In this paper we argue that institutional conditions should be taken into consideration when consumers trust in food safety is analysed. Our survey results demonstrate that levels of trust in food safety varies considerably across our three selected countries: Russian consumers expressing the lowest level of trust, Norwegian consumers the highest and Danish consumers expressing levels of trust in food safety which were in between. We find empirical evidence in all countries that consumers trust in food safety is related to their evaluation of how their national food control authorities perform, as well as to what extend they trust market mechanisms to secure food quality. However, while trust in food safety in the Scandinavian countries is more likely to rest on trust in public food control, trust in food safety more often depend on trust in market mechanisms in the St. Petersburg region.
1 Introduction
In most European countries an increasing number of risks associated with food have appeared on the public agenda during recent decades. Topics like pesticide residuals in fruits and vegetables, food colouring and other additives in processed foods, hormones in beef and BSE have been subject to media debate and consumer concern. Lately, BSE has led to turmoil in the market for beef and to institutional change in public food regulatory and administrator bodies – both within European nations and at the EU level (Tacke 2001).
Whether and how consumers can trust the foods they eat have become an urgent matter with important political and economic implications. In this article, we study how consumer trust in food safety varies within and between three European populations. We analyse data from surveys conducted in Denmark, Norway and the St. Petersburg region of Russia.
Within the area of food consumption, most studies of trust have concentrated on how trust in food varies according to individual variables, to types of food risks and to specific information sources (Hansen et al. 2003). However, recent studies indicate that trust in food safety varies systematically between countries (Berg 2000; Glitsch 2000). This calls for broader contextualisation of variations in trust. Drawing on sociological literature on trust, we argue that nationally specific institutional conditions must be included in order to understand variations and changes in trust within the area of food safety. This is further explored in this article, where variations according to individual variables are analysed in combination with variables reflecting trust in food system institutions in the three populations.
In this study we have selected two countries, Denmark and Norway that are similar in many respects, including general conditions of trust. The debate on food safety issues, however, is quite differently laid out. A third country, Russia, is geographically not very far away, but it is quite dissimilar in many respects, including general conditions of trust. In order to limit and be able to catch this wide variation, and also to make the three cases comparable in terms of size, the St. Petersburg region has been selected.
The quantitative analysis is based on surveys conducted among representative samples of the adult populations in St. Petersburg, Denmark and Norway during 2001. In the discussion and interpretation of results existing qualitative and quantitative material and information are drawn upon. We thereby aim to complement our quantitative cross-national project with more case-oriented data describing the different national and cultural contexts (Rokkan 1955; Ragin 1987; Mjøset 2000).
The article begins with a discussion of why we prefer the concept ‘trust’ instead of ‘risk perception’ in the study of consumer reactions to food safety. It is followed by a short and summary description of important characteristics of the food sector in the three countries/region. After a description of the methods adopted, we present results about variations in consumer trust in food in the three populations. We thereafter raise questions and assumptions about these variations, which are further investigated in multivariate analyses. The following main questions are addressed: How can possible differences in consumer trust in food between the three national populations be explained? On what do consumers base their trust in food? Is it primarily on their understanding of how the food market operates or is it on how they evaluate the public food control system? Does trust in food in the three populations vary the same way according to individual background variables, such as gender, age and educational level?
2 Consumer trust in food safety
A number of studies have approached the field of food safety analysing risk perceptions. It has been shown that public understanding of food risks differ from expert understanding (Slovic 1992). Types of information, credibility of sources of information and risk perception have been studied extensively, as well as the relationship between types of risk and perceived severity (Hansen et al. 2003). Socio-demographic variation with respect to risk perception has been demonstrated in different countries (Schafer et al. 1993; Sjödén 1993).
The above-mentioned types of studies analyse consumers’ perception of food safety as dependent on individual properties such as cognitive risk perception and socio-demographic background factors or on types of food risk, credibility of sources of information, etc. (Hansen et al. 2003). They rarely address the societal and institutional conditions under which perception of food safety is produced and reproduced. Feeling confident eating specific foods such as eggs, beef, fish and vegetables is relying on a long chain of actors and institutions operating in the food system, such as food producers, food retailers and public food control systems (Kjærnes 1999; Berg and Kjærnes 2001).
The concept ‘risk perception’ signals a reflected awareness of some external risk, but according to empirical studies, many consumers demonstrate tacit trust in food safety (Berg 2000/2004; Halkier and Holm 2004). In this article we aim to compare and understand consumers’ perception of food safety within the context of their food system institutions. For this analytical purpose we find ‘trust’ a more appropriate concept than ‘risk perception’, since theories of trust capture both perception of food safety based on trust in persons as well as trust in institutions, on reflected trust as well as on unreflected confidence. In this sense, using the term ‘trust’ represents a social relational approach to consumers and food safety.
Approaching the field of food safety in this manner, implicates that levels of trust in food can be interpreted and partly explained by the differences in the working of the institutional arrangements in the food sector, and by the extent to which the population base their trust in food on these institutions.
The existing literature on trust and institutions suggest that the differences in the workings of the national institutional arrangements affect consumer trust in food in numerous ways. Rothstein (2000) argues in general that the ways in which institutional performance is evaluated by groups in society often has to do with such actors’ considerations of self-interest as well as broader cultural historical processes. Kaase (1999) claims more specifically that evaluation of institutions can be explained by broader cultural dynamics if the institutional and societal context is characterised by stable conditions. But if the societal context is characterised by social change, explanations must be related to the performance of specific institutions. A recent European survey study (Poppe and Kjærnes, 2004) concludes that cultural features, social practices and institutional performance have substantial impact on consumers’ assessments about trust in food.
In relation to the empirical study presented later in this article, different political conditions could imply that countries with a stable and clear institutional structure for public food control would exhibit higher levels of general consumer trust in food than countries with unstable and complex public food control institutions. Differences in economic conditions could imply, that countries with a middle class large enough to support a market of alternative products (such as organic products) would gain higher levels of direct consumer trust in specific kinds of foodstuff than countries with big social differences and a small middle class. Further, cultural differences could mean that countries with many media-stories about lack of food safety would show relatively lower levels of consumer trust in specific foods.
The concept of trust has a long implicit history within sociology even though it never has been of central focus to sociological theory (Misztal 1996). It has been silently present in many classical models of society discussing the changing basis for social integration (e.g. the discussion about Gemeinschaft versus Gesellschaft (Tönnies 1972 [1912]). Recently the concept of trust is explicitly thematized in discussions about late modernity and current tendencies of individualisation, reflexivity and increasing institutional complexity (Beck 1992; Giddens 1994).
The concept of trust has no firm or generally agreed definition. One typical and very general definition states that trust may be seen as ‘confidence in the reality of a person or system regarding a giving set of outcome or events’ (Giddens 1990: 34).
Different levels and types of trust have been suggested. First, trust can be basic in the sense that a lot of peoples daily activities draw upon a ‘taking-for-granted’ understanding of social reality, where knowing what to do is based on habits and tacit knowledge (Bourdieu 1984; Kjærnes and Dulsrud 1998; Ilmonen 2001). This kind of trust, sometimes also defined as ‘trust as habitus’ (Misztal 1996), is fundamental for identity and necessary for carrying out daily life (Giddens 1990; Luhmann 1999). Second, trust can be relational in the sense that it can be based on direct personal contact (Kjærnes and Dulsrud 1998) and may draw upon familiarity or direct personal bonds and obligations. This type of trust is sometimes also defined as ‘trust as passion’ (Misztal 1996) and is based on expectations of benevolent mutuality. The concept relational trust is also applied to analyses of decisions and cooperation. Third, trust can be system oriented, i.e. based on the well functioning of anonymous societal institutions such as the market or the political regulatory system (Kjærnes and Dulsrud 1998). This type of trust is sometimes defined as ‘trust as policy’ (Misztal 1996) and is seen as increasingly necessary in modern large-scale societies as a means of reducing complexity (Luhmann 1999). Further, trust may be tacit, i.e. based in practical consciousness and not intentionally formulated, or it may be reflexive, i.e. based in discursive consciousness (Giddens 1984: 5–14). The different levels and types of trust are not mutually exclusive as they all contribute to the building of concrete trust relationships (Misztal 1996: 101).
In recent sociology of food, trust is brought up in relation to analyses of risk (Kjærnes 1999; Berg 2000/2004; Hansen et al. 2001). When risks are uncertain and contested, trust in food producers, regulatory bodies and communicators becomes increasingly important. In modern societies food consumption is an activity enmeshed in complex interactions between large abstract systems and the intimate social relations and structures that form everyday life.
How consumer trust in food safety is affected by national conditions.
Consumer trust in food in a country is related to which food safety events have taken place and how this is represented in the public debates (in media); how the arrangements of the food system (market and regulatory bodies) work; and how this is viewed by consumers. In the article important food safety issues in each country are shortly presented and the structure of each national food system is described. Survey data are presented about: a) Consumer trust in food safety and b) Consumer trust in institutions (shadowed boxes).
3 National backgrounds for understanding of food safety issues
When comparing understandings of food safety in the three populations it is important to be aware of the big differences in living as well as in societal and institutional conditions in Russia, Denmark and Norway. While the average private standard of living is high in Denmark and Norway, the situation in Russia at the time of the data collection was characterized by social polarization and financial deprivation of a significant part of the population. The two Scandinavian countries have, for many decades, been relatively stable political and social systems, whereas Russia is characterized by dramatic political and social changes. The general background conditions influence the situation in the food sector and how it is perceived by all three populations. Table 1 presents some characteristics of the food sector in the three countries. The table demonstrates several principal points of difference.
. | Denmark . | Norway . | Russia . |
---|---|---|---|
Average share of income spent on foods | 11%1 | 12%2 | 52%3 60% spend almost all their money on food4 |
Income distribution5 | Gini-index 25 | Gini-index 26 | Gini-index 49 |
Recent change in diets | No dramatic change | No dramatic change | Reduced consumption of meat, fish, fruits and vegetables. Increased consumption of bread, potatoes, sugar. |
Agriculture | Large, industrialised farms Export oriented 25% of export income derives from agriculture | Small and geographically spread farms Oriented towards domestic market. Subsidised and protected by import restrictions. | Oriented towards domestic market. |
Food export/import | Export of pork and diary produce. Import of fruit and vegetables6 | Export of farmed fish. Import of grains, fruits, vegetables, sugar. Hardly any import of meat and dairy produce | Import of meat and other foods7 Food import is considerably outstripping the rate of food production |
Food control and legislation | One administrative body in charge of all inspection and control of food from stable to table. One Food Act regulates food control for both export and home market. Inspections carried out by 11 regional Food Control Authorities8 Member of EU. | Responsibility for food control and safety shared by three ministries. Harmonisation with EU regulations.9 | Responsibility for food control and safety divided between five ministries and administrative bodies. Currently unclear delimitations between responsibilities of bodies10 |
Histories of food safety | Public debates about food additives and pesticide residuals in fruit and vegetables during 1970'ies. | Public debate about additives, pesticides, BSE, food and mouth disease. | Public debates about harmful ingredients in food, stale and fake foods and infringement of sanitary requirements. |
Salmonella major problem in Danish eggs, poultry and pork in the 1990'ies.11 | Media and consumers more concerned about high food prices than food safety matters. | False caviar holds 20–40% of market, false tea 43%, false instant coffee 34%.14 | |
Ambitious plans against salmonella decreased number of notified cases of infections on humans from 5000 to 2000 (estimated real number: 50.000 to 20.000) | No cases of BSE12, hardly any Salmonella13 Strict measures against foot and mouth disease when outbreaks abroad (no incidences in Norway since 1952). | No reports of BSE or Foot-and-mouth disease in Russian media. | |
Since mid 1980'ies infections with campylobactor increased from less than 1000 to 4000 notified cases (estimated real number: 10.000 to 40.000). | Scrapie scandal in late 1990'ies. National surveillance and control program was initiated in 1997. Scrapie in sheep decreased from 31 infected herds in 1996 to 3 in 1999. | ||
Few cases of BSE. Strict plan of action exceeded measures agreed upon in EU. | |||
Consumer orientations | Preferences for natural foods, foods without additives GMO or chemicals. | Preferences for natural foods, foods without additives GMO or chemicals | Preferences for natural foods, foods without additives GMO or chemicals |
Consumption of organic foods seen as way to minimize food risk.15 | Preferences for foods of Norwegian origin, which are considered safer. | Preferences for locally produced foods, especially vegetables from own allotments, which are considered ecological and clean. | |
Wide variety of organic products available in major supermarket chains. | Only 1–2% of agriculture devoted to organic farming.17 | No market for certified organic foods. | |
Consumption of organic produce increased by factor 10 from 1993 to 2000. 29% of Danish consumers claim always to buy organic dairy products, 18% organic vegetables and 9% always organic meat.16 | Only in the last two years increase in numbers of consumers preferring ecologically cultivated foods (from 13% in 1999 to 20% in 2001)18 | Substantial self-provision of foods. Two thirds of Russians provide themselves with potatoes, one third with milk and meat. | |
Due to small and geographically spread farms many Norwegians think of their agriculture as almost organic. | Share of private farms, which is associated with organic methods, increased from 24% to 57% in total production during last decade.19 |
. | Denmark . | Norway . | Russia . |
---|---|---|---|
Average share of income spent on foods | 11%1 | 12%2 | 52%3 60% spend almost all their money on food4 |
Income distribution5 | Gini-index 25 | Gini-index 26 | Gini-index 49 |
Recent change in diets | No dramatic change | No dramatic change | Reduced consumption of meat, fish, fruits and vegetables. Increased consumption of bread, potatoes, sugar. |
Agriculture | Large, industrialised farms Export oriented 25% of export income derives from agriculture | Small and geographically spread farms Oriented towards domestic market. Subsidised and protected by import restrictions. | Oriented towards domestic market. |
Food export/import | Export of pork and diary produce. Import of fruit and vegetables6 | Export of farmed fish. Import of grains, fruits, vegetables, sugar. Hardly any import of meat and dairy produce | Import of meat and other foods7 Food import is considerably outstripping the rate of food production |
Food control and legislation | One administrative body in charge of all inspection and control of food from stable to table. One Food Act regulates food control for both export and home market. Inspections carried out by 11 regional Food Control Authorities8 Member of EU. | Responsibility for food control and safety shared by three ministries. Harmonisation with EU regulations.9 | Responsibility for food control and safety divided between five ministries and administrative bodies. Currently unclear delimitations between responsibilities of bodies10 |
Histories of food safety | Public debates about food additives and pesticide residuals in fruit and vegetables during 1970'ies. | Public debate about additives, pesticides, BSE, food and mouth disease. | Public debates about harmful ingredients in food, stale and fake foods and infringement of sanitary requirements. |
Salmonella major problem in Danish eggs, poultry and pork in the 1990'ies.11 | Media and consumers more concerned about high food prices than food safety matters. | False caviar holds 20–40% of market, false tea 43%, false instant coffee 34%.14 | |
Ambitious plans against salmonella decreased number of notified cases of infections on humans from 5000 to 2000 (estimated real number: 50.000 to 20.000) | No cases of BSE12, hardly any Salmonella13 Strict measures against foot and mouth disease when outbreaks abroad (no incidences in Norway since 1952). | No reports of BSE or Foot-and-mouth disease in Russian media. | |
Since mid 1980'ies infections with campylobactor increased from less than 1000 to 4000 notified cases (estimated real number: 10.000 to 40.000). | Scrapie scandal in late 1990'ies. National surveillance and control program was initiated in 1997. Scrapie in sheep decreased from 31 infected herds in 1996 to 3 in 1999. | ||
Few cases of BSE. Strict plan of action exceeded measures agreed upon in EU. | |||
Consumer orientations | Preferences for natural foods, foods without additives GMO or chemicals. | Preferences for natural foods, foods without additives GMO or chemicals | Preferences for natural foods, foods without additives GMO or chemicals |
Consumption of organic foods seen as way to minimize food risk.15 | Preferences for foods of Norwegian origin, which are considered safer. | Preferences for locally produced foods, especially vegetables from own allotments, which are considered ecological and clean. | |
Wide variety of organic products available in major supermarket chains. | Only 1–2% of agriculture devoted to organic farming.17 | No market for certified organic foods. | |
Consumption of organic produce increased by factor 10 from 1993 to 2000. 29% of Danish consumers claim always to buy organic dairy products, 18% organic vegetables and 9% always organic meat.16 | Only in the last two years increase in numbers of consumers preferring ecologically cultivated foods (from 13% in 1999 to 20% in 2001)18 | Substantial self-provision of foods. Two thirds of Russians provide themselves with potatoes, one third with milk and meat. | |
Due to small and geographically spread farms many Norwegians think of their agriculture as almost organic. | Share of private farms, which is associated with organic methods, increased from 24% to 57% in total production during last decade.19 |
1Danmarks Statistik 2003.
2Statistical Central Bureau Norway, SSB 2000.
3Goscomstat 2000.
4Comcom SPb 2001.
5Gini-index: 0=perfect equality, 100=1% has everything, 99% have nothing.
6Danmarks Statistik 2002, p. 378.
7Arguments and Facts, 37, 2002/Apгyмeнты и фaкты № 37, 2002; Economy of Russian Agriculture 1, 2000/Экoнoмикa ceльcкoгo xoзяйcтвa Poccии № 1, 2002.
8http://www.fdir.dk
9SNT 2001.
10Demand No. 3, 2001/Cпpoc № 3, 2001/http://www.aris.ru
11Bro et al. (1995).
12Scientific Steering Committee 2000.
13Liven (1999).
14Arguments and facts 38, 2000/Apгyмeнты и фaкты № 38, 2000.
15Halkier (2000), Holm (1999).
16Danmarks Statistik (2002a).
17Torjusen et al. (1999).
18Berg (2001b).
19http://www.aris.ru; Arguments and Fact, 2001, no. 18/Apгyмeнты и фaкты № 18, 2001.
Firstly, no big differences appear in living standards in terms of share of income spent on foods between Denmark and Norway, whereas in Russia economic restraints have more impact on eating habits. Whereas food is abundant in the Scandinavian countries and dietary intake excessive, a substantial part of the Russian population suffer from poverty and economic inequality. This have resulted in inadequate eating patterns with widespread consumption of low quality and fake foods, as well as insufficient intake of proteins and calories. Secondly, Russia is a food import-oriented country while food export volume in Norway is somewhat larger and in Denmark much larger. Thirdly, the structure of food safety control appears more simple and transparent in the Scandinavian countries and rather intricate and uncoordinated in Russia. Fourthly, food safety issues are widely debated in all three countries albeit there are different focuses in each of them. Finally, despite our consumers’ similarity with respect to the preference for natural products, there are more similarities between Norwegians and Russians who perceive their own domestic agriculture as almost organic, whereas Danish consumers consider choosing organic foods instead of domestic conventional foods as a means to avoid food-related risks.
As expected, the Russian food sector differs much from those of the Scandinavian countries. However, in spite of many similarities between living conditions, political systems and social structures there are marked differences in the food safety situation in Denmark and Norway. While Danish consumers have to deal with a comprehensive problem with salmonella and campylobactors, most Norwegian consumers still take food safety for granted (Berg 2000).
4 Methods
The quantitative analysis to be presented is composed by three independent national surveys. The Danish and Norwegian surveys share a core of common questions, while the Russian survey is a smaller version of the Norwegian survey (comparisons presented in Berg (2001)), The variables that are included in the following analyses have been used and tried out in previous surveys (Berg 2000; Kjærnes et al. 2001).
As demonstrated in Table 1, our three selected countries have similarities as well as supposedly different conditions for food safety. Since Russia is more like a continent than a homogenous small country like Denmark and Norway, we decided to concentrate the Russian material to the St. Petersburg area, which has about 4.5 million residents (Norway about 4 million and Denmark 5 million). In each country, 1000 consumers aged 18–80 were interviewed. The material was collected during the months of April and May, 2001, parallel to the outbreak of the foot and mouth disease, with heavy import restrictions on meat and milk products in all three countries. Common collection procedures, by CATI–telephone interviews, have been followed in all countries.1
The phone numbers were picked randomly in each country. The response rates were 37 in Russia, 55 in Denmark and 56 in Norway. In order to promote comparability, the materials have undergone weighting procedures, and the materials are now considered representative by age and gender in all three countries/region. We notice, however, that while Denmark and Norway have a 50:50% distribution between male and female citizens in the age group 18–80 years, the distribution in Russia is 43:57, due to a higher death rate among younger men in Russia than in Denmark and Norway. This means that female relations to food safety matters have a larger weight on the total scores in Russia than in Denmark and Norway, and we know that female consumers tend to be more critical in food issues than male consumers (Wandel 1994; Berg 2000). These gender distinctions also affect the age distributions. In the multivariate analyses, however, gender and age are included among the independent variables, so that effects from these variables are controlled.
5 Empirical results
Initially we shall present results that document the differences between our selected countries, first by measures on direct trust in specific foods and second by measures on general trust in food system institutions (market mechanism and public authority food control). Multivariate analyses then allow us to compare the different patterns that emerge within each country. Finally we try to grasp the situation in each country by combining the results from these different steps of the analyses in one illustrative figure.
5.1 Direct trust in foods
To feel and expect that the food you eat is sufficient, safe and healthy is basic for quality of life itself. Are there differences in how consumers in Russia, Denmark and Norway trust the food they eat in their everyday lives?
‘Trust in food safety’ is a comprehensive concept, and it is far from obvious what is the most valid operationalisation. Since trust in food not only vary between consumers, but also differ from food to food for the same person (Halkier, 2001b), we found it convenient to ask about several different specific foods, in order to get a more differentiated picture. The consumers were asked whether they found six different foods ‘rather safe’ or ‘not quite safe’ to eat.
Consumers evaluation of food safety in Russia, Denmark and Norway. Based on the question: ‘For each of the following foods, please tell me if you think that eating them is rather safe or not quite safe?’. For significance levels see footnote 7 (N=1001, 980, 1001).
Consumers evaluation of food safety in Russia, Denmark and Norway. Based on the question: ‘For each of the following foods, please tell me if you think that eating them is rather safe or not quite safe?’. For significance levels see footnote 7 (N=1001, 980, 1001).
The main impression from Figure 2 is that consumer trust in foods is high in Norway; medium in Denmark; and low – on three of the selected foods – in Russia. As mentioned, the percentage of women is higher in Russia than in Denmark and Norway, and this could affect the total scores. The lower Russian trust, however, still persists if we split on men and women (data not shown).
Figure 2 also reveals another distinction: in Russia, the trust pattern is very differentiated; some foods are trusted, while others are not, while the Danish and Norwegian trust patterns are more even. In Russia as many as 83% thought that fruits and vegetables were rather safe to eat, while only 35% thought that fish was safe (percent difference 48). Danish consumers’ evaluation, on the other hand, only varies from 77% on fish to 66% on beef and eggs (percent difference 11). The Norwegian consumers’ evaluation varies somewhat more, but still only from 92% on fish to 70% on chicken (percent difference 22).
Some of these differences can be explained by different national food system conditions: fish shows the biggest discrepancy; from almost everybody in Norway trusting fish (92%) to only one out of three (35%) in Russia. Norway is a large producer and exporter of fish, and fresh fish is widely available, whereas the fish in St. Petersburg is mostly imported, frozen and relatively expensive. The variations in trust in fish is reasonable, given that in Norway, food safety crises have been few, and in Russia there have been problems with, e.g., stale products and sanitary requirements, both important features in relation to fish. In addition Russian consumers are concerned and worried about product origin, processing technology and storing conditions for fish, as well as water pollution. In St. Petersburg fish remains one of the worst products by microbiological parameters. By experts’ opinion, in absolutely all the city's stores temperature norms are violated (Petersburg Quality/Пeтepбypгcкoe кaхecтвo №. 11, 12; 2002).
It is tempting to question whether the difference between Denmark and Norway has something to do with their predominant food industry: fish in Norway and meat in Denmark. While fish in Norway is considered indisputable healthy, there has been a media debate in Denmark focusing on alarming high residues of dioxins in fish. Lassen (1996) claims that people in Denmark are normally quite reluctant to eat fish. However, nutrition experts in all three countries recommend people to eat fish for health reasons and our results show that people in both Denmark and Norway still more often think that fish is healthy to eat, than beef and pork.2
The figure also shows other similarities among the countries in consumers’ trust in specific foods – such as chicken – which cannot be immediately explained by similarities in performance of the national food systems institutions or their histories of food safety. Danish and Norwegian consumers’ trust in chicken is expressed at the same level, although the food safety histories in Denmark are almost dominated by salmonella in chicken and has played next to nothing of a role in the Norwegian food safety histories. Other more socio-cultural or normative differences in how eating chicken is interpreted could be at play, such as the fact that in Norway chicken is associated with modern ready-made fast food, as well as with industrialized large scale meat production with low animal welfare (Berg 2002). The Russian results, which show higher trust in chicken than in meat and fish, is logical: St. Petersburg poultry industry is the leader among Russian regions, and the local product is considered as tasty, fresh and of high quality. In addition, eating chicken in Russia is traditionally associated with dietetic nutrition (Peasants’ Lists/Кpecтьянcкиe вeдoмocти № 11–12, 2002; Peasants’ Lists/Кpecтьянcкиe вeдoмocти №13–14, 2002; Expert. Northwest/Экcпepт. Ceвepo-зaпaд № 14, 2002).
If we focus on the level of distrust, Figure 2 shows that Danish consumers express more scepticism than both Russian and Norwegian consumers on fruits and vegetables (24%), chicken (28%) and eggs (31%), while Russian consumers are the most sceptical of pork (31%), beef (39%) and fish (48%). The Norwegian results show the lowest distrust on all, except one food: chicken. One out of four Norwegian consumers does not think it is quite safe to eat chicken.
In order to get a single general measurement on consumers’ trust in food, the responses to the six foods (as presented in Figure 2) are merged into one food safety index.3 On this index reaching from 0 (no trust) to 12 (high trust), the average Russian consumer expresses a trusting level of 7.8, the average Danish consumer scores 8.7, while the average Norwegian consumer scores 10.3.
5.2 Trust in food system institutions
As illustrated in our introductory model presented in Figure 1, we expect individual consumers trust in food to be affected by the qualities of their surrounding food systems. If the food market and the food control function satisfactory and food safety is well taken care of, it is likely that consumer trust in food is high. Our survey material does not contain measures on how well or bad the different national food system institutions actually perform, only measures based on how consumers trust or evaluate the performance of their food system institutions. We distinguish between whether the consumers do trust – or distrust – the food market on the one hand, and public food control on the other hand.
Before investigating whether direct trust in foods is affected by trust in food system institutions, we would like to present how consumers in our three countries evaluate their food system by expressed trust in food market and public food control respectively. Based on the national variations in direct trust in foods as presented in Figure 2, we would expect that Norwegian consumers also show the highest trust in their food system institutions, while the Russian consumers show the lowest trust in their food system institutions, with Danish consumers in between.
Table 2 indicates that our expectations were wrong: the tendency to trust food system institutions seems to be higher in both Russia and Norway than in Denmark. While 54% of the Norwegian consumers and 40% of the Russian consumers fully agree that ‘you can rely on the food control authorities that the food is safe to eat’, only 18% of the Danish consumers fully agree with this statement. There are also fewer Danish than Norwegian and Russian consumers that fully agree to the statement: ‘The market mechanisms secure that the quality of the food is good enough’.
. | Russia . | Denmark . | Norway . |
---|---|---|---|
Trust in food control | 40 | 18 | 54 |
Trust in market mechanisms | 28 | 16 | 28 |
. | Russia . | Denmark . | Norway . |
---|---|---|---|
Trust in food control | 40 | 18 | 54 |
Trust in market mechanisms | 28 | 16 | 28 |
Based on positive response to the statements: ‘You can rely on the food control authorities that the food is fit to eat’ and ‘The market mechanisms secure that the quality of the food is good enough’. Percentages (N=979, 938, 948).
5.3 What affects consumers’ trust in food safety within the countries?
Our results show that trust in food safety (Figure 2) varies considerably, while evaluations of the food safety conditions (Table 2) do not vary, as expected, from country to country. What patterns do these preliminary results hint at?
These results could indicate that the Danish trust in food is less dependent on trust in the food market and the public authority food control than the Russian and Norwegian trust. In other words, these results could indicate that Danish consumers’ relatively high trust in food depends on other factors than food control and a trustworthy food market. For example, people could feel safe because they shop alternative foods through alternative channels. If this explanation is supportable, we will find lower effects in the coming multivariate analyses from trust in food system institutions on trust in food safety among the Danish consumers.
Another possibility could be that it is the Russians, and not the Danish consumers, that are breaking the pattern. Historically, the functioning and division of responsibilities between state and market have been differently arranged in Russia than in our two Scandinavian countries. Since the dissolution/disorganisation of the Soviet state in 1991, there have been great alterations in the Russian society with appurtenant changes in the food system. This could very well affect how citizens evaluate such institutions. The Russians, compared to the Scandinavians, might be too generous with their trust in the market mechanisms and in the public food control system, when considering their experiences with food safety matters. If this is the case, the coming multivariate analysis will show more similar patterns in Denmark and Norway compared to Russia.
A third possibility is that neither of these explanations hold, because there are no significant correlations between the feeling of food safety and the evaluation of the institutions in the food systems in any of the countries. If this is true, our presupposition – trust in food is affected by trust in food system institutions – simply failed.
Table 3 shows the results from three multivariate analyses, where we can compare how trust in food system institutions affect how consumers trust the food they eat in our three selected countries. Alongside the independent variables for the national food system institutions, we have included three individual social background variables that feature strongly as creating patterns of differences in consumers’ food practices in existing sociology of consumption research – gender, age and education (Wandel 1995; Warde 1997; O'Doherty et al. 1999; Berg 2000). The dependent variable is the additive trust in food index constructed by adding the responses to the six ordinal variables measuring trust in fruits and vegetables, chicken, eggs, pork, beef and fish, as shown in Figure 2. In linear regression analyses it is recommended to use continuous variables on interval or ratio level; however, linear regression has proven to be rather robust and Labovitz (1967) also justifies the use of continuous variables measured on ordinal level, as frequently seen in sociological analyses. Our dependent trust in food index goes from 0 (no trust) to 12 (high trust).
. | Russia r2=0.08 . | Denmark r2=0.12 . | Norway r2=0.12 . |
---|---|---|---|
. | β (B) . | β (B) . | β (B) . |
Female=1 | −0.18** (−1.10) | −0.10** (−0.68) | −0.08** (−0.40) |
Age 18–80 | 0.08** (0.02) | 0.11** (0.02) | −0.08** (−0.01) |
Higher education=1 | −0.04 (−0.22) | −0.03 (−0.18) | −0.01 (−0.04) |
Trust in food control (1–3) | 0.11** (0.42) | 0.25** (1.22) | 0.27** (1.00) |
Trust in market forces(1–3) | 0.17** (0.69) | 0.14** (0.68) | 0.12** (0.38) |
(Constant) | ** (5.50) | ** (4.69) | ** (7.87) |
. | Russia r2=0.08 . | Denmark r2=0.12 . | Norway r2=0.12 . |
---|---|---|---|
. | β (B) . | β (B) . | β (B) . |
Female=1 | −0.18** (−1.10) | −0.10** (−0.68) | −0.08** (−0.40) |
Age 18–80 | 0.08** (0.02) | 0.11** (0.02) | −0.08** (−0.01) |
Higher education=1 | −0.04 (−0.22) | −0.03 (−0.18) | −0.01 (−0.04) |
Trust in food control (1–3) | 0.11** (0.42) | 0.25** (1.22) | 0.27** (1.00) |
Trust in market forces(1–3) | 0.17** (0.69) | 0.14** (0.68) | 0.12** (0.38) |
(Constant) | ** (5.50) | ** (4.69) | ** (7.87) |
**P<0.001.
Linear regression. Standardized (β) and un-standardized (B) regression coefficients (N=979, 948, 938).
Comparative data material allows us to investigate whether the same mechanisms seem to be in operation in several countries in spite of different national conditions. The factors that influence consumers’ trust in food show similar, but not identical, patterns in the three countries. We will first comment on how gender, age and education seem to affect trust in food in the three countries, and finally discuss how the results support or reject our proposed hypotheses related to what institutions consumers tend to base their trust in food.
We find support in all three countries for the assumption that female consumers are more sceptical than male consumers. When age and educational level, as well as consumers’ evaluation of the food system institutions are held constant, the un-standardised coefficients (B) show that female consumers in Russia score 1.10 lower on the trust in food-index than male consumers in Russia. Gender differences appear to be bigger in Russia than in Denmark, where the differences appear to be bigger than in Norway. Since we know that average trust increases in reverse order, we can hypothesis that the more scepticism towards food safety in a country, the bigger the gender differences in trust. If this is true, it follows that it is the female consumers who react the most when food safety is threatened.
Age shows contradictory results: while ageing is positive – measured by trust – in Russia and Denmark, it is negative in Norway. This could imply that different mechanisms pulling in opposite directions can be attached to age (age-related concerns and age-related vulnerability): On the one hand, while younger people often have important age-related concerns other than food safety on their minds (mating, music, education, childbirth, etc.), older people have more time for pragmatic issues like health and food safety concerns. As years go by we might therefore develop a more sceptical attitude towards food safety. This is reflected in the Norwegian results, where older people show more distrust towards food safety than younger people. On the other hand, when we find more scepticism among the youngest consumers in Russia and Denmark, this can be the result of another activated age related mechanism: if food safety is considered seriously threatened, younger people could be more vulnerable and more flexible in their conceptions. The reason why ageing shows different effect in Norway could be that Norwegians during last years have been less exposed to serious national collective food events than Danes and Russians, and that the age related vulnerability mechanism has not been activated.4
According to Table 3, higher education does not affect consumer trust in food in any of our countries. This could be a result of two or more mechanisms pulling in opposite directions. Other studies show contradictory results on how educational level affect food safety issues (Sellerberg 1990; Schafer et al. 1993; Wandel 1994; Berg 2000).
5.4 Food system institutions’ impact on consumers trust in food safety
We shall now turn to the main concern of this article, first: Is it relevant to include societal institutions in analyses of consumer trust in food safety? As expected, we find positive significant effects from trust in public authority control as well as trust in market mechanisms on consumer trust in food safety in all countries. The estimated β-coefficients also indicate that such context evaluative variables are even more important to consider than individual characteristics like gender, age and educational level.5
How the food safety systems works in Russia, Denmark and Norway. Proportions of consumers in Russia, Denmark and Norway who report they trust their food control and the market mechanisms (left axis), and average trust (0–12) in food based on the six foods presented in Figure 2 (right axis), as well as arrows showing correlations between institutional trust on the one hand and ‘trust in food-level’ on the other.8 Percentages, mean, Pearsons correlation (N=976, 938, 948).
How the food safety systems works in Russia, Denmark and Norway. Proportions of consumers in Russia, Denmark and Norway who report they trust their food control and the market mechanisms (left axis), and average trust (0–12) in food based on the six foods presented in Figure 2 (right axis), as well as arrows showing correlations between institutional trust on the one hand and ‘trust in food-level’ on the other.8 Percentages, mean, Pearsons correlation (N=976, 938, 948).
Figure 3 illustrates the main findings: it is the Norwegian consumers who most frequently say they trust their food control, and accordingly they also show the highest average trust in food safety. But even though it is the Danes who express the lowest trust in their food control and their food market, it is the Russian consumers who show the lowest average trust in foods.
Due to very low trust in the food system institutions in Denmark relative to the Danish consumers’ reported trust in food, our first expectation was that effects from these institutions should be weaker in Denmark than in Russia and Norway. Our analyses do not confirm this. According to our results it is the Russian consumers – not the Danish – who show the lowest correspondence between trust in food system institutions on the one hand and food safety considerations on the other. Trust in food in Denmark is not less dependent on trust in food safety control and market mechanisms than in Russia and Norway. Or, in Denmark disadvantageous food-safety events have resulted in more articulated and conscious scepticism towards the food system institutions than the case is in Russia.
Still, in all three countries there are significant correlations between conceptions of food safety on the one hand and trust in the public food control and the food market on the other. But in Denmark and Norway trust in food safety is stronger influenced by trust in the food control than trust in the food market, while in Russia we find the very opposite pattern.6 We think that the national patterns illustrated in Figure 3 reflect that the Russian food system is different from the food systems in Denmark and Norway, which seem to have more common features.
In other words, there is reason to believe that trusting the public food control is more decisive for consumers trust in food in Denmark and Norway, while opinions on whether the power of the market forces promote food safety are the strongest predictor on whether consumers in Russia feel that the food is safe to eat or not. We could mention that in England and Belgium, as in Russia, trust in market mechanisms seem to be more decisive than trust in public food control for direct trust in food (Berg 2000).
6 Discussion
In this article we have argued that institutional conditions should be taken into consideration when consumers trust in food safety is analysed. We find empirical evidence in all countries that consumers trust in food safety is related to their evaluation of how their national food control authorities perform, as well as to what extend they trust market mechanisms to secure food quality. Trust in food safety in the Scandinavian countries is more likely to rest on trust in public food control. Russians do not show more trust in market than in food control (40% of the Russians say they trust food control, while 28% say they trust the market mechanisms). However, the impact from trusting the market on direct trust in food is stronger than the impact from trusting the food control, while in Denmark and Norway it is the food control that gives the best prediction on whether you trust the food or not. Russian consumers who trust the market mechanisms more often than other Russians also feel safe about the food they eat, while differences in trust in food between those who trust in market mechanisms and those who do not in Denmark and Norway is much lower. How can these results be understood?
As shown in Table 1, income distribution as measured by the Gini-index, as well as the average share of income spent on foods, signal much larger differences in living conditions in Russia than in the Scandinavian countries. For a large part of the Russian population, food is very expensive, and for many families economic deprivation has resulted in decreased food quality and dietary problems. We believe that the Russian pattern also reflects a very differentiated food market measured by food safety and quality, and that availability of safe and high quality foods is strongly connected to individual economic resources. It is rather obvious that rich people who can afford to buy safe foods also have more personal reasons to say they trust the market mechanisms, while the poor have less reason to trust the market mechanisms as well as the food they can afford to buy.
During last ten years, the Russian diet has worsened, and at the time of our data collection 40% of the population suffers from protein calorie insufficiency. In Denmark and Norway food supply is abundant and food prices do not form a significant obstacle for a healthy diet. While many Russians cannot afford to give priority to eat safe and quality foods, the low food prices in Scandinavia relative to other expenditures, makes food safety more of an optional thing. In both Denmark and Norway most worried or sceptical consumers can desist from foods considered to be unsafe or unhealthy, and give priority to buy alternative, perhaps more expensive, but safer foods, on the expenses of some other consumption goods. In Denmark, with more scepticism towards food safety, consumption of organic foods is seen as a way to minimise food risks (Holm and Kildevang 1996; Halkier 1999, 2001a; Holm 1999).
But why do Danish consumers show lower trust in food safety than Norwegian consumers? In both countries there is a rather strong correlation between trust in food and trust in public food control. In other words, both Danish and Norwegian consumers seem to consider food safety as a central public responsibility. But the results also show that in Denmark more consumers have lost their trust: only 18% trust their food control, compared to 54% in Norway.
Table 1 gives examples on how the national food systems differ. In Norway the food sold on the home market is largely of domestic origin and contrary to the fishing industry, the agricultural sector in Norway is solely domestically oriented. Moreover the agriculture, which consists of many small-scale farms, is protected by import restrictions and subsidies. Consequently, consumers in Norway are used to seeing their foods as local products, which are exclusively for Norwegian consumes and which are protected by the state.
In Denmark food is first and foremost associated with export and Danish food production has a long history of industrial large-scale production. Correspondingly a centralised food control system has been established which controls both foods that are exported and the food sold on the Danish home market. Accordingly consumers in Denmark are used to being exposed to a centralised food system, which in many ways is very effective, at the same time as it is vulnerable. When they appear, food safety problems most often do so as widespread national problems and very rarely just as local incidents.
Recent histories of food safety differ between the three countries in ways that correspond with the differences in food safety levels. Russian consumers were faced with numerous facts related to falsifications and poor quality of imported foods and Russian media report widely about harmful ingredients in foods and infringements of sanitary requirements. Danish consumers must handle widespread and persistent problems with salmonella and other bacterial infections and there is a public debate about this and other food safety problems in the Danish food production sector (Halkier 2001a). In Norway on the other hand, no major food safety scandals affecting human health have occurred during recent years. Rather food safety problems are considered as problems abroad which are not relevant for Norwegian consumers.
In Norway it seems as if consumers base their trust in foods on a rather strong belief in the public authorities’ ability and capability to control the food market and thereby secure safe foods for everyone. In Russia it is the individual consumer's economic ability that decides whether or not one can take advantageous of the supply of safe quality foods on the market. It is the newly released market forces, not the food control, that make the economic elite capable of finding and buying safe quality foods. In Denmark, consumer trust has been heavily contested by different food scandals, and consumers seem to respond to this with a loss in trust in their public food control.
The preceding analysis suggest that system trust, i.e. trust in anonymous societal institutions such as the market or the state is important when accounting for national differences in levels of consumers trust in food. It is reason to believe that also relational trust, i.e. trust in direct personal contacts, may be of varying importance in the three countries and should be studied further in forthcoming research. In Denmark there are signs – like increased interests for farmers’ markets – indicating a shift towards more relational trust. Also in Norway alternative food channels appear, but still to a far lesser extent than in Denmark. The widespread home-produce in Russia suggests that informal networks play an important part in Russian food provisioning. Strategies of survival during all transformation years have been based more on non-formal structures of mutual trust than on formal contracts. However, modern Russian society is characterized by the limitation of trust within private social networks (Eremicheva and Simpura 1999: 145–59).
The results from the surveys that were presented here show that trust in food is not to be taken for granted. However, the results do not indicate that basic trust is an irrelevant type of trust when analysing consumers’ trust in food. For various foods the majority of consumers in Norway, Denmark and St. Petersburg expressed confidence that the foods in question were rather safe to eat. Consequently consumers need not reflect about safety when buying these foods. This suggests that large parts of food consumption in all three countries may be seen as matters of routine or habit. These routines may in turn be based on tacit trust, where the safety of foods is taken for granted, and perhaps not even questioned. Or, on the other hand, the routines may be established on the basis of conscious deliberations about personal food practices, i.e. reflexive trust.
The material presented here, indicates that food safety is a recurrent theme in public debate and media in Norway, Denmark and Russia. This suggests that, even though the discursive framing of food and food safety differs between the three countries, trust in food is likely to become an increasingly reflexive type of trust. Shopping, preparing and eating food is no longer only a matter of tradition and consumers direct experience, but is also a matter of mediated experience (Thompson 1995), i.e. knowing what to do on the basis of expert advice and information systems. Public debates about the efficiency and safety of food production and food control systems are part of this kind of knowledge. This means that when accounting for differences in consumers’ relations and reactions to food, individual socio-demographic background factors are not sufficient. It is necessary to study how the societal institutional framing of food issues influence consumers trust in food.
Footnotes
Thanks to the poll institutes: COMCON-SPb (St. Petersburg), Opinion (Oslo) and Gallup (Copenhagen).
Instead of showing significance estimates for all the results presented in Figure 2, please note that Margins of error for sample size 1000=1.4 (5/95 distribution) −3.2 (50/50 distribution).
It should be mentioned, however, that Norwegians do not appreciate farmed fish as much as wild fish. Almost one out of three thinks that farmed fish is not quite safe to eat (Berg 2001).
The index is constructed by adding the values on the six foods presented in Figure 2, where ‘trust’ gives 2, ‘uncertain’ gives 1 and ‘distrust’ gives 0.
Separate analyses of the Norwegian material, which contains more variables than this comparable material, show that this situation is about to change also in Norway: in 2001 younger people were more sceptical to eat meat than older people, while the opposite was true in 1999.
Only being a woman in Russia showed greater effect on trust in food.
Using the β-coefficients from Table 3 would give same interpretation.
This distinction is even more explicit in the multivariate analyses results.
References
Lisbet Berg, Dr. Polit. in Sociology, University of Oslo 1997. Head of Research at National Institute for Consumer Research (SIFO), Oslo.
Unni Kjaernes, Cand.real in Nutrition, University of Oslo 1981. Senior Researcher at the National Institute for Coinsumer Research (SIFO), Oslo. Project coordinator for ‘Consumer Trust in Food’ (1992–94), funded by the European Commission.
Elena Ganskau, PhD. in Sociology, St. Petersburg State University, 2001. Researcher at the St. Petersburg State University.
Vera Minina, PhD in Economics, St. Petersburg State University 1984. Grand PhD in Sociology, St. Petersburg State University 1999. Professor, Faculty of Sociology, St. Petersburg State University. Editor-in-chief of Russian journal Personal-Mix.
Ludmila Voltchkova, Professor, Head of the Chair of Public Administration and Social Planning. PhD in Economics 1983. Doctor in Sociology 1994.
Bente Halkier, PhD. Sociology, Roskilde University 1998. Associate Professor at the Department of Communication, Journalism and Computer Science, Roskilde University Denmark.
Lotte Holm, M.Sc. and PhD. in Sociology, University of Roskilde 1991. Associate Professor, Department of Human Nutrition.