the categories and the total number of launched products, respectively.Then, the location quotient for the category i is defined as in Equation ( 1) Finally, the location quotients were subjected to a trend analysis using linear regressions: where α i and β i are the trend regression parameters to be estimated, y i is the location quotient, and t is a trend variable that takes the number 0 in year 2000, 1 in 2001, and so forth.A positive value of slope β i indicated that the proportion of products with health nutrition and food safety attributes in category i was growing faster than the proportion of those products when all the categories were included.

Overview of the Launching of New Food and Drink Products in China
As shown in Figure 1, the Chinese market has been experiencing a significant positive trend in terms of the number of new food and drink products launched, particularly since 2009 after recovering from the 2007-2008 financial crisis (Li et al. 2012).This picture illustrates the transformation of the Chinese food system (e.g., Hawkes 2008;Ortega et al. 2014).
Table 1 presents a frequency distribution of launched products by category (Figure 3 presents the evolution since 2000 of the top 10 food and drink categories in terms of the number of launches).The snacks and bakery categories were the most active ones, representing 36% of the total launched products.The data supported Hawkes and Ruel (2011) and Zhai et al. (2014) views that Chinese consumers were increasingly interested in the consumption of snacks.Whilst in 2000, only 116 new snack products were introduced, in 2016 that number had increased to 3219.The figure also showed a positive trend for 'processed fish, meat and egg products' and 'fruits and vegetables'.These trends mirrored those observed in the Chinese diet literature (e.g., Zhai et al. 2014).where  and  are the trend regression parameters to be estimated,  is the location quotient, and t is a trend variable that takes the number 0 in year 2000, 1 in 2001, and so forth.A positive value of slope  indicated that the proportion of products with health nutrition and food safety attributes in category i was growing faster than the proportion of those products when all the categories were included.

Overview of the Launching of New Food and Drink Products in China
As shown in Figure 1, the Chinese market has been experiencing a significant positive trend in terms of the number of new food and drink products launched, particularly since 2009 after recovering from the 2007-2008 financial crisis (Li et al. 2012).This picture illustrates the transformation of the Chinese food system (e.g., Hawkes 2008;Ortega et al. 2014).
Table 1 presents a frequency distribution of launched products by category (Figure 3 presents the evolution since 2000 of the top 10 food and drink categories in terms of the number of launches).The snacks and bakery categories were the most active ones, representing 36% of the total launched products.The data supported Hawkes and Ruel (2011) and Zhai et al.'s (2014) views that Chinese consumers were increasingly interested in the consumption of snacks.Whilst in 2000, only 116 new snack products were introduced, in 2016 that number had increased to 3219.The figure also showed a positive trend for 'processed fish, meat and egg products' and 'fruits and vegetables'.These trends mirrored those observed in the Chinese diet literature (e.g., Zhai et al. 2014).Source: Own elaboration based on Mintel's GNPD data.
In contrast with what has been observed in some western countries (e.g., see Revoredo-Giha 2014 for the UK case), there was a low concentration in terms of the number of companies launching new products.Thus, the 50 most important companies launching products for the period of 1997-  In contrast with what has been observed in some western countries (e.g., see Revoredo-Giha 2014 for the UK case), there was a low concentration in terms of the number of companies launching new products.Thus, the 50 most important companies launching products for the period of 1997-2016 only represented about 13% of all the new products launched in the Chinese market (the top 10 cover only 5%).With regards to their origin, of the top 20 companies, 12 were Asian (i.e., Chinese or from another Asian country) and 8 were western.However, when the entire sample was considered, about 67% of the products were found to have been introduced by Chinese companies.

Number of products
Figure 4 shows the evolution of products since 2000 according to claims.As of 2016, the most popular claims were those associated with demographic groups (e.g., children).Health and nutrition was the most popular until 2012; however, this has showed a slight decrease since 2014.The figure also indicates a steady increase in the number of products making food safety claims, which is in line with consumers' increasing concerns about food safety issues (The Lancet 2014) and with the introduction of the Food Nutrition Labelling Act in 2011.Table 2 focuses on the products making health, nutritional, and food safety claims (which account for about 27% of all the claims made for the period of 1997-2016).The food safety claim of 'no additives or preservatives' is the most frequent, accounting for almost 27.1% off all products considered in the table.This can easily be explained as a reaction to the number of food-related scandals in China (e.g., The Lancet 2014).Furthermore, it is important to highlight the link between food additives and obesity identified by the literature (Chassaing et al. 2015).This is followed by 'low or reduced sugar' and 'vitamin mineral fortified' claims, which represent 10.6% and 10%, respectively.

Trends in the Launching of Products Making Health, Nutritional, and Food Safety Claims
Figure 5 shows the evolution of the top 10 health, nutrition, and food safety claims over time.There is a substantial difference between the number of products making the claim 'no additives and preservatives' and the rest.Regarding the claims other than 'no additives and preservatives', they also show a positive trend, but this trend is more modest.Table 3 illustrates the trends in the launching of products making health, nutritional, and food safety claims within each product category.As was expected, not all the claims were present in all the product categories.The table shows that for the case of products making food safety claims (i.e., 'all natural', 'GMO-free', 'hormone-free', and 'no additives/preservatives'), the products increasing fastest are those with the 'no additives/preservatives' claims, followed by 'GMO-free' products and 'hormone-free' products.In contrast, 'all natural' products show below average growth rates, with a statistically significant coefficient around 0.2 for the juice drinks, breakfast cereals, and water categories.The 'GMO-free' claim was found to be important for sauces and seasoning, with a statistically significant coefficient of almost 0.5.Moreover, products making the 'no additives or preservatives' claim were above the mean growth for several categories, especially for juices and baby drinks.Finally, the claim 'hormone-free' was found to be important for new products in the processed fish, meat, and egg products category, and also for breakfast cereals and the fruit and vegetables categories.
One category of particular interest is dairy products due to the food scandals that have been associated with it (The Lancet 2014).For this category, the two claims that were found to be important were 'all natural' and 'no additive/preservatives', both showing growth that was statistically significant.Table 3 also includes a trend analysis for the case of health and nutrition claims.It should be noted that all the estimated coefficients were statistically significant and positive in almost all cases, revealing that new food products making health and nutrition claims were growing faster than the average.However, important differences between the claims and the food categories were observed.For instance, meals and meal center products made only four of the studied claims that were growing-gluten-free, reduced allergens, low transfat, and vitamin and mineral fortified.This was important because these products were the most representative of a transition towards an unhealthy western diet high in fat, salt, and glycemic load (Baker and Friel 2016;Monteiro et al. 2013).The results also showed that the claims suppliers were introducing the most (i.e., in most food categories) were 'vitamin and mineral fortified', followed by products with 'other function' claims (e.g., good for the throat, with ingredients that improve health), 'low or no allergens', 'low or reduced transfat', and finally, those with 'antioxidants'.).With regards to their origin, of the top 20 companies, 12 were Asian (i.e., Chinese or from another Asian country) and 8 were western.However, when the entire sample was considered, about 67% of the products were found to have been introduced by Chinese companies.
Figure 4 shows the evolution of products since 2000 according to claims.As of 2016, the most popular claims were those associated with demographic groups (e.g., children).Health and nutrition was the most popular until 2012; however, this has showed a slight decrease since 2014.The figure also indicates a steady increase in the number of products making food safety claims, which is in line with consumers' increasing concerns about food safety issues (The Lancet 2014) and with the introduction of the Food Nutrition Labelling Act in 2011.Table 2 focuses on the products making health, nutritional, and food safety claims (which account for about 27% of all the claims made for the period of 1997-2016).The food safety claim of 'no additives or preservatives' is the most frequent, accounting for almost 27.1% off all products considered in the table.This can easily be explained as a reaction to the number of food-related scandals in China (e.g., The Lancet 2014).Furthermore, it is important to highlight the link between food additives and obesity identified by the literature (Chassaing et al. 2015).This is followed by 'low or reduced sugar' and 'vitamin mineral fortified' claims, which represent 10.6% and 10%, respectively.shows that for the case of products making food safety claims (i.e., 'all natural', 'GMO-free', 'hormone-free', and 'no additives/preservatives'), the products increasing fastest are those with the 'no additives/preservatives' claims, followed by 'GMO-free' products and 'hormone-free' products.
In contrast, 'all natural' products show below average growth rates, with a statistically significant coefficient around 0.2 for the juice drinks, breakfast cereals, and water categories.The 'GMO-free' claim was found to be important for sauces and seasoning, with a statistically significant coefficient of almost 0.5.Moreover, products making the 'no additives or preservatives' claim were above the mean growth for several categories, especially for juices and baby drinks.Finally, the claim 'hormonefree' was found to be important for new products in the processed fish, meat, and egg products category, and also for breakfast cereals and the fruit and vegetables categories.Source: Own elaboration based on Mintel's GNPD data.
One category of particular interest is dairy products due to the food scandals that have been associated with it (The Lancet 2014).For this category, the two claims that were found to be important were 'all natural' and 'no additive/preservatives', both showing growth that was statistically significant.Table 3 also includes a trend analysis for the case of health and nutrition claims.It should be noted that all the estimated coefficients were statistically significant and positive in almost all cases, revealing that new food products making health and nutrition claims were growing faster than the average.However, important differences between the claims and the food categories were observed.For instance, meals and meal center products made only four of the studied claims that were growing-gluten-free, reduced allergens, low transfat, and vitamin and mineral fortified.This was important because these products were the most representative of a transition towards an unhealthy western diet high in fat, salt, and glycemic load (Baker and Friel 2016;Monteiro et al. 2013).The results also showed that the claims suppliers were introducing the most (i.e., in most food categories) were 'vitamin and mineral fortified', followed by products with 'other function' claims (e.g., good for the throat, with ingredients that improve health), 'low or no allergens', 'low or reduced transfat', and finally, those with 'antioxidants'.
When considering health claims by product category, the food category with a major number of health and nutrition claims was baby food, which showed growth for 18 out of the 28 health and nutrition claims considered, with the digestion, immune system, and diabetic being the ones with the highest coefficients.For the categories of fruit and vegetables and RTDs, 14 claims had been observed   When considering health claims by product category, the food category with a major number of health and nutrition claims was baby food, which showed growth for 18 out of the 28 health and nutrition claims considered, with the digestion, immune system, and diabetic being the ones with the highest coefficients.For the categories of fruit and vegetables and RTDs, 14 claims had been observed and all of these showed significant activity.For fruit and vegetables, the most relevant claims were immune system, brain and nervous system, added calcium, bone health, and digestion.RTDs were primarily associated with antioxidants and slimming claims.The categories of sauces and seasoning, snacks, juice drinks, and dairy followed in importance regarding the number of associated claims (13, 12, 11, and 11 respectively, out of 28 health and nutrition claims).

Number of products
For the product category 'sauces and seasoning', the most important growth was in the products with low or reduced sodium and low or reduced transfats.Snacks displayed low coefficients for all claims (i.e., low growth), except for the case of 'high protein'.This category also had a negative coefficient for the claim 'low/reduced sugar and vitamin fortified', indicating a reduction in the number of products introduced in this category and claims compared with the average products introduced.Given the increasing importance of snacks in Chinese consumption (Zhai et al. 2014), the fact that there were no claims with regards to calories or fat levels reflected that the new products that were launched aimed only at exploiting their increasing importance, but without aiming at improving their nutritional quality.
The health and nutrition claims that showed growth for juice drinks category were antioxidant, vitamin fortified, low/reduced transfat, and slimming.In addition, dairy products displayed the following claims-bone health, brain and nervous system, antioxidants, and anti-ageing.Finally, the product category associated with less number of claims was savory spreads, with no activity for any claim, followed by sports and energy drinks, carbonate soft drink drinks, other beverages, and alcoholic beverages.
If we focus on the five top health and nutrition claims, it can observed that the 'low, no or reduced sugar' claim is especially important for sports and energy drinks, baby food, water, and sugar and gum confectionary.The 'vitamin, mineral fortified' claim is associated with new sweeteners and sugars, fruit and vegetables, baby food, and sweet spread products.'Other function' claims are employed for new sweeteners and sugars, fruit and vegetables, baby food, and RTDs.For the claims 'added calcium' and 'digestive (improves digestion)', a clear link to the categories baby food and fruits and vegetables can be observed.Zhai et al. (2014) had characterized China's changes in food consumption between 1991 and 2011 as a state in which macronutrient composition had shifted towards fats, and protein and sodium intakes remained high and potassium intake low.In this context, the results indicated that the introduction of new products might only partially offset these trends due to the fact that the number of products with low or reduced calories, fats, sodium, and sugar was still very limited.Moreover, growing categories associated with western diets, such as ready meals (i.e., meals and meal centers) and snacks, did not reflect these claims.

Major Companies Introducing Products with Health, Nutritional and Food Safety Claims
Table 4 illustrates which companies are leading the introduction of new products, making health and safety claims, and the importance of foreign retailers and manufacturers.As expected, due to the low number of private label products, all of the top five companies launching products making the selected claims were manufacturers and not retailers.It could be observed that the concentration of the number of products introduced by the top five companies was low in most of the cases except for three claims-low/reduced glycemic, low/reduced carbohydrates, and high satiety.For these claims, the products introduced by the top five companies represented about 50% or more of the products introduced.However, they were made by very few companies.The claims of low/reduced lactose, gluten-free, prebiotic, bone health, caffeine-free, low/reduced saturated fat, and hormone-free products introduced by the top five companies also represented from 20% to 50% of all products introduced in these categories.For the rest of the claims, the introduction of new products with health claims was distributed among a large number of companies.This situation revealed a low concentration on the introduction of food products by companies in the Chinese market, i.e., many firms introduced few products each.Despite the above, some companies are particularly active in the development of products associated with different claims, such as the Chinese dairy companies Inner Mongolia Yili Industrial Group and Inner Mongolia Mengniu Dairy Group.Moreover, the top five companies launching products with anti-ageing, diabetic, hormone-free, and GMO-free claims are Chinese.
Regarding the role of western suppliers in the Chinese market, it is clear that some multinational companies play an important role in introducing new products making health, nutrition, and food safety claims.Thus, firms such as Nestlé and Heinz are very active, showing in the top five ranking for several claims (14 out of 34 selected claims).Others such as PepsiCo and Coca-Cola are also leading the introduction of new products in some of the claim categories (e.g., high protein, high added fiber, and reduced calories).Western manufacturers are also particularly important for the introduction of wholegrain products.

Conclusions
The purpose of this paper was to explore the launching of food and drink products with health, nutrition, and food safety attributes by product category in China.This is important because the literature points out that China's food consumption patterns and eating and cooking behaviours have been changing dramatically towards an unhealthy western diet since the 1990s.Therefore, the availability of products with healthier and safer attributes, if adopted by consumers, can help to some extent to offset this trend.
It is important to highlight that the analysis indicated that about 50% of the products introduced in the Chinese market do not make any associated claim.This shows issues regarding the communication of food attributes to consumers, which can be very useful to consumers as a guide to improve their diet.The data showed that the Chinese retail market is dynamic, and it is continuously launching new food and drink products extending consumers' preference, with snacks and bakery as the leading categories.This is expected to reinforce the observed increase in snacking, which is one of the highlighted trends in Chinese consumption.According to Statista (2018), the retail sales value of the snack food market in China increased from 191.7 billion yuan in 2009 to 544.5 billion yuan in 2018.Whilst the data showed some positive trends towards the introduction of snacks with reduced or low sugar, snacks with reduced calories, fat, cholesterol, or saturated fats were not found.The most popular animal-source food in China is pork, and the consumption of poultry and eggs are also increasing.This is reflected in the new product data by an increase in the introduction of processed fish, meat and egg products, and ready meals (i.e., meals and meal centres).However, similar to snacks, these products do not reflect many attributes that may improve Chinese diets.
Products making food safety claims, particularly those with the 'no additives or preservatives' claim, were found to be increasingly important.This was not surprising due to the history of food scandals in China, and products making these claims certainly aimed at restoring consumers' trust in food.The food categories where health and nutrition attributes were more actively communicated were baby food, fruit and vegetables, and RTDs, and the five top health and nutrition claims were 'low, no or reduced sugar', 'vitamin, mineral fortified', 'other functional', 'added calcium', and 'digestive (improves digestion)".
Regarding which companies were leading the introduction of new food making health, nutrition, and food safety claims, it was observed that the introduction of new products was mostly in the hands of Chinese firms, and it was very fragmented (the top launching companies represented a small percentage of the total number of products introduced in a category).Western firms played an important role in the introduction of new products in several claim categories, but they also represented a small percentage of the total number of launched products.With regards to branded versus private labels, retailers did not play a significant role in developing new products under their private labels, and they represented less than 4%.This result was in line with that of Xu et al. (2018), and explained the lack of awareness that Chinese consumers, especially those aged over 40, had regarding private labels (Mintel 2016;Lupton et al. 2010).
Overall, whilst the analysed data showed some trends towards an increase in the number of products making health, nutrition, and food safety claims, it was clear that the new food and drink products being launched followed the consumption trends highlighted in the literature.Manufacturers and retailers were extending the offer of food to fulfill the needs of a population looking for convenient and on-the-go food such as snacks and bakery with limited associated health claims.Therefore, the data showed that food suppliers were not trying to improve the Chinese diet by launching a higher proportion of healthy products.Thus, companies were making a significant effort on differentiating safe and natural products to conquer consumer trust.

Figure 1 .
Figure 1.Number of products launched during 1997-2016.Source: Own elaboration based on Mintel's Global New Products Database (GNPD) data.

Figure 2 .
Figure 2. Products launched without any claim during 2000-2016.Source: Own elaboration based on Mintel's GNPD data.Note: As percentage of the total number of launched products.

Figure 2 .
Figure 2. Products launched without any claim during 2000-2016.Source: Own elaboration based on Mintel's GNPD data.Note: As percentage of the total number of launched products.

Figure 3 .
Figure 3. Number of products launched for the top 10 food and drink categories during 2000-2016.

Figure 3 .
Figure 3. Number of products launched for the top 10 food and drink categories during 2000-2016.Source: Own elaboration based on Mintel's GNPD data.

Figure 4 .
Figure 4. Number of launched products by claim group during 2000-2016.Source: Own elaboration based on Mintel's GNPD data.

Figure 5 .
Figure 5. Number of launched products by top 10 health and nutrition claims during 2000-2016.

Figure 5 .
Figure 5. Number of launched products by top 10 health and nutrition claims during 2000-2016.Source: Own elaboration based on Mintel's GNPD data.

Table 1 .
Frequency distribution of launched products by category 1 .
Source: Own elaboration based on Mintel's GNPD database.Note: 1 Considers data from January 1997 to December 2016.

Table 2 .
Frequency distribution of products with health, nutrition, and food safety claims 1 .
Source: Own elaboration based on Mintel's GNPD database.Note: 1 Considers the entire database, i.e., data from 1997 to December 2016.

Table 3 .
Results of trend regressions on location quotient of products with health, nutrition, and food safety claims by product category.

Table 4 .
Top 5 companies introducing products with safety, health and nutrition claims during the period 1997-2016.