New Trends in the Chinese Diet: Cultural Influences on Consumer Behaviour.

China is one of the most dynamic regions in the world in terms of economic growth and development. Such development has inevitably influenced the structure and habits of Chinese society. Whilst the economic condition of the middle class and high-income segment has steadily improved, cultural changes are also under way: ancient Chinese traditions now include major elements from other cultures, most notably the West. The above scenario is the background to this paper. A structured research-administered survey was developed to investigate the changes in the Chinese consumer food culture: 500 urban participants were randomly selected from six reference cities, covering geographically almost the whole country. This study aims not only to analyze the propensity of consumers to include food products from other countries in their ancient Chinese culinary culture, but also represents an initial attempt to perform a market segmentation of Chinese consumers according to their degree of cultural openness towards non-Chinese food, taking into account socio-demographic, cognitive and psychographic variables.

Introduction

When considering emerging markets, China is one of the most dynamic regions in the world in terms of economic growth and development. It is no coincidence, therefore, that this area, together with India, Russia and Brazil, has become a significant challenge for companies from the old continent (Horwitz and Budhwar, 2015), generating extensive discussion also in academic circles on how best to exploit the new market opportunities.

Economic development has inevitably influenced the structure and habits of Chinese society: while there is still a great disparity between the richer and poorer classes, a growing middle class and high-income segment have emerged (Cheung and Wu, 2012). These groups are changing in cultural terms, including major elements from other cultures, most notably the West, in the ancient Chinese tradition (Wang et al., 2015). This broadening of cultural horizons is also affecting Chinese food consumption behavior: although the Chinese market has a gastronomic tradition which is profoundly different from its Western counterpart, the emerging social classes are expressing a growing demand for processed food, opening up to foreign influences, which are neither accepted with prejudice nor viewed passively. Therefore, the dynamic and vibrant food Chinese market is evolving, creating room for example for European products which are imbued not only with strong cultural value but also enjoy high standards of quality and food safety (Grunert et al., 2011; Zhang et al., 2012). Given the above scenario, a structured research-administered survey was developed to investigate the change in Chinese consumer food culture. The survey was one of the research activities of the Q-PorkChains EU FP6 Project, which focuses on the development of the European pork industry, testing and supplying innovations in order to improve global competitiveness. Our study had a twofold objective. The first was to analyse the propensity of consumers to include in their ancient Chinese culinary culture food products from other countries. The second was a first attempt to perform a market segmentation of Chinese consumers according to their degree of cultural openness toward non-Chinese food, with due consideration taken of socio-demographic, cognitive and psychographic variables. In particular, the applied framework relies on the analysis of values (Schwartz et al., 2001) and lifestyle (Bruns and Grunert, 1995). This approach, far removed from the classic schemes, allows fairly objective consumer characterization with little influence from the researcher’s assumptions and knowledge. In our study, the hypothesis being tested is that the propensity of consumers to consume foreign food products from countries (the dependent variable) is a function both of the socio-demographic characteristics of consumers and of the values of individuals and several dietary habits.

Materials and Methods

Survey and data

A structured research-administered survey was developed to investigate Chinese consumer preferences. In all, 500 urban participants were randomly interviewed from six reference cities, representing extensive coverage of the country’s disparate regions: Nanjing, in the south-eastern coastal region, Chengdu, in the southwest, Wuhan in the middle, Changchun in the northeast, Beijing in the north, and Guangzhou in the south of China. Nevertheless, the field survey was restricted to two kinds of retailing: supermarkets and local markets. The data we will analyse come from five sections of the questionnaire: i) Schwartz Portrait Value Questionnaire (PVQ) questions (Schwartz et al., 2001); ii) food-related lifestyle (FRL) (Bruns and Grunert, 1995); iii) a reduced version of the CETSCALE to measure ethnocentrism (Sharma, 2014); iv) consumer behavior related to pork consumption; v) socio-demographic characteristics of the respondent. Several approaches have been proposed to characterize and measure personal values. Among the most frequently used are the Rokeach value survey (RVS) (Rokeach, 1973) and the list of values (LOV) developed at the University of Michigan Survey Research Center (Beatty et al., 1985). Schwartz (1992) defined ten values, namely: self-direction, stimulation, hedonism, achievement, power, security, conformity, tradition, universalism and benevolence. The ten values are abstract guides in human life and are able to fulfill three essential needs: i) those of individuals as a biological organism; ii) requisites of social interaction; iii) and the survival and welfare needs of the group (Schwartz, 1992). The PVQ was implemented using the theory of Schwartz and subsequently validated at the international level (Saris et al., 2013). The instrument consists of 56 operational items. Each provides a brief description (portrait) of the individual goals, aspirations or desires. Everything is structured so that it can describe the above values proposed by Schwartz, represented in a circular spatial way, organized in two pairs of different and opposite dimensions: openness to change vs conservation; self-transcendence vs self-enhancement. Values that are juxtaposed in the circular structure are similar in terms of meaning. The relationship between values, attitudes and behavior has been widely investigated (Bernard et al., 2003; Dreezens et al., 2005; Thøgersen and Ölander, 2002) and the flow of causality that seems to emerge from different studies is from values through attitudes to behavior, so that values have an impact on attitudes, which in turn influence behavior (Homer and Kahle, 1988, Maio and Olson, 1994). However, the relation between values and behavior seems to be indirect. Individual values influence behavior mostly indirectly through mediator variables. The intermediate level is played by food-related values, attitudes or lifestyles related to food consumption (Maio and Olson, 2004; Brunsø and Grunert, 1995). Among the different mediators proposed to improve the relationship between values and behavior in the food area is the FRL construct defined as the system of cognitive categories, scripts and their associations which relate a set of products to a set of values (Grunert, 1993; Brunsø and Grunert, 1995).

The hypothesis that FRL could be used as moderators in the relationship between values and food-related behaviors was first tested in two studies (Brunsø et al., 2004), using the so-called LOV instrument (Beatty et al., 1985). Both studies confirmed the role of the FRL construct as mediators. Brunsø, Scholderer and Grunert (2004) used the more sophisticated and complex Schwartz value survey (SVS). Their analysis confirms the influence of values on attitudinal constructs such as FRL, having found a meaningful relationship between values and FRL in Spain, Germany and Asia (Grunert et al., 2011). In this study, FRL variables were used to segment consumers and build a profile through the knowledge of their actual, or also perceived, way of life. The model assumes that the approach to the different individual life dimension is related to a specific mental construct that often does not identify the real lifestyle. The theoretical approach (Brunsø and Grunert, 1995) was implemented operationally through the construction of an instrument, validated for inter-cultural analysis, capable of binding the cognitive structure of consumers and the food choices of these values. The five cognitive categories identified are: i) buying patterns – the importance of product information, sensitivity to advertising, price sensitivity, purchase impulses or shopping lists; ii) methods of food preparation – for the culinary appetite, looking for innovative techniques and recipes, ready meals, improvisation; iii) quality aspects – health, value for money, news, organic products, taste and freshness; iv) consumption situations – rapid consumption vs full meals; v) buying motivations – personal satisfaction, security, social relations. A reduced version of the CETSCALE (Sharma, 2014) to measure ethnocentric attitudes was also used in the questionnaire. More specifically a six-item scale related to the hypothetical negative impact of foreign products on the local economy/local employment was used. In other words, the consumption of products from other countries could be considered wrong because it could have a negative impact on the local economy/local employment (Table 1). One part of the questionnaire was devoted to pork consumption habits. In one of the questions, consumers were asked which drinks could go with pork. The alternatives included Chinese wine vs non-Chinese wine. China is at present ranked fifth in the world for red wine consumption. In the Chinese market more than half of this product is produced domestically, but there are also imported wines widely available, especially from France and the US, which have a higher reputation than locally produced wines (Muhammad et al., 2014).

Table 1.

Ethnocentric scale.

Statement	Mean	SD
Buying goods made in China helps the circulation of China’s economy	5.26	1.23
Buying goods made in China supports local economy and employment	5.43	1.14
A real Chinese person should always buy products which are made in China	4.79	1.46
We should buy products which are made in China and not let other countries prosper before us	4.38	1.55
Chinese consumers who buy products made in other countries may be considered responsible for their Chinese compatriots losing jobs	3.83	1.46
Chinese products are the first and most important choice	4.94	1.39

SD, standard deviation. Cronbach’s alpha=0.85.

Methodology and empirical framework

In this study, the hypothesis being tested is that the propensity of Chinese consumers to consume foreign food depends on dietary habits (FRL), values of individuals (PVQ), ethnocentric attitudes, and the socio-demographic characteristics of consumers. The statistical model adopted is that of multinomial logit. This model represents a generalization of the logit model when the dependent variable may assume more than two outcomes that are not ordered. In this paper, the dependent variable consists of a set of four cases:

The model assumes that the probability of the response for the i-th observation being equal to the outcome g is as follows: where x is the vector of explanatory variables characterizing the i-th respondent, and β is the vector of unknown parameters for outcome g obtained by the maximum likelihood estimator (Greene, 2012). The dependent variable was constructed using the scores assigned by consumers to six different items included in the food-related lifestyle section chosen among those most related to the research aims (Table 2).

Table 2.

Items used for the dependent variable.

Statements		Mean	SD
New way (Likert scale 1-7)*	I love to try recipes from abroad	4.95	1.34
	I like to try new recipes	4.86	1.35
	I look for various ways to prepare uncommon meals	4.84	1.43
Security (Likert scale 1-7)°	I dislike anything that may change my habits in food and drink	4.62	1.56
	I only buy and eat the food I know well	5.76	1.43
	Dishes that I am familiar with give me a sense of security	5.16	1.4

*Cronbach’s alpha: 0.60

°Cronbach’s alpha: 0.54.

Results and Discussion

According to the average scores assigned by the consumers to the factors new way and security (including 6 items), the sample was divided into three tertiles for each of the two factors used (high/average/low score for new way; high/average/low score for security). This procedure gave us four classes of consumers: traditionalist, high score for new way and low score for new way; indifferent, average score for both security and new way; nationalist, innovators-high score for both security and new way; innovator, high score for new way and low score for security.

The adopted approach allows a first Chinese market segmentation according to the categories described above, showing the propensity of consumers to foreign food.

As shown in Table 3, innovators represent over 14.2% of the sample, 21.75% of the sample can be identified as nationalist-innovators, while 41.75% are indifferent and 22.34% of the sample can be termed traditionalists.

Table 3.

Propensity to foreign food, sample share of the categories.

Segment	Sample share (%)
Innovator	14.2
Nationalist-innovator	21.71
Indifferent	41.75
Traditionalist	22.34
Total	100

The gradient of the propensity to foreign food shown in this segmentation may be considered as an indirect index of change and openness that is affecting the economy and Chinese society. In this context, identification of a segment of pure innovators of around 14%, consumers ready for exotic tastes and perceptions, confirms the potential of the Chinese market as a potential destination for European food products in general and for Italian products in particular. The market share increases if we include the first mentioned segment, that of nationalist-innovators, thus constituting a potential market of more than 1/3 of the total population.

The explanatory variables are divided into four groups: i) PVQ; ii) ethnographic attitude; iii) behaviour toward national and international wine; iv) socio-demographic variables.

Principal component analysis with orthogonal (Varimax) rotation was performed on the 10 Schwartz values obtained from the PVQ answers collected in the questionnaire. The scores of the extracted factors were subsequently interpreted following the meta-value dimensions and included as regressors in the econometric model.

Table 4 shows Schwartz metavalues obtained from the polymerase chain reaction: i) conservation (comp. 1); ii) openness to change (comp. 2); iii) self-enhancement (comp. 3); iv) benevolence (comp. 4). Estimated meta-values are consistent with the Schwartz circle (Schwartz et al., 2001).

Table 4.

Principal component analysis loading matrix.

	Component 1	Component 2	Component 3	Component 4	Unexplained
Benevolence	-0.04	-0.08	-0.03	0.83	0.15
Universalism	0.38	0.15	-0.11	0.24	0.30
Self-direction	0.07	0.75	-0.13	-0.08	0.20
Stimulation	-0.19	0.58	0.24	-0.02	0.31
Hedonism	0.07	-0.09	0.61	0.11	0.33
Achievement	-0.06	0.19	0.30	0.39	0.34
Power	0.03	0.01	0.65	-0.13	0.36
Security	0.41	0.13	-0.13	0.19	0.34
Conformity	0.49	0.07	0.03	-0.03	0.39
Tradition	0.63	-0.12	0.14	-0.17	0.25

The socio-economic variables used in the econometric model and statistically significant in at least one model are as follows: i) age; ii) gender; and iii) living in rural areas.

Also a variable related to preferences for Chinese vs foreign wine was used. Our hypothesis is that the choice between Chinese wine and imported wine is, among other factors, affected by values such as traditionalism.

Table 5 shows the results of the estimates of the parameters, indicating the direction and magnitude of the statistical associations between the explanatory variables (x vectors) and the probability for the i-th respondents to be assigned to the g-th group or Prob (y=g). The group of traditionalists, which makes up 22.34% of the sample, shows a positive association with the Meta value, which includes the conservation values, namely tradition, conformity and security. This meta-value emphasizes order, self-restriction, preservation of the past, and resistance to change. This segment of consumers tends to be older: if they had to choose a wine to go with pork they would buy a Chinese wine.

Table 5.

Multinomial logit estimates.

		Coefficient	SD	z	P
Traditionalist (high score for Security and low score for New way)	Conservation	0.58	0.13	4.37	0
	Openness to change	-0.13	0.15	-0.88	0.38
	Self-enhancement	-0.1	0.14	-0.73	0.46
	Benevolence	-0.16	0.15	-1.09	0.27
	Age	0.02	0.01	1.73	0.08
	Gender	-0.45	0.31	-1.44	0.15
	Living in rural areas	-0.17	0.15	-1.14	0.25
	Chinese wine	0.7	0.31	2.26	0.02
	Ethnocentrism	0.19	0.16	1.2	0.23
	Constant	-2.48	0.92	-2.71	0.01
Nationalist-innovator (high score for both Security and New way)	Conservation	-0.09	0.12	0.8	0.43
	Openness to change	0.13	0.13	1.01	0.31
	Self-enhancement	0.21	0.12	1.74	0.08
	Benevolence	0.23	0.17	1.38	0.17
	Age	0.01	0.01	0.55	0.58
	Gender	0.26	0.27	0.96	0.34
	Living in rural areas	-0.35	0.14	-2.44	0.02
	Chinese wine	-0.28	0.26	-1.05	0.29
	Ethnocentrism	0.6	0.14	4.32	0
	Constant	-3.64	0.84	-4.34	0
Innovator (high score for New way and average or low score for Security)	Conservation	0.01	0.11	0.04	0.97
	Openness to change	0.27	0.12	2.28	0.02
	Self-enhancement	-0.07	0.12	-0.59	0.55
	Benevolence	-0.03	0.12	-0.27	0.79
	Age	-0.01	0.01	-0.71	0.48
	Gender	0.41	0.26	1.61	0.11
	Living in rural areas	-0.17	0.13	-1.31	0.19
	Chinese wine	-0.26	0.25	-1.03	0.3
	Ethnocentrism	-0.21	0.13	-1.6	0.11
	Constant	0.39	0.72	0.54	0.59

SD, standard deviation.

The group of nationalist-innovators, which makes up 21.71% of the sample shows a positive association with the meta-value of self-enhancement, including achievement and power values. This meta-value emphasizes pursuit of one’s own interests and relative success and dominance over others. This group of consumers is highly ethnocentric and lives in urban areas.

The last group, that of innovators (14.2% of consumers in the sample), shows a positive association with the meta-value openness to change, which includes stimulation and self-direction. Centered on independence and readiness for change this meta-value emphasizes independence of thought, action, and feelings and readiness for change. This group sees the prevalence of the female gender, and has a negative correlation with ethnocentric attitudes.

Conclusions

The analysis presented in this study on the propensity for innovation in the food habits of Chinese consumers provided interesting insights, despite showing some limitations. The latter concern essentially the sample that, albeit representative of heavily urbanized areas of the country, is really small compared to the Chinese population of one billion. This problem is not extremely significant when considering some further aspects: the extent of economic and social disparities in China, the lack of knowledge of the Chinese market currently possessed by analysts and Western food business operators. For the above-mentioned factors, this study can be considered an initial analysis to investigate the ancient, diverse and little-known Chinese culture. Moreover, from a marketing strategic point of view, it would be not very useful to investigate the rural population, still in an early stage of economic development and cultural openness.

Within the limits of the proposed analysis, the results seem to confirm the major potential of certain segments of the Chinese market (namely, the urban middle class and upper income bracket) as a destination for quality Western products. Identifying the segment of pure innovators – the consumers ready to buy and consume foreign foods – by three types of variables (FRL, PVQ and socio-demographic) provides valuable information not only for analysts but also for food business operators in the West and in Italy in particular.

In addition, using a cognitive, modern approach in market segmentation provides an open-minded Chinese consumer profile that is very useful for the implementation of policies for specific marketing and promotion strategies. The analysis also assumes a major role if we consider the level of closure that characterized China in the past and the presence of a strong and meaningful cultural identity in all dimensions of Chinese existence. Therefore, the challenge of any marketing strategy for penetrating the Chinese market will rely on offering a different approach to life.

Latent demand for Westernization is growing in all emerging countries. From a marketing point of view, this opportunity could well be captured by food products, which can confer valuable intangible characteristics such as a different life-style and a feeling of modernity.