Organic consumption and diet choice: An analysis based on food purchase data in France

Organic consumption and diet choice: An analysis based on food purchase data in France

Accepted Manuscript Organic consumption and diet choice: An analysis based on food purchase data in France Christine Boizot-Szantai, Oualid Hamza, Lou...

893KB Sizes 1 Downloads 50 Views

Accepted Manuscript Organic consumption and diet choice: An analysis based on food purchase data in France Christine Boizot-Szantai, Oualid Hamza, Louis-Georges Soler PII:

S0195-6663(17)30595-0

DOI:

10.1016/j.appet.2017.06.003

Reference:

APPET 3508

To appear in:

Appetite

Received Date: 20 April 2017 Revised Date:

29 May 2017

Accepted Date: 4 June 2017

Please cite this article as: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

INRA-Aliss UR 1303, France

Oualid Hamza ([email protected])

M AN U

Christine Boizot-Szantai ([email protected])

TE D

Louis-Georges Soler ([email protected])

Corresponding author :

Louis-Georges Soler ([email protected])

EP

INRA-Aliss UR 1303, 94205 Ivry sur Seine, France

AC C

SC

Christine Boizot-Szantai1, Oualid Hamza1, Louis-Georges Soler1

RI PT

Organic consumption and diet choice: an analysis based on food purchase data in France

ACCEPTED MANUSCRIPT

1

Organic consumption and diet choice: an analysis based on food purchase data in France

2 3

RI PT

4 5 1. Introduction

7

The food consumption patterns observed in developed countries raise important concerns related to

8

health and environmental issues. In this context, many studies have aimed to assess the potential

9

benefits of organic consumption and to determine the extent to which the promotion of organic foods

10

could be a relevant strategy to address these issues. Crinnion (2010) and Lairon (2010) reviewed multiple

11

studies and showed that organic varieties provide significantly greater levels of vitamin C, iron,

12

magnesium, phosphorus, and antioxidant phytochemicals than did non-organic varieties of the same

13

foods. Whereas Smith-Spangler et al. (2012) considered that the published literature lacks strong

14

evidence that organic foods are significantly more nutritious than conventional foods, Forman and

15

Silverstein (2012) concluded that in terms of health advantages, organic diets have been convincingly

16

demonstrated to expose consumers to fewer pesticides associated with human disease. Regarding

17

environmental issues, organic farming has been demonstrated to have less of an environmental impact

18

than conventional approaches (Heerwargen et al., 2014). This point, however, is still controversial

19

because Life-Cycle Analysis (LCA) studies comparing agricultural products from conventional and organic

20

farming systems have reported a wide variation in the resource efficiency of products from these

21

systems. Some studies have shown that the impacts per area of farmed land are usually less in organic

22

systems but the impacts per produced quantity are often higher (Meier et al., 2015). However, this result

23

is not solely due to the usually lower yields in organic systems but also to inaccurate modeling within

24

LCA. The consequence is that compared to conventional agriculture, organic production with its greater

25

land use requirements appears environmentally less favorable, although its energy use is markedly

26

reduced because of the reduction of the input use of fertilizer industry, and organic production increases

27

the value added of agriculture (Risku-Norja and Mäenpää, 2007). Although additional research is

28

required to clearly determine the differences in the biological and toxicological status of organic food

29

consumers and conventional food consumers (Dangour et al., 2010; Hoefkens et al., 2010; Jensen et al.,

30

2013), these potential benefits have justified many investigations aimed at better understanding the

AC C

EP

TE D

M AN U

SC

6

ACCEPTED MANUSCRIPT

31

motivations of organic consumers and characterizing the determinants and conditions of organic market

32

expansion.

33 On the one hand, studies have been conducted to identify both organic consumers and the main factors

35

that explain their selection of organic vs. non-organic foodstuffs (see Yiridoe et al. (2005) for a literature

36

review). Consumers’ attitudes towards and motivations for buying organic food are strongly linked to

37

beliefs about its healthiness, taste and environmental friendliness (Bryla, 2016; Chen and Lobo, 2012;

38

Costanigro et al., 2014; Dowd and Burke, 2013; Goetzke et al., 2014; Honkanen et al., 2006; Hughner et

39

al., 2007; Kihlberg and Risvik, 2007; Lee et al., 2013; Magnusson et al., 2003; Thøgersen et al., 2015).

40

Ethical motivations and animal welfare also affect the purchase intentions of regular organic consumers,

41

whereas food safety concerns influence those of occasional organic consumers (Pino et al., 2012; Zander

42

and Hamm, 2010). As shown by Aertsens et al. (2009), related attributes of organic food with values,

43

such as ‘security’, ‘hedonism’, ‘universalism’, and ‘benevolence’, positively influence attitudes towards

44

organic food consumption. Among the factors that undermine organic consumption, insufficient

45

knowledge and a lack of trust in organic labels seem to play a key role (Honkanen et al., 2006; Hughner

46

et al., 2007; Janssen and Hamm, 2012; Smed et al., 2013). Indeed, labels and information signaling

47

credence are important search attributes. Although credence cues cannot be accurately evaluated by

48

consumers, the expectations that they generate affect consumers’ perceived quality and sensory

49

experiences (Fernqvist and Ekelund, 2014). Psychological dimensions, such as cognitive and affective

50

attitudes, personality traits, personal norms, and perceived behavioral control, also influence the

51

intentions to purchase organic products (Aertsens et al., 2009; Chen, 2007; Lee and Yun, 2015; Scalco et

52

al., 2017; Schleenbecker and Hamm, 2013; Welsch and Kühling, 2009). Social norms influence organic

53

consumption because people are likely to be affected by what others think. Ruiz de Maya et al. (2011)

54

stress the role of reference groups and routine behavior. They show that economic and cognitive factors

55

are significant covariates of pro-environmental consumption and that the consumption patterns of

56

reference persons strongly affect the decision to consume organic food. Interestingly, overall, the same

57

drivers seem to explain the increase in organic consumption in emerging countries (Bruschi et al., 2015;

58

Chen and Lobo, 2012; Grannis et al., 2003; Marques Vieira et al., 2013).

AC C

EP

TE D

M AN U

SC

RI PT

34

59 60

On the other hand, a large amount of research has focused on the economic issues (such as the high

61

prices of organic foods) affecting consumers’ decisions and discourage repeat purchases of organic food

62

products (Avitia et al., 2015; Marian et al., 2014). On the basis of a marketing literature review,

ACCEPTED MANUSCRIPT

Hemmerling et al. (2015) show that the most investigated topics are cost to the consumer, consumer

64

value and benefits. Rodiger and Hamm (2015) review studies addressing the price elasticity of organic

65

food demand. Despite an impressive number of papers published over the last decade, robust

66

conclusions about the own-price elasticity of organic products and the cross-price elasticity of organic vs.

67

non-organic products remain elusive (Bernard and Bernard, 2009; Bezawada and Pauwels, 2013; Bunte

68

et al., 2010; Lopez and Lopez, 2009; Mosnier et al., 2009; Schrock, 2012). Concerning consumers’

69

willingness-to-pay (WTP) for organic foods, Rodiger and Hamm (2015) conclude that a large share of

70

consumers was willing to pay a higher price for organic food, although WTP differs by product categories

71

and consumer segments. Finally, concerning the influence of socio-demographic data on WTP,

72

contradictory results were obtained regarding the influence of family size, gender, income, and marital

73

status. However, most studies have found that education and age variables had a strong and positive

74

influence on WTP.

75

M AN U

SC

RI PT

63

When considering all of these studies, it is worth noting that most of the available literature has

77

addressed consumers’ trade-offs between organic and conventional products, focusing on the reasons

78

people buy organic foods and their WTP for health and environmental attributes. However, the literature

79

excludes two important questions.

80

TE D

76

First, no consumer purchases only organic foods. Consequently, it is important (i) to determine the place

82

of these foodstuffs within overall consumption patterns and (ii) to assess whether an increase in organic

83

consumption is associated with modified consumption patterns. If so, then it is impossible to understand

84

organic consumers’ motivations without considering their consumption patterns or the substitutions and

85

complementarity relationships between organic and non-organic products within the diet. Recent

86

studies have explored this issue; by using large surveys, they have shown that organic consumption is

87

often associated with modified consumption patterns (Kesse-Guyot et al., 2013; Petersen et al., 2013).

88

For instance, a small but statistically significant correlation between increasing organic budget shares

89

and decreasing meat budget shares was found. The first goal of our paper is to pursue this analysis by

90

using purchase data from a large representative sample of French households.

AC C

EP

81

91 92

Second, the prices of organic foods are generally higher than those of conventional foods because they

93

rely on stricter environmental specifications and more demanding production methods. However, if

94

higher levels of organic consumption are associated with modified consumption patterns, what are the

ACCEPTED MANUSCRIPT

consequences of such modifications on household food expenditures? When considering both modified

96

consumption patterns and price differentials between organic and non-organic foods, is an increase in

97

organic consumption associated with larger or smaller total food budgets? In other words, does the

98

modification of consumption patterns compensate for the increase in food budgets related to the

99

purchase of more expensive organic products? This second point is important because, depending on the

100

answer, research should focus not only on consumers’ trade-offs between organic and conventional

101

products (as the majority of existing studies do) but also on their willingness to change their

102

consumption patterns (Micheelsen et al., 2014). The second goal of our paper is to answer these

103

questions and better characterize the impacts of organic purchases on food budgets by accounting for

104

variations in the overall consumption patterns. This will help us distinguish between ‘quantity’ and

105

‘quality’ effects, namely, the differences in the food budgets of non-organic and regular organic

106

consumers caused by variations in the quantities of the different food groups purchased and the

107

differences in the food budgets of non-organic and regular organic consumers caused by variations in the

108

prices of the products purchased in the different food groups. This analysis will be able to provide

109

interesting insights on organic consumers’ preferences, not only about the trade-offs between organic

110

and conventional foods, but also about their choices of dietary patterns.

M AN U

SC

RI PT

95

111

Three important results follow. First, organic consumption seems to be one element in a wider ‘quality

113

strategy’ of consumers because these consumers purchase, on average, not only more organic foods but

114

also more national brands (compared to private labels) and more non-organic products of higher quality

115

than non-organic consumers do. Second, we show that dietary patterns change considerably along the

116

gradient of the organic penetration rate. Indeed, regular organic consumers have dietary patterns

117

comprising more plant-based products, less meat, fewer alcoholic beverages and fewer prepared meals

118

than do occasional and non-organic consumers. Independent of the specific characteristics of organic

119

products, their diets are likely to be healthier and more environmentally friendly than those of other

120

consumers. Third, we show that the quantity effect resulting from modified consumption patterns

121

partially compensates for the quality effect resulting from the purchase of more expensive organic and

122

non-organic foods by regular organic consumers. This means that the price barrier is perhaps not so high,

123

with the major issue being the consumers’ willingness to change their dietary habits, the potential loss of

124

consumer welfare caused by a shift toward more sustainable dietary patterns, and the consumption of

125

less-preferred foods.

126

AC C

EP

TE D

112

ACCEPTED MANUSCRIPT

127 128

2. Data and methods

129 The original data set is based on the records of a representative consumer panel of French households

131

collected by Kantar Worldpanel. It is a home-scan data set that provides detailed information on all food

132

product purchases. Among other things, the data set provides the characteristics of the goods (e.g.,

133

brand, size, regular or diet product, organic or conventional), quantities purchased, and related

134

expenditures for each household. The data set also provides information on each household’s

135

socioeconomic characteristics, such as its demographic composition, socioeconomic status and income

136

class. To analyze the consumption practices related to organic foods, we considered food purchases in

137

2012. After excluding non-active consumers (that is, consumers who did not record their purchases

138

throughout 2012), we obtained a representative sample of 12,000 households.

M AN U

SC

RI PT

130

139

Previous studies have used different criteria to identify a gradient in the level of organic consumption. In

141

survey-based studies, the characterization of organic consumers often relies on frequency

142

questionnaires, and consumer types are defined using clustering methods (see, for instance,

143

Kesse-Guyot et al., 2013). In the present study, we used purchase data, which allowed us to define

144

consumer types based on real food expenditures. Following Wier et al. (2008), we defined an index

145

based on the ‘organic penetration rate’, that is, the organic expenditure share of each household. In this

146

way, we aimed to identify the consumers who are more or less involved in organic consumption

147

independent (a priori) of the absolute values of the expenditures and household incomes.

150

EP

149

2.1. Organic penetration rate, prices and socio-demographic characteristics of consumers

AC C

148

TE D

140

151

As a first step, we investigated the relationships between the organic expenditure shares and economic

152

and socio-demographic characteristics of the households. First incomes and prices must be taken into

153

account as they may have an effect on demand and consumption patterns. Family status (single, couple,

154

couple with/without children) and composition (e.g., age of family members, number and age of

155

children) may also affect food expenditure. In addition, other criteria, such as education level, location of

156

household, and obesity, may be associated with different food preferences and contribute to explaining

157

the variability of food and diet choices.

158

ACCEPTED MANUSCRIPT

To measure the effects of these variables, standard least squares regression techniques can be used to

160

provide summary point estimates that allow one to calculate the average effect of the independent

161

variables on the organic penetration rate. However, this focus on the average effect may hide important

162

features of the underlying relationship. Quantile regression techniques can provide a more complete

163

picture of this relationship by better taking into account consumer heterogeneity. Indeed, unlike

164

Ordinary Least-Square (OLS) regression, quantile regression is not limited to explaining the mean of the

165

dependent variable and can be employed to explain the determinants of the dependent variable at any

166

point in the distribution of the dependent variable (Chamberlain, 1994; Koenker, 2005). This seems to be

167

relevant, given the very high variability of organic consumption in the whole population (Buder et al.,

168

2014) and the need to pay special attention to the consumer price behavior of heavy and medium buyers

169

of organic foods (Rodiger et Hamm, 2015).

M AN U

170

SC

RI PT

159

171

For this reason, we performed a quantile regression (command ‘qreg’, Stata 13) in which the dependent

172

variable was the organic expenditure share. This allows estimating, on average and for different points of

173

the organic penetration rate, the distribution of the effects of the variables considered in the analysis.

174

0 Thus, we estimated the following equation, where the dependent variable ln wh is the logarithm of

175

organic share expenditure of household h and the explanatory variables are the conventional p h and

176

organic ph prices, the vector of socio-demographic variables Z h , and u 0h is the error term:

TE D

0

c

ln w 0h = γ 0 ln p0h + γ c ln pch + βh Zh + u 0h

177

The Kantar data set provides, for each household, the purchased quantities and expenditures but not the

179

prices. Price indexes are set by computing the unit values of conventional and organic purchases, which

180

are the ratio of total organic (conventional) expenditures to total organic (conventional) quantities. To

181

limit the risk of endogeneity1, organic and conventional unit values have been calculated per living area2

182

by considering the total organic (conventional) expenditures of households living in a specific living area

183

divided by the total organic (conventional) quantities purchased by these households. This index

184

expresses, to some extent, the average organic and conventional prices faced by all households living in

185

the same area.

AC C

EP

178

1

See S. Lecocq, J-M. Robin, « Estimating demand response with panel data », Empirical Economics (2006) 31:1043– 1060. 2 In the Kantar data set, France is divided into 1300 living areas. By computing the unit values per living areas, we consider that two households living in the same area face the same prices.

ACCEPTED MANUSCRIPT

186

To account for the size of the family, we included the number of consumption units (CU) as an

187

independent variable. This number was computed by using the scale in which the first adult counts for

188

one CU, each additional adult aged 15 years or older counts for 0.7 CU and each child less than 15 years

189

counts for 0.5 CU.

RI PT

190

In addition, we included the following variables: income level (modest, lower average, upper average,

192

well-off); age of the household shopper; education level (low/medium/high); family status (single,

193

married or couples without children, married or couple with children); location of the household (rural,

194

small city, medium city, large city); presence of at least one obese member in the household; and

195

presence of children under 3 years old. Some studies have suggested that organic consumption might be

196

associated with greater consumer involvement in cooking and meal preparation. In our dataset, we had

197

some indicators related to kitchen equipment that we also integrated into the quantile regression, such

198

as indicators of culinary practices (fridge, freezer, microwave oven, electric fryer).

199 200

2.2. Consumption patterns analysis

201

M AN U

SC

191

In a second step, we investigated the variation in consumption patterns along the gradient of the

203

organic penetration rate. To characterize different types of consumers that were more or less involved

204

in organic consumption, we split the entire sample of 12000 households into quintiles of the organic

205

penetration rate.

206

TE D

202

To compare the consumption patterns of these 5 quintiles, we calculated the average expenditures and

208

purchased quantities of organic and non-organic foodstuffs in the 23 food groups for each household3.

209

This food categorization is similar to those used in various papers dealing with consumers’ diet analysis.

210

These values were expressed per household and per home consumption unit (CU) using the Oxford scale

211

mentioned above. ANOVA tests (using the F distribution) were performed to accept or reject the

212

hypothesis that the quintile means were significantly different from each other.

AC C

EP

207

213 214

2.3. Decomposing the quality and quantity effects

215

3

Bottled water was not considered in this study due to the fact that water is not available in organic quality.

ACCEPTED MANUSCRIPT

In a third step, we analyzed the differences in the food budgets of the various quintiles to obtain a

217

clearer vision of the economic constraints that may limit organic consumption. In fact, differences in

218

food budgets can be due to differences in consumption patterns, that is, the quantities purchased in the

219

various food groups or they can be due to differences in the prices of the foodstuffs purchased in each

220

food group. To identify the magnitude of these two effects, we decomposed the variation of the total

221

average food budget of each quintile into a ‘quantity effect’ and a ‘quality effect’.

RI PT

216

222

Decomposition methods for group comparisons have been used in many economic studies (Fortin et al.,

224

2011). Here, we used the same method used by Griffith et al. (2017) for decomposing the variation in the

225

food budget by quintile. This decomposition critically relies on counterfactual exercises corresponding to

226

the following questions:

SC

223

M AN U

227 228

(i) For the ‘quantity effect’: what would the total food expenditure of the average consumer of the first

229

quintile be if he/she purchased the same quantities in each food group as the average consumer of the

230

fifth quintile without changing his/her food prices (i.e., without changing his/her quality strategy)?

231

(ii) For the ‘quality effect’: what would the variation in the total food expenditure of the average

233

consumer of the first quintile be if he/she continued to purchase the same quantities in each food group

234

but at the same average prices as the average consumer of the fifth quintile (i.e., without changing

235

his/her quantity strategy)?

236

TE D

232

To disentangle the quantity and quality effects, we proceeded as follows. Let us first consider the

238

variation in the total food budget ∆D T(5−1) between the first and fifth quintiles. As explained above, this

239

can be decomposed into several components using the following expression:

240 241 242 243

(1)

AC C

EP

237

n

n

i =1

i

∆D T(5−1) = ∑ ∆Di(5−1) = ∑  ∆Qi(5−1) pi1 + Qi1∆pi(5−1) + ∆Qi(5−1) ∆pi(5−1) 

244 245

where i=1,…, n are the food groups, ∆Qi(5−1) is the variation in the purchased quantities of each food

246

group between the first and the fifth quintiles, ∆p i(5−1) is the variation in the average prices of each food

ACCEPTED MANUSCRIPT

247

group between the first and the fifth quintiles, pi1 is the average price of each food group in the first

248

quintile, and Q i1 is the quantity purchased from each food group in the first quintile.

249 In expression (1), the first term denotes the share of the variation of total food budget ∆D T(5−1) that

251

results from the variation in the consumption patterns between the first and the fifth quintiles, assuming

252

the same average prices in each food group in both quintiles (the quantity effect). This term will be

253

negative if the average quantities purchased by the fifth-quintile consumers are lower than the

254

quantities purchased by consumers of the first quintile. The second term expresses the share of the

255

variation of food budget ∆D T(5−1) that is due to variation in the average prices of food groups between

256

the first and the fifth quintiles, assuming the same consumption patterns (the quality effect). This term

257

will be positive if regular organic consumers purchase, on average, more expensive foods in the different

258

food groups compared with non-organic consumers. The third term expresses the cross effect of the

259

quantity and the quality effects. It captures the covariance of the variations in quality and quantity. If,

260

compared to the first quintile, consumers in the fifth quintile shift toward food groups that are on

261

average less expensive, this term will be negative.

M AN U

SC

RI PT

250

262

To take a step forward, we considered the quality effect (the second term of expression (1)) more

264

precisely. Because the food budget is higher in the fifth quintile than in the first, this difference could

265

result from the purchase of (i) more expensive organic foods (e.g., if regular consumers buy organic

266

foods at specialty stores and occasional consumers buy organic foods at hypermarkets), (ii) more

267

expensive non-organic foods (e.g., if regular organic consumers buy more non-organic foods with

268

national brands and non-organic consumers buy more non-organic foods with private labels), or (iii) a

269

higher organic penetration rate in some or all food groups (because organic foods are more expensive

270

than non-organic foods). To disentangle these factors, we decomposed the quality effect using the

271

following expression4:

AC C

EP

TE D

263



272

=



+



+



+



(2)

273 4

The ‘quality effect’ estimated with expression (1) is computed for given quantities purchased from each food group.

ACCEPTED MANUSCRIPT

274

where

275

proportion of non-organic purchases made by the fifth quintile in food group i;

276

in the proportion of organic purchases from food group i between the fifth and the first quintiles; and

277

is the variation in the proportion of non-organic purchases from food group i between the fifth and the first quintiles. Thus, in expression (2):

279 280

is the

is the variation

RI PT

278

is the proportion of organic purchases made by the fifth quintile in food group i;

is the expenditure variation caused by differences in non-organic prices for each



food group between quintiles, assuming unchanged purchased quantities of non-organic foods. This

282

term expresses the share of the quality effect related to the fact that regular consumers buy more

283

expensive non-organic foods than occasional consumers do. −

is the expenditure variation related to differences in organic prices for each food

M AN U

284

SC

281

285

group between the fifth and the first quintiles, assuming unchanged purchased quantities of organic

286

products in each food group. This term expresses the share of the quality effect related to the fact

287

that regular consumers buy more expensive organic foods than occasional consumers.

288 289



is the expenditure variation related to differences in the share of organic vs. non-organic purchases between quintiles for each food group (assuming unchanged

291

prices of organic and non-organic foods in each food group between quintiles). This value expresses

292

the share of the quality effect caused by the fact that regular consumers buy more organic foods

293

(and fewer non-organic foods) than occasional consumers do.

294 295

297 298 299

3. Results

AC C

296

EP

TE D

290

3.1. Characteristics of organic consumers

300

First, we present the quantile regression estimates of the effect of prices and socio-demographic

301

characteristics on variations in the organic penetration rate. In Table 1, the first column (OLS) provides

302

the average effect for the entire sample. The other columns provide the effect by quantile of the

303

penetration rate distribution. It is then possible to assess the strength of the relationships between the

304

dependent and explanatory variables by considering consumer heterogeneity.

ACCEPTED MANUSCRIPT

305 306

[Insert Table 1]

307 The price effects are highly significant both on average and for all levels of the distribution of the organic

309

penetration rate. On average, the organic expenditure decreases by 0.55% and increases by 1.3% when

310

the organic and conventional prices increase by 1%, respectively.

311

RI PT

308

Regarding age, education and income, the results are highly significant in the OLS treatment and in the

313

quantile regression. A higher income, a higher education level and an older household shopper increase

314

the probability of observing a higher organic penetration rate at each level of the distribution. The higher

315

the quantile, the greater the effect of age, education and income level.

SC

312

M AN U

316 317

The number of consumption units is significant and affects negatively the organic penetration rate. The

318

presence of young children (under 3 years old) is positively correlated to the organic penetration rate.

319

However, this effect seems to be weaker among regular organic consumers.

320

Regarding the households’ locations, the OLS coefficients are significant for medium and large towns.

322

The presence of obese members in the household has a very significant and negative effect on the

323

probability of someone being an occasional or regular organic consumer. This effect is much stronger

324

among regular consumers.

325

TE D

321

Regarding kitchen equipment, ownership of microwaves and electric fryers significantly decreases the

327

probability of someone being an organic consumer, regardless of the level of organic penetration rate.

328

This effect is stronger for regular organic consumers.

330 331 332

AC C

329

EP

326

3.2. Consumption patterns and expenditures

333

It is worth noting that only 654 households bought no organic products in 2012, which is less than 6% of

334

the sample. This finding indicates that organic consumption is widespread among the population and

335

that many households purchase organic products at least a few times a year. The consequence is that

ACCEPTED MANUSCRIPT

336

the average organic penetration rate is very low, approximately 2.2%, which is similar to the values

337

obtained in other studies5.

338 The average food budgets and shares of organic expenditures for each quintile are given in Table 2.

340

These values are expressed per household and per consumption unit. The organic penetration rate

341

ranges between 0.08% and 8.1% from the first to the last quintile. Note that even in the last quintile,

342

organic food expenditure remains a small part of the total budget6. The total food budget, excluding

343

mineral water, ranges from 2280 to 2551 euros per consumption unit per year and depends on the

344

quintile; the food budget of the fifth quintile is 11.9% higher than that of the first quintile.

345 346 [Insert Table 2]

M AN U

347

SC

RI PT

339

348

Table 3 displays the average quantities of foods (organic and non-organic) purchased in each quintile of

350

the organic penetration rate distribution. It is interesting to note that except in the fresh fruits and

351

vegetables (F&V) group (and to a lesser extent in the meal substitutes and the fish and seafood groups),

352

purchased quantities decrease from the first to the last quintiles for most food groups. In particular, we

353

observe significant reductions in the purchased quantities of fresh and processed meats, prepared

354

meals, alcoholic and non-alcoholic beverages and fat7. In general terms, the higher the organic

355

penetration rate, the greater the purchased quantities of fresh fruit and vegetables, meal substitutes

356

and, to a lesser extent, fish and seafood and the lower the purchased quantities of fresh and processed

357

meats, prepared meals, alcoholic and non-alcoholic beverages and fat. When we compare the first and

358

the last quintiles, fresh F&V purchases increase by almost 45%, and the purchases of fresh meat,

359

processed meat, alcoholic and non-alcoholic beverages decrease by 18.9%, 29%, 44.6% and 38% from

360

the first to the last quintile. For the other food groups, the differences between quintiles do not show a

361

uniform trend or are not significant.

AC C

EP

TE D

349

362 363

Some of these trends are observed even among occasional organic consumers: this is the case for fresh

364

F&V and all beverages. Other trends are mainly observed among more involved organic consumers:

365

meat and processed meat consumption, processed F&V consumption and prepared meals consumption 5

See http://www.agencebio.org/la-bio-en-france For the last decile, the average penetration rate is 13.3%. 7 It should be noted that for fresh meats, this downward trend starts in the 2nd quintile. 6

ACCEPTED MANUSCRIPT

are only significantly lower among regular organic consumers. Thus, whereas the decrease in beverage

367

purchases and the increase in fresh F&V appear to be associated with an increase in the organic

368

penetration rate even among occasional consumers, reductions in meat, processed meat, processed F&V

369

consumption and prepared meals purchases seem to be specific markers of more involved organic

370

consumers.

RI PT

366

371

Let us now consider food expenditures. Significant decreases are observed mainly in three food groups:

373

fresh meat, processed meat, and alcoholic beverages8. However, in several food groups (e.g., cheese, hot

374

drinks), we observe an increase in expenditures although the purchased quantities are similar or even

375

lower in the last quintiles. This suggests that the variation in the total and food group expenditures by

376

quintile results in a mix of quantity and quality effects that must be disentangled. It is interesting to note

377

that these trends, though weaker, are also observed among occasional organic consumers, especially for

378

eggs, fish and seafood, desserts and starchy foods.

M AN U

SC

372

379 380

[Insert Table 3]

381

Table 4 displays the average quantities of organic products purchased from each food group. From the

383

first to the last quintile, the expansion of organic consumption relies on the following:

TE D

382

384

- An increase in the number of food groups from which consumers purchase organic foods. On average,

386

non-organic consumers purchase foods in 1.2 food groups, mainly fresh F&V, bread and flour and, to a

387

lesser extent, culinary ingredients and eggs. For regular organic consumers in the last quintile alone,

388

purchases are made in more than 10 food groups.

EP

385

AC C

389 390

- An increase in the quantities purchased from each food group, particularly fresh F&V, milk, eggs,

391

desserts, baby foods, bread and starchy food groups. Note that the higher level of organic baby foods in

392

the last quintile is consistent with the greater involvement of households with young children in regular

393

organic consumption.

394 8

In fact, we first observe an increase in meat consumption (quantities and expenditures) when moving from the first to the second decile. This likely indicates that economic constraints limit the meat consumption of the first decile.

ACCEPTED MANUSCRIPT

395

[Insert Table 4]

396 In most food groups, we observe the substitution of organic for non-organic products from the first to

398

the fifth quintiles. This is not the case for the fresh F&V group: quantities of both fruits and vegetables

399

increase among regular organic consumers compared with occasional consumers. Paradoxically, this

400

increase is such that regular organic consumers consume, on average, more non organic F&V than

401

non-organic consumers.

RI PT

397

402

Are regular organic consumers also different in their purchases of non-organic foods? In Table 5, we

404

examine the variations in non-organic foodstuffs by quintile of the organic penetration rate distribution.

405

The upper part of Table 6 displays the shares of non-organic food quantities dedicated to products with

406

nutritional claims. These claims provide additional guarantees about nutritional values and health

407

benefits, and products making such claims generally have higher prices than conventional products. The

408

middle part of Table 6 presents the shares of non-organic food quantities dedicated to non-organic

409

quality certifications (e.g., ‘Protected Designation of Origin’ [PDO], ‘Label Rouge’, Country of Origin

410

certifications). These certifications are generally related to taste and hedonic dimensions but may also be

411

linked to environmental or animal-welfare concerns. The lower part of the table displays the shares of

412

national brands and private labels among the non-organic purchases in each food group by quintile.

TE D

M AN U

SC

403

413

The purchase of non-organic products with nutritional claims and quality certifications9 and of national

415

brands rather than private labels can also reveal quality strategies implemented by organic consumers

416

with health, environmental and quality concerns.

EP

414

417

[Insert Table 5]

AC C

418 419 420

The higher the organic penetration rate, the higher the use of nutritional claims, of quality certifications,

421

and of national brands. Foods that are light in fat, sugar and salt or enriched with omega 3 products are

422

positively and significantly correlated with high levels of the organic penetration rate, suggesting that

423

organic consumers have strong concerns about health. Red labels and certificates of origins also appear 9

Aschemann-Witzel et al. (2013) considered the combination of organic and health claims. Their results show that products with a claim were neither significantly preferred nor rejected. Occasional organic buyers, however, were significantly more likely to choose products with a health claim. Here, we do not address this issue but focus instead on the health claims of non-organic foods purchased by organic consumers.

ACCEPTED MANUSCRIPT

more frequently in regular organic consumers’ baskets, indicating some interest in more demanding

425

production processes (for either taste or environmental reasons). Regarding local vs. global supply, we

426

only had this information for tomatoes. Local (French) supplies of (non-organic) tomatoes are much

427

more frequently purchased by regular organic consumers. In addition, organic consumers appear to buy

428

national brands (private labels) more (less) frequently than non-organic or occasional consumers do.

RI PT

424

429 430

3.3. Decomposing the variation in food budgets

431

As already noted, the average total food budget is 11.9% higher in the last quintile compared to the first.

433

Using expression (1) we decomposed this variation of food budget into a quality and a quantity effects

434

(Figure 1). The quality effect, that is, the effect due to differences in the average prices of each food

435

group between the two quintiles (assuming the same quantities purchased from each food group by

436

both quintiles), induces a 26.7% increase in the food budget. However, (i) regular organic consumers

437

purchase lower quantities in most food groups (this effect reduces the total food budget by 11.2%), (ii)

438

compared to non-organic consumers, regular organic consumers shift their purchases from more

439

expensive (meat, alcohol) to less expensive food groups (fresh F&V). This effect reduces the total food

440

budget by 3.6%. Thus, half of the budget increase resulting from the purchase of more expensive foods

441

(26.7%) is offset by changes in consumption patterns (11.2% + 3.6%).

TE D

M AN U

SC

432

442 443 444

[Insert Figure 1]

EP

445 446

Using expression (2), it is possible to more precisely describe the quality effect, which induces a 26.7%

448

increase in the total food budget of regular organic consumers (fifth quintile) compared with non-organic

449

consumers (first quintile). This increase in the food budget of regular organic consumers compared with

450

non-organic consumers is the result of three components (see Figure 2):

AC C

447

451 452

-

The increase in the average prices of non-organic products in each food group leads to a 22.6%

453

increase in the total food budget. Thus, all else being equal, regular organic consumers buy more

454

expensive non-organic foods than non-organic consumers. This is consistent with the observations

455

in Table 3, which shows that regular organic consumers purchase more products with nutritional

ACCEPTED MANUSCRIPT

456

claims and quality certifications, more national brands and fewer private-label brands than

457

non-organic or occasional organic consumers do.

458 459

-

The increase in the average price of organic products in each food group leads to a 2.4% increase in the food budget. This might occur because regular organic consumers buy more expensive organic

461

products in specialized organic stores whereas occasional consumers buy more often organic

462

products in non-specialized stores (see Table 6).

463 −

465

The increase in the share of organic products in each food group leads to a 1.7% increase in the food

SC

464

RI PT

460

budget.

466

The two last components, which are directly related to organic purchases, seem to have small impacts

468

compared to the first one. However, the effects induced by the prices of organic foods must be

469

considered by keeping in mind that even for regular organic consumers, the organic penetration rate

470

remains quite modest (approximately 8%).

471

[Insert Figure 2]

473

[Insert Table 6]

475 476 477

4. Discussion

478

EP

474

TE D

472

M AN U

467

Organic consumption has two important characteristics. On the one hand, it is widespread throughout

480

the population. Only 6% of households in our sample bought no organic foods in 2012. On the other

481

hand, it is very concentrated, as the expenditures of the fifth quintile of the organic penetration rate

482

distribution represent nearly 80% of total organic expenditures10. This remark highlights the structure of

483

the organic market, which is composed of a very large number of occasional consumers who buy few

484

products a year (by default or without paying really attention to organic claims), and a very small number

485

of committed consumers who have ‘food quality strategies’ and chose their stores accordingly. This very

AC C

479

10

This seems to be the case in many countries. For instance, in Germany, 17% of the population is responsible for 76% of all organic food purchases (Buder et al., 2014).

ACCEPTED MANUSCRIPT

486

strong heterogeneity in food practices imposes to consider differently consumers depending on their

487

involvement in organic consumption. It is for this reason that we tried to identify consumption practices

488

and dietary patterns for different points of the organic penetration by performing a quantile regression.

489 Socio-demographic characteristics, such as gender, income and education, often effectively describe the

491

typical organic consumer (Hughner et al., 2007; Krystallis and Chryssohoidis, 2005; Torjusen et al., 2010

492

and 2012; Wier et al., 2008). Despite some contradictory results in the literature, most studies report

493

that higher education levels and higher incomes increase the probability of being a regular organic

494

consumer. In our study, some socio-demographic characteristics have a positive impact, although some

495

variables are more strongly correlated with regular organic consumption than with occasional

496

consumption. We show that education level and household income are positively correlated with the

497

tendency to buy organic foods, and the strength of the correlation is greater for regular consumers.

M AN U

SC

RI PT

490

498

Concerning shopper age, the results are more ambiguous. Our results tend to show that age is positively

500

and significantly correlated with the organic penetration rate, and the effect is much higher for

501

consumers who frequently purchase organic products. Urbanization appears to have a strong impact on

502

organic consumption in many studies. Metropolitan areas in Great Britain and Denmark have

503

significantly higher organic budget shares (Wier et al., 2008). We also found this effect in France, with a

504

quite strong effect associated to medium and large cities.

505

TE D

499

The impact of family status is also ambiguous because it depends on the level of the organic penetration

507

rate (Riefer and Hamm, 2011). Like Wier et al. (2008), we show that the propensity to regularly purchase

508

organic foods is related to the presence of young children. One can assume that the presence of young

509

children leads parents to be more demanding about the quality (and safety) of foods.

AC C

510

EP

506

511

We have also shown that regular organic consumers have very different consumption patterns

512

compared with non-organic consumers; organic consumers purchase more fresh F&V, less fresh and

513

processed meat, and fewer alcoholic and non-alcoholic beverages. In addition, our observations indicate

514

that organic consumers are more involved in cooking and culinary practices, as suggested by their

515

purchase of fewer prepared meals and more fresh F&V (compared to processed F&V) and their reduced

516

likelihood of owning microwave ovens.

517

ACCEPTED MANUSCRIPT

Regarding consumption patterns, our results based on purchase data confirm the observations of a few

519

studies based on consumer surveys. Kesse-Guyot et al. (2013) suggested stepwise changes in food group

520

consumption among clusters of organic product consumers, with increased consumption of whole

521

grains, vegetables, fruits, soups, dried fruits, legumes, fruit and vegetable juices, sweet products,

522

vegetable oils and nuts and lower consumption of meats and processed meats, milk and dairy products,

523

sodas, alcoholic beverages, sweets and fat products, added fats and fast foods. In Norway, Torjusen et al.

524

(2012) noticed that food patterns and dietary quality differed between pregnant organic and

525

non-organic consumers. In Denmark, a survey showed that the highest quartile in relation to organic

526

preferences consumed 50% less meat than non-organic consumers did (Petersen et al., 2013). All these

527

results suggest that independent of the characteristics of organic foods, organic consumers have

528

healthier consumption patterns. This observation must be related to the fact that the probability of

529

being overweight or obese is significantly lower among regular organic consumers.

M AN U

SC

RI PT

518

530

Hoefkens et al. (2010) suggested that the health benefits experienced by organic consumers could be

532

related to their consumption patterns rather than the characteristics of the organic foods they

533

purchased. A demonstration of this hypothesis was beyond the scope of this article; however, the fact

534

that, on average, regular organic consumers purchase more conventional F&V than non-organic

535

consumers do reinforces the plausibility of this explanation.

536

TE D

531

Regarding environmental issues, the same conclusions can be suggested. Indeed, independent of the

538

environmental benefits of organic foods, regular organic consumers seem to better comply with

539

sustainable diet concepts that minimize environmental impact (Magkos et al., 2006) by consuming fewer

540

meat and animal-based products and more plant-based products (Aston et al., 2012; De Boer et al.,

541

2014; MacDiarmid et al., 2012; Masset et al., 2014).

AC C

542

EP

537

543

By considering the consumption practices of organic consumers more widely than many other studies

544

have, we show that regular organic consumers differ from non-organic and occasional consumers not

545

only because they buy more organic foods but also because they do not buy the same non-organic

546

products (e.g., they purchase more products with nutritional claims and quality certifications and more

547

national brands). Our results suggest that for organic consumers, organic food purchases are one

548

component of a more general ‘quality strategy’ that affects all food purchases and their entire diet. This

549

‘quality strategy’ is implemented through both organic and non-organic purchases and relies not only on

ACCEPTED MANUSCRIPT

a larger share of organic products but also on larger shares of high-quality non organic products and

551

more sustainable consumption patterns. It is also likely that, as shown by Hempel and Hamm (2016),

552

‘organic-minded’ consumers have stronger preferences for local products, and that local production

553

(here weakly identified the country of origins of some products) complements organic food production

554

fro them. This ‘quality strategy’ also affects the preferences for retail outlets, as organic foods are much

555

more purchased in specialized stores by regular organic consumers. This statement is in line with other

556

studies that stress on the connections between types of products, preferred retail outlets and consumer

557

affinities (de Boer and Schösler, 2016). Our analysis did not explicitly address the organic consumers’

558

motivation to pursue this ‘quality strategy’; however, as many authors suggest, both the consumption

559

patterns and the types of non-organic purchases indicate that hedonistic, health and environmental

560

concerns play key roles in organic consumers’ decisions.

M AN U

561

SC

RI PT

550

562

Regarding the food budgets of regular organic consumers, it turns out that they are higher than those of

563

non- and occasional organic consumers. However, we have shown that the gap is not that wide and that

564

the additional costs of more organic purchases (and higher quality non-organic products) are partially

565

offset by modified consumption patterns. The differences in household dietary habits appear to strongly

566

moderate the direct impacts of organic prices.

TE D

567

In our study, income is a variable that is clearly correlated with higher levels of organic food

569

consumption. This is not always the case, and in other studies, income does not seem to explain

570

differences in organic purchasing behavior. In fact, our results suggest that income might be interpreted

571

less as a budgetary constraint that limits organic purchases than as a characteristic of some consumer

572

groups (much the same as educational level) whose lifestyle choices prioritize healthier, more

573

environmentally friendly, and higher quality foods and dietary patterns.

575 576

AC C

574

EP

568

5. Conclusion

577

The goal of this article was to analyze the overall consumption patterns of organic consumers and their

578

impact on household food expenditures. We characterized how complete consumption patterns vary

579

according to the organic penetration rate; we linked these variations to the households’ socio-

580

demographic characteristics; and we estimated the impact of consumption patterns on food budgets.

581

ACCEPTED MANUSCRIPT

It is worth noting some original characteristics of the present study. First, it relies on purchase data of a

583

large consumer panel of French households. As noted by Wier et al. (2008), the use of such data is not

584

very frequent, although it might avoid some biases of declarative surveys. We also considered the

585

complete households’ consumption patterns, whereas many studies deal only with substitutions

586

between organic and conventional products within specific food groups. In addition, we analyzed the

587

overall food expenditures, including organic and non-organic purchases.

588

RI PT

582

This study also had some limitations. The analysis only took into account home-consumption, and

590

away-from-home consumption was excluded due to a lack of data. The results could be modified

591

because

592

socio-demographic characteristics. We conducted a static comparative study of consumption patterns,

593

whereas it would be very interesting to analyze changes in consumption patterns to obtain a better

594

understanding of the ‘trajectories’ of consumption changes associated with an increase in organic

595

purchases. For instance, do consumers shift to organic purchases because they already consume a large

596

quantity of fruits and vegetables, or do they increase their consumption of fruits and vegetables as they

597

begin to purchase more organic products? The implications for the promotion of organic consumption

598

would likely be different, depending on the answer to this question.

consumption

varies

dramatically

depending

on

the

consumers’

M AN U

away-from-home

SC

589

TE D

599

We have shown that household consumption patterns and diets vary dramatically along the gradient of

601

the organic penetration rate. In addition, for consumers who purchase organic food regularly, the

602

quantity effect resulting from modified consumption patterns partially compensates for the quality

603

effect of purchasing organic and non-organic foods that are more expensive. The consequence is that the

604

crucial question raised when promoting organic consumption is related not only to consumers’ WTP for

605

the health and environmental attributes of organic foods but also to their willingness to change their

606

dietary habits and shift toward more sustainable dietary patterns. In this perspective, an analysis of the

607

variation of consumer welfare stemming from the adoption of a more sustainable diet by non-organic

608

consumers is an important challenge for future research (de Boer et al., 2014; de Boer et al., 2016).

610

AC C

609

EP

600

Acknowledgements

611 612

The authors would like to thank Fabrice Etilé and Sébastien Lecocq for their suggestions and comments

613

on a first version of the article. This research was part of research program conducted in the ‘Bionutrinet’

ACCEPTED MANUSCRIPT

614

project, which was funded by the French National Research Agency (ANR /Coord. Emmanuelle Kesse-

615

Guyot).

617

Conflicts of interest

618

No conflict of interest to declare

RI PT

616

619 620 621 References

623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643

Aertsens, J., Verbeke, W., Mondelaers, K., Van Huylenbroeck, G. (2009). Personal determinants of organic food consumption: A review. British Food Journal, 111(10), 1140–1167.

SC

622

M AN U

Aschemann-Witzel, J., Maroscheck, N., Hamm, U. (2013). Are organic consumers preferring or avoiding foods with nutrition and health claims? Food Quality and Preference, 30(1), 68-76. Aston, L. M., Smith, J. N., Powles, J. W. (2012). Impact of a reduced red and processed meat dietary pattern on disease risks and greenhouse gas emissions in the UK: a modelling study. BMJ open, 2(5).

TE D

Avitia, J., Costa-Font, M., Gil, J. M., Lusk, J. L. (2015). Relative importance of price in forming individuals’ decisions toward sustainable food: A calibrated auction-conjoint experiment. Food Quality and Preference 41(0): 1-11. Bernard, JC., Bernard, DJ. (2009). What is it about organic milk? An experimental analysis. American Journal of Agricultural Economic, 91(3):826-836.

EP

Bezawada R., Pauwels K. (2013). What is special about marketing organic products? How organic assortment, price, and promotions drive retailer performance. Journal of Marketing , 77(1):31-51. de Boer, J., Schösler, H., Aiking, H. (2014). ”Meatless days” or “less but better”? Exploring strategies to adapt Western meat consumption to health and sustainability challenges. Appetite, 76:120-128. de Boer, J., De Witt A., Aiking H. (2016). Help the climate, change your diet: A cross-sectional study on how to involve consumers in a transition to a low-carbon society. Appetite, 98:19-27

646 647

de Boer, J., Schösler, H. (2016). Food and value motivation: Linking consumer affinities to different types of food products, Appetite, 103, 95-104.

648 649 650 651 652 653

Bruschi, V., Shershneva, K., Dolgopolova, I., Canavari, M., Teuber, R. (2015). Consumer Perception of Organic Food in Emerging Markets: Evidence from Saint Petersburg, Russia. Agribusiness, 31(3), 414– 432. doi:10.1002/agr.21414

AC C

644 645

Bryla, P. (2016). Organic food consumption in Poland: Motives and barriers. Appetite, 105 (1), 737-746.

ACCEPTED MANUSCRIPT

Buder, F., Feldmann, C., Hamm, U. (2014). Why regular buyers of organic food still buy many conventional products: Product-specific purchase barriers for organic food consumers. British Food Journal, 116(3), 390-404.

RI PT

Bunte, FH., van Galen, MA., Kuiper, WE., Tacken, G. Limits to growth in organic sales. De Economist 2010;158(4):387-410. Chamberlain, G. 1994. Quantile regression, censoring, and the structure of wages. In Advances in Economics Sixth World Congress, ed. Christopher A. Sims, 171-209. Cambridge University Press: Cambridge.

SC

Chen, J., Lobo, A. (2012). Organic food products in China: determinants of consumers’ purchase intentions. The International Review of Retail, Distribution and Consumer Research, 22(3), 293–314. doi:10.1080/09593969.2012.682596

M AN U

Chen, M.-F. (2007). "Consumer attitudes and purchase intentions in relation to organic foods in Taiwan: Moderating effects of food-related personality traits." Food Quality and Preference, 18(7): 1008-1021. Costanigro, M., Kroll, S., Thilmany, D., Bunning, M. (2014). "Is it love for local/organic or hate for conventional? Asymmetric effects of information and taste on label preferences in an experimental auction." Food Quality and Preference, 31(0): 94-105. Crinnion, W. J. (2010). Organic foods contain higher levels of certain nutrients, lower levels of pesticides, and may provide health benefits for the consumer. Alternative Medicine Review: A Journal of Clinical Therapeutic, 15(1), 4–12.

TE D

Dangour, A. D., Lock, K., Hayter, A., Aikenhead, A., Allen, E., Uauy, R. (2010). Nutrition-related health effects of organic foods: a systematic review. The American journal of clinical nutrition, 92(1), 203-210.

EP

Dowd, K. , Burke, K. J. (2013). "The influence of ethical values and food choice motivations on intentions to purchase sustainably sourced foods." Appetite, 69(0): 137-144. Fortin, N., Lemieux, T., Firpo, S. (2011). Decomposition methods in economics. Handbook of labor economics, 4, 1-102. Fernqvist, F., Ekelund, L. (2014). Credence and the effect on consumer liking of food – A review. Food Quality and Preference, 32(PC), 340–353.

AC C

654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700

Forman, J., Silverstein, J. (2012). Organic Foods: Health and Environmental Advantages and Disadvantages. Pediatrics, 130(5), e1406–e1415. doi:10.1542/peds.2012-2579 Goetzke, B., Nitzko, S., Spiller, A. (2014). "Consumption of organic and functional food. A matter of well-being and health?" Appetite, 77(0): 96-105. Grannis, J., Hine, S., Thilmany, D. (2003). Marketing Premium Food Products in Emerging Economies. Journal of International Food & Agribusiness Marketing, 13(2-3), 59–76.

ACCEPTED MANUSCRIPT

Griffith, R., O'Connell, M., Smith, K. (2017). The Importance of Product Reformulation versus Consumer Choice in Improving Diet Quality. Economica, 84: 34–53 Heerwagen, L. R., Andersen, L. M., Christensen, T., Sandøe, P. (2014). "Can increased organic consumption mitigate climate changes?" British Food Journal, 116(8): 1314-1329.

RI PT

Hemmerling, S., Hamm, U., Spiller, A. (2015). Consumption behaviour regarding organic food from a marketing perspective—a literature review. Organic Agriculture, 5(4), 277–313. Hempel, C., Hamm, U. (2016). How important is local food to organic-minded consumers? Appetite, 96:309-318.

SC

Hoefkens, C., Sioen, I., Baert, K., De Meulenaer, B., De Henauw, S., Vandekinderen, I., Devlieghere, F., Opsomer, A., Verbeke, W., Van Camp, J. (2010). Consuming organic versus conventional vegetables: The effect on nutrient and contaminant intakes. Food and Chemical Toxicology, 48(11), 3058-3066.

M AN U

Honkanen, P., Verplanken, B., Olsen, S. O. (2006). Ethical values and motives driving organic food choice. Journal of consumer behaviour, 5(5), 420-430. Hughner, R. S., McDonagh, P., Prothero, A., Shultz, C. J., Stanton, J. (2007). Who are organic food consumers? A compilation and review of why people purchase organic food. Journal of consumer behaviour, 6(2-3), 94-110. Janssen M., Hamm U. (2012). Product labelling in the market for organic food: Consumer preferences and willingness-to-pay for different organic certification logos. Food Quality and Preference, 25(1):9-22.

TE D

Jensen, M. M., Jørgensen, H., Lauridsen, C. (2013). Comparison between conventional and organic agriculture in terms of nutritional quality of food - a critical review. CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources, 8(045).

EP

Kesse-Guyot, E., Peneau, S., Mejean, C., de Edelenyi, F. S., Galan, P., Hercberg, S., Lairon, D. (2013). Profiles of organic food consumers in a large sample of French adults: Results from the nutrinet cohort study. PloS one, 8(10) Kihlberg, I. E. Risvik (2007). Consumers of organic foods – value segments and liking of bread. Food Quality and Preference, 18(3): 471-481.

AC C

701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748

Koenker, R. (2005). Quantile Regression. Cambridge University Press: New York. Krystallis, A., Chryssohoidis, G. (2005). Consumers' willingness to pay for organic food: Factors that affect it and variation per organic product type. British Food Journal, 107(5), 320-343. Lairon, D. (2010). Nutritional quality and safety of organic food. A review. Agronomy for sustainable development, 30(1), 33-41. Lee, H.-J., Z.-S. Yun (2015). Consumers’ perceptions of organic food attributes and cognitive and affective attitudes as determinants of their purchase intentions toward organic food. Food Quality and Preference, 39(0): 259-267.

ACCEPTED MANUSCRIPT

Lee, W.-c. J., Shimizu, M., Kniffin, K. M., Wansink, B. (2013). "You taste what you see: Do organic labels bias taste perceptions?" Food Quality and Preference, 29(1): 33-39.

RI PT

Lopez E., Lopez RA. (2009). Demand for differentiated milk products: implications for price competition. Agribusiness, 25(4):453-465. Macdiarmid, J. I., Kyle, J., Horgan, G. W., Loe, J., Fyfe, C., Johnstone, A., MacNeill, G. (2012). Sustainable diets for the future: can we contribute to reducing greenhouse gas emissions by eating a healthy diet? The American journal of clinical nutrition, 96(3), 632-639.

SC

Magkos, F., Arvaniti, F., Zampelas, A. (2006). Organic food: buying more safety or just peace of mind? A critical review of the literature. Critical reviews in food science and nutrition, 46(1), 23-56.

M AN U

Magnusson, M. K., Arvola, A., Hursti, U.-K. K., Åberg, L., Sjödén, P.-O. (2003). Choice of organic foods is related to perceived consequences for human health and to environmentally friendly behaviour. Appetite, 40(2), 109-117. Marian, L., Chrysochou, P., Krystallis, A., Thøgersen, J. (2014). The role of price as a product attribute in the organic food context: An exploration based on actual purchase data. Food Quality and Preference, 37, 52-60. Marques Vieira, L., Dutra De Barcellos, M., Hoppe, A., Bitencourt da Silva, S. (2013). An analysis of value in an organic food supply chain. British Food Journal, 115(10), 1454–1472.

TE D

Masset, G., Vieux, F., Verger, E. O., Soler, L.-G., Touazi, D., Darmon, N. (2014). Reducing energy intake and energy density for a sustainable diet: a study based on self-selected diets in French adults. The American journal of clinical nutrition, 99(6), 1460-1469.

EP

Meier, M. S., Stoessel, F., Jungbluth, N., Juraske, R., Schader, C., Stolze, M. (2015). Environmental impacts of organic and conventional agricultural products-Are the differences captured by life cycle assessment? Journal of Environmental Management, 149, 193–208. Micheelsen, A., Havn, L., Poulsen, S. K., Larsen, T. M., Holm, L. (2014). The acceptability of the New Nordic Diet by participants in a controlled six-month dietary intervention. Food Quality and Preference, 36, 20-26.

AC C

749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795

Monier S., Hassan D., Nichèle V., Simioni, M. (2009). M. Organic food consumption patterns. Journal of agricultural & food industrial organization, 7(2). Petersen, S. B., Rasmussen, M. A., Strøm, M., Halldorsson, T. I., Olsen, S. F. (2013). Sociodemographic characteristics and food habits of organic consumers–a study from the Danish National Birth Cohort. Public health nutrition, 16(10), 1810-1819. Pino, G., Peluso, A. M., Guido, G. (2012). Determinants of regular and occasional consumers' intentions to buy organic food. Journal of Consumer Affairs, 46(1), 157-169.

ACCEPTED MANUSCRIPT

796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830

Riefer, A., Hamm, U. (2011). Organic food consumption in families with juvenile children. British Food Journal, 113(6), 797-808.

831 832 833 834 835 836 837 838 839 840 841 842

Torjusen, H., Lieblein, G., Næs, T., Haugen, M., Meltzer, H. M., Brantsæter, A. L. (2012). Food patterns and dietary quality associated with organic food consumption during pregnancy; data from a large cohort of pregnant women in Norway. BMC public health, 12(1), 612.

Rödiger, M., Hamm, U. (2015). How are organic food prices affecting consumer behaviour? A review. Food Quality and Preference, 43, 10-20.

RI PT

Risku-Norja, H., Mäenpää, I. (2007). MFA model to assess economic and environmental consequences of food production and consumption. Ecological Economics, 60(4), 700–711. Ruiz de Maya, S., López-López, I., Munuera, J. L. (2011). Organic food consumption in Europe: International segmentation based on value system differences. Ecological Economics, 70(10), 1767– 1775.

SC

Scalco, A., Noventa, S., Sartori, R., Ceschi, A. (2017). Predicting organic consumption : A meta-analytic structural equation model based on the theory of planned behavior. Appetite, 112 (1), 235-248.

M AN U

Schröck R. (2012).The organic milk market in Germany is maturing: A demand system analysis of organic and conventional fresh milk segmented by consumer groups. Agribusiness, 28(3):274-292. Schleenbecker, R., Hamm, U. (2013). Consumers’ perception of organic product characteristics. A review. Appetite, 71, 420-429. Smed, S., Andersen, L. M., Kærgård, N., Daugbjerg, C. (2013). A matter of trust: how trust influence organic consumption. Journal of Agricultural Science, 5(7), p91.

TE D

Smith-Spangler, C., Brandeau, M. L., Hunter, G. E., Bavinger, J. C., Pearson, M., Eschbach, P. J., Sundaram, V., Liu, H., Schirmer, P., Stave, C., Olkin, I., Bravata, D.M. (2012). Are organic foods safer or healthier than conventional alternatives? a systematic review. Annals of Internal Medicine, 157(5), 348-366.

EP

Thøgersen, J., de Barcellos, M. D., Perin, M. G., Zhou, Y. (2015). Consumer buying motives and attitudes towards organic food in two emerging markets. International Marketing Review, 32(3/4), 389–413.

AC C

Torjusen, H., Brantsæter, A. L., Haugen, M., Lieblein, G., Stigum, H., Roos, G., Holmboe-Ottesen, G. Meltzer , H.M.(2010). Characteristics associated with organic food consumption during pregnancy; data from a large cohort of pregnant women in Norway. BMC public health, 10(1), 775.

Vieux, F., Darmon, N., Touazi, D., Soler, L. G. (2012). Greenhouse gas emissions of self-selected individual diets in France: Changing the diet structure or consuming less? Ecological Economics, 75, 91-101. Wier, M., Jensen, K. O. D., Andersen, L. M., Millock, K. (2008). The character of demand in mature organic food markets: Great Britain and Denmark compared. Food Policy, 33(5), 406-421. Welsch, H., Kühling, J. (2009). Determinants of pro-environmental consumption: The role of reference groups and routine behavior. Ecological Economics, 69(1), 166–176.

ACCEPTED MANUSCRIPT

Yiridoe, E. K., Bonti-Ankomah, S., Martin, R. C. (2005). Comparison of consumer perceptions and preference toward organic versus conventionally produced foods: A review and update of the literature. Renewable Agriculture and Food Systems, 20(4), 193–205. Zander, K., Hamm, U. (2010). Consumer preferences for additional ethical attributes of organic food. Food Quality and Preference, 21(5), 495-503.

RI PT

843 844 845 846 847 848 849 850 851 852

SC

853 854 855

M AN U

856 857 858 859 860 Highlights

862

864 865 866

• There is a strong variation in household consumption patterns along the gradient of the organic

penetration rate.

EP

863

TE D

861

• Regular organic consumers have a ‘quality strategy’ that affects both organic and nonorganic

purchases. Not only do they buy more organic products, but regular organic consumers purchase also

868

non-organic items with more health claims and quality certifications and purchase more national

869

brands than occasional and non-organic consumers.

870

AC C

867

871

• The food budget of organic consumers is higher than those of occasional and non-organic consumers,

872

but the variation in consumption patterns along the gradient of the organic penetration rate largely

873

compensates for the additional costs of more organic purchases.

874

ACCEPTED MANUSCRIPT

Upper average Well-off

Non Declared

Low Education Degree Medium Education Degree High Education Degree

Consumption units

Rural Small cities Medium cities

Large cities

0.013

***

(0.001)

0.817

***

(0.116)

0.012

***

(0.001)

(0.030)

0.958

***

(0.089)

0.013

***

(0.001)

Q(0.30)

-0.378

***

Q(0.40)

-0.471

(0.026)

1.160

***

(0.076)

0.013

***

(0.001)

***

(0.026)

1.073

***

(0.077)

0.015

***

(0.001)

Q(0.50)

-0.526

***

(0.050)

(0.052)

(0.040)

0.005

-0.067

*

(0.036)

(0.039)

(0.030)

Ref

Ref

-0.128

0.275

***

***

(0.046)

-0.437

**

(0.139)

0.332

Ref

***

(0.047)

-0.464

**

(0.179)

(0.035)

-0.127

***

(0.035)

-0.109

(0.026)

0.267

Ref

***

(0.036)

-0.724

***

(0.137)

0.221

***

(0.031)

-0.531

***

(0.118)

Q(0.60)

-0.608

(0.029)

1.117

***

(0.086)

0.015

***

(0.001)

-0.204*** -0.151** -0.136*** -0.219*** -0.182*** -0.117**

SC

Lower Average

(0.105)

(0.039)

-0.345

***

M AN U

Modest

1.314

***

Q(0.20)

***

(0.039)

-0.144

***

***

(0.026)

1.192

***

(0.077)

0.013

***

(0.001)

-0.098** (0.035)

-0.090

***

Q(0.70)

-0.717

***

(0.029)

1.238

***

(0.084)

0.012

***

(0.001)

Q(0.80)

-0.842

***

(0.033)

1.599

***

(0.097)

0.014

***

(0.001)

Q(0.90)

-0.934*** (0.042)

1.908*** (0.122)

0.012*** (0.001)

-0.112** -0.294*** -0.530*** (0.038)

-0.165

***

(0.044)

-0.238

***

(0.055)

-0.255***

(0.026)

(0.029)

(0.026)

(0.029)

(0.033)

(0.042)

Ref

Ref

Ref

Ref

Ref

Ref

0.236

***

(0.032)

-0.618

***

(0.120)

0.208

***

(0.035)

-0.731

***

(0.132)

0.272

***

(0.031)

-0.557

***

(0.118)

0.422

***

(0.034)

-0.406

**

(0.130)

0.298

***

(0.040)

-0.529

***

(0.150)

0.161** (0.050)

0.21 (0.189)

-0.304*** -0.359*** -0.266*** -0.238*** -0.212*** -0.272*** -0.249*** -0.311*** -0.451*** -0.253***

TE D

Age

(0.035)

-0.293

***

(0.040)

(0.046)

(0.035)

(0.030)

(0.031)

(0.034)

(0.031)

(0.034)

(0.039)

(0.049)

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

0.146

***

(0.039)

-0.016

(0.043)

EP

Log price of non organic products

-0.557

-0.009

***

(0.002)

0.004

0.122

***

(0.033)

-0.006

**

0.106

***

(0.028)

-0.007

***

0.140

***

(0.028)

-0.009

***

0.196

***

(0.031)

-0.014

***

0.254

***

(0.028)

-0.019

***

0.208

***

(0.031)

-0.021

***

(0.002)

0.178

***

(0.036)

-0.015

***

(0.002)

0.370*** (0.045)

-0.003

(0.003)

(0.002)

(0.002)

(0.002)

(0.002)

(0.002)

-0.008

-0.134*

0.042

0.067

0.016

-0.007

-0.072*

0.026

0.058

0.008

(0.046)

(0.053)

(0.041)

(0.035)

(0.036)

(0.039)

(0.035)

(0.039)

(0.045)

(0.056)

AC C

Log price of organic products

Q(0.10)

***

RI PT

OLS

(0.003)

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

0.125*

0.075

0.168***

0.152***

0.118**

0.154***

0.131**

0.198***

0.138**

0.068

(0.054)

(0.062)

(0.047)

(0.041)

(0.041)

(0.046)

(0.041)

(0.045)

(0.052)

(0.065)

*

**

0.114

**

(0.043)

0.02 (0.048)

0.134

***

(0.036)

0.076

(0.031)

0.096

(0.032)

0.130

***

(0.035)

0.079

*

(0.031)

0.140

***

(0.035)

0.171

***

(0.040)

0.245*** (0.050)

ACCEPTED MANUSCRIPT

-0.277*** -0.125*** -0.182*** -0.270*** -0.237*** -0.237*** -0.275*** -0.346*** -0.380*** -0.393*** (0.030)

No children less than 3 years old

No microwave

(0.024)

(0.022)

(0.024)

(0.028)

(0.035)

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

-0.048

-0.135*

-0.026

0.143**

0.202***

0.06

0.120*

0.116*

0.170*

(0.062)

(0.069)

(0.053)

(0.045)

(0.046)

(0.051)

(0.045)

(0.050)

(0.058)

(0.072)

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

***

-0.525

***

-0.407

***

-0.473

***

(0.053)

(0.056)

(0.043)

(0.037)

Ref

Ref

Ref

Ref

1.056

Non declared

(0.022)

0.136*

-0.608

Microwave

(0.022)

***

(0.248)

1.248

***

(0.203)

1.096

***

(0.156)

RI PT

Children less than 3 years old

(0.025)

-0.495

***

0.729

***

(0.134)

-0.672

***

-0.706

***

-0.713

***

-0.838

***

-0.997***

(0.037)

(0.041)

(0.037)

(0.041)

(0.047)

(0.059)

Ref

Ref

Ref

Ref

Ref

Ref

SC

No Obese member

(0.033)

0.957

M AN U

Obese member

***

(0.136)

1.271

***

(0.150)

1.598

***

(0.134)

1.229

***

(0.148)

0.738

***

(0.171)

0.32 (0.215)

-0.267*** -0.122*** -0.104*** -0.168*** -0.299*** -0.270*** -0.374*** -0.399*** -0.410*** -0.332***

Fryer No Fryer Non declared

(0.034)

(0.036)

Ref

Ref

-0.159

0.122

(0.136)

(0.144)

(0.028)

(0.024)

(0.024)

(0.027)

(0.024)

(0.026)

(0.031)

(0.038)

Ref

Ref

Ref

Ref

Ref

Ref

Ref

Ref

0.139

adj. R2

0.138

Standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1

AC C

Table 1. Results of the quantile regression

EP

R2

(0.153)

-0.214

-0.083

-0.042

-0.162

-0.062

-0.062

(0.095)

(0.096)

(0.106)

(0.095)

(0.105)

(0.121)

(0.152)

0.155

0.703***

1.012***

1.363***

(0.101)

(0.111)

(0.128)

(0.161)

TE D

(0.145)

-0.334

**

(0.110)

-0.653*** -2.610*** -2.190*** -1.710*** -1.141*** -0.543***

Constant

(0.117)

(0.101)

(0.102)

(0.113)

-0.313

**

ACCEPTED MANUSCRIPT Q2 Q3 Q4

Q1

Q5

P-value1

0.00000

2280 (1,054)

2377 (981)

2426 (976)

2488 (1,015)

2551 (984)

Annual organic expenditure per household and per consumption units (euros)

1.8

7.0

15.0

32.0

206.0

Average organic penetration rate (%)

0.08

0.30

0.62

1.30

1

RI PT

Total annual expenditure per household and per CU (euros)

8.10

P-value is based on anova test for heterogeneity between quantiles. Standard errors in parentheses.

AC C

EP

TE D

M AN U

SC

penditures per household and consumption unit and average organic penetration rate for the 5 quintiles of the organic penetration rate distribution

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

rate distribution

ACCEPTED MANUSCRIPT

Mean quantities (kg or l/year) Q1

Q2

Q3

Q4

Q5

Mean expenditures (euros/year) P-value

1

Q1

Q2

Q3

Q4

Q5

P-value1

Fresh Fruits and Vegetables

109.00 126.30 137.84 147.57 158.40

0.00000

197.54 229.89 262.14 295.12 351.45 0.00000

p<0.0001

p<0.0001

Processed Fruits and Vegetables

30.08

30.46

30.55

30.85

26.08

0.00000

65.84

0.00000

p<0.0001

p<0.0001

Fresh meat

34.96

37.54

36.45

34.76

28.33

0.00000

308.74 340.67 335.93 335.63 307.06 0.00030

p<0.0001

0.0003

Processed meat

21.67

21.48

20.41

18.38

15.37

0.00000

205.79 209.96 207.83 199.71 190.74 0.00060

p<0.0001

0.0006

Fish and Seafood

13.07

14.14

14.97

15.49

15.08

0.00000

121.56 133.03 148.92 160.25 178.65 0.00000

p<0.0001

p<0.0001

Eggs

6.85

7.75

7.86

7.71

7.79

0.00030

20.36

0.0003

p<0.0001

Prepared Meals

33.47

30.73

29.79

28.54

26.27

0.00000

169.27 163.16 163.89 162.99 163.93

ns

p<0.0001

0.8297

Meal substitutes

0.03

0.08

0.09

0.18

0.52

0.00000

0.67

0.95

1.42

2.37

4.91

0.00000

p<0.0001

p<0.0001

Milk

55.92

55.01

56.39

53.12

52.20

ns

40.53

40.86

42.82

42.96

51.08

0.00000

0.1494

p<0.0001

75.54

80.78

RI PT

71.51

M AN U

SC

24.44

26.33

28.59

79.28

34.46

0.00000

49.05

47.21

49.66

49.60

48.95

ns

109.98 112.87 121.60 124.79 136.68 0.00000

0.5051

p<0.0001

Cheese

18.37

19.25

18.63

18.18

17.25

0.00000

155.60 167.11 166.22 169.49 176.87 0.00000

p<0.0001

p<0.0001

Starchy foods

13.16

13.38

12.82

12.87

12.42

ns

27.42

29.77

31.28

34.90

37.88

0.00000

0.1213

p<0.0001

Bread, Flour

13.48

11.80

11.77

13.00

12.71

0.03810

29.67

28.61

29.96

35.82

41.54

0.00000

0.0381

p<0.0001

Fat

12.10

12.10

11.70

11.14

10.53

0.00000

50.58

51.32

51.18

50.78

54.00

0.09820

p<0.0001

0.0982

Sweeteners

13.42

12.26

12.07

12.30

12.79

0.03460

39.42

37.27

39.07

43.44

51.48

0.00000

0.0346

p<0.0001

Culinary ingredients

19.14

19.98

0.0003

p<0.0001

Baby foods

10.82

7.32

Non Alcoholic beverages

19.39

17.99

0.00030

64.34

71.18

74.28

78.07

82.72

0.00000

7.45

7.15

11.07

0.00520

49.80

34.57

33.23

29.38

49.04

0.00120

0.0052

0.0012

104.76 101.23 89.96

82.43

64.25

0.00000

86.21

87.71

83.50

85.42

80.22

0.09030

p<0.0001

0.0903

5.64

5.27

0.00060

57.32

62.21

63.45

69.12

73.78

0.00000

0.0006

p<0.0001

49.15

32.56

0.00000

242.63 237.99 230.97 216.30 163.25 0.00000

p<0.0001

p<0.0001

5.94

Alcoholic beverages

58.76 18.28

Confectionary products.

13.31

1

58.65

5.11

5.63

53.52

18.94

17.80

16.20

14.66

0.00000

95.23 100.41 96.94

95.05

ns

p<0.0001

0.1566

13.45

13.20

12.62

11.35

0.00000

104.04 107.95 106.67 110.82 108.31

ns

p<0.0001

0.2648

4.54

4.35

4.68

4.45

0.00000

37.26

0.00000

p<0.0001

p<0.0001

AC C

Biscuits, cakes and pastries

5.97

19.58

EP

Hot drinks

Appetizers

TE D

Desserts

33.91

33.04

94.06

36.92

38.65

P-value is based on anova test for heterogeneity between quantiles; ns indicates a P-value > 0.1 Table 3. The total quantities and expenditures per food group for the 5 quintiles of the organic penetration rate distribution

ACCEPTED MANUSCRIPT

Mean quantities of organic products (kg or l/ year and/ CU) Q4

Q5

P-value1

0.36 0.02 0.00 0.00 0.00 0.02 0.01 0.00 0.05 0.03 0.00 0.03 0.05 0.02 0.01 0.05 0.00 0.05 0.00 0.00 0.01 0.00 0.00

1.07 0.06 0.00 0.01 0.02 0.07 0.06 0.00 0.26 0.08 0.02 0.07 0.15 0.04 0.03 0.19 0.01 0.20 0.02 0.01 0.03 0.01 0.00

1.90 0.14 0.02 0.02 0.03 0.15 0.10 0.01 0.66 0.19 0.03 0.11 0.22 0.07 0.08 0.33 0.08 0.52 0.04 0.08 0.05 0.02 0.01

2.50 0.25 0.02 0.03 0.07 0.50 0.19 0.02 1.93 0.49 0.04 0.30 0.44 0.17 0.20 0.69 0.23 1.10 0.10 0.10 0.09 0.03 0.03

10.64 1.45 0.34 0.27 0.21 2.13 1.37 0.27 13.26 4.08 0.34 1.68 2.08 1.05 1.33 1.80 2.52 5.54 0.57 0.50 0.69 0.27 0.15

0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000

SC

RI PT

Q3

P-value is based on anova test for heterogeneity between quantiles.

TE D

1

Q2

M AN U

Fresh Fruits and Vegetables Processed Fruits and Vegetables Fresh meat Processed meat Fish and Seafood Eggs Prepared Meals Meal substitutes Milk Desserts Cheese Starchy foods Bread, Flour Fat Sweeteners Culinary ingredients Baby foods Non Alcoholic beverages Hot drinks Alcoholic beverages Biscuits, cakes and pastries Confectionary products. Appetizers

Q1

AC C

EP

able 4. Purchased quantities of organic foods in the 5 quintiles of the organic penetration rate distribution

ACCEPTED MANUSCRIPT

Mean shares of products with nutrition claims Q2

Q3

Q4

Q5

P-value1

8.03 25.45 8.88 7.57 0.92 0.28 0.56 0.18 1.52

8.32 26.94 8.66 4.48 1.22 0.22 0.25 0.13 1.11

8.17 30.76 8.70 5.80 2.21 0.43 0.34 0.09 1.90

9.93 34.74 8.84 8.40 2.00 0.43 0.79 0.16 1.91

11.71 41.66 12.47 8.22 2.98 0.58 0.55 0.62 3.16

0.00000 0.00000 0.00410 0.00000 0.00040 0.00050 0.00530 0.00010 0.00000

RI PT

Light in sugar desserts Enriched in omega3 fat Light in sugar jam Sweeteners without sugars Light in salt chips Light in fat cheese Light in fat ham Enriched in omega3 ham Light in salt ham

Q1

Mean shares of products with official quality certifications Q2

Q3

Q4

0.28 13.39 6.44 0.57 29.70

0.22 13.26 6.01 0.69 27.21

0.43 17.12 6.23 0.87 31.39

0.43 25.44 6.77 0.74 29.97

Q5

P-value1

0.58 35.82 8.85 1.07 39.19

0.00070 0.00000 0.00000 0.00070 0.00000

SC

Labelled cheeses Labelled eggs Labelles meat Labelled fish Labelled tomatoes from France

Q1

Q2

80.75 27.02 65.40 69.33 61.35 61.35 60.88 67.28 66.55 42.31 44.39

80.31 23.92 63.75 67.80 59.85 60.09 59.72 68.79 64.58 38.94 44.16

Q3

Q4

Q5

P-value1

79.79 23.65 62.03 67.04 59.52 58.97 59.41 68.21 64.40 38.54 43.40

76.91 23.38 58.84 65.59 56.40 54.46 55.06 63.82 59.11 35.25 39.69

76.19 22.59 52.56 56.46 52.83 52.32 53.37 58.57 58.67 33.88 37.85

0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000

P-value is based on anova test for heterogeneity between quantiles.

EP

1

Q1

TE D

Processed Fruits and Vegetables Fresh meat Processed meat Eggs Prepared Meals Desserts Cheese Bread, Flour Culinary ingredients Hot drinks Confectionary products.

M AN U

Mean shares of private labels

AC C

ble 5. Characteristics of the non-organic purchases of the 5 quintiles of the organic penetration rate distribution (% of non-organic purchases)

RI PT SC M AN U TE D EP AC C

organic purchases)

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

30%

RI PT

26.7%

SC

25%

M AN U

20%

15%

11.9%

0%

-5%

Variation of the food budget between the first and the fifth quantiles

Quantity effect

AC C

5%

EP

TE D

10%

-10%

-11.2% -15%

Quality effect

Cross Quantity-Quality effect -3.6%

ACCEPTED MANUSCRIPT

30%

RI PT

26.7% 25%

SC

22.6%

M AN U

20%

TE D

15%

AC C

EP

10%

5%

2.4%

1.7%

0%

Quality effect

Non-organic price effect

Organic price effect

Ratio Organic/non-organic effect

ACCEPTED MANUSCRIPT Q2

Q3

Q4

Q5

Non-specialized stores (hypermarkets, supermarkets, harddiscounts)

98.90%

98.10%

97.40%

95.20%

66.90%

Organic specialized stores

0.10%

0.30%

0.80%

2.60%

31.40%

Other

1.00%

1.60%

1.80%

2.20%

1.80%

RI PT

Q1

AC C

EP

TE D

M AN U

SC

Table 6. Repartition of organic expenditures in organic specialized and non-specialized stores in the 5 quintiles of the organic penetration rate distribution

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

tration rate distribution