C2.6: Domestic market

• C2.6.1 Introduction
• C2.6.2 Approaches
  • C2.6.2.1 Household panels
  • C2.6.2.2 Retail panels
  • C2.6.2.3 Surveys among retailers
  • C2.6.2.4 Surveys of farm shops and farmers’ markets
  • C2.6.2.5 Expert estimates
• C2.6.3 Concluding remarks

Organic products are sold through a variety of outlets, such as supermarkets (i.e. multiple retailers), independent shops including specialist organic shops, artisan food producers (e.g. bakers and butchers), farmers markets, farm shops and by mail order. 

All of the data collection approaches described below have the weakness that they do not cover the full organic market. Household panels show the whole variety of marketing channels, but with different coverage rates. Retail panels based on electronic point of sale (EPOS) data focus on sales in participating multiple retailers and other outlets that use scanners at check-out. Retail sales through non-multiple channels can be important in the organic market (up to 40% of all organic sales in some countries). Sales through direct marketing through farm shops and weekly markets or sales through independent and specialist organic shops can be difficult to track without conducting specific surveys unless specific panels exist. The challenge of achieving good coverage of all sales channels, and thus a reliable estimate of the total organic market, is discussed further below.

Different sources are used to estimate the sales values in specific outlets and of the total domestic market, as well as the breakdown by product, and include:

• household panels,
• retail panels (multiple retailers, specialist panels of independent/organic shops in some countries),
• surveys of retailers,
• surveys of farm shops and farmers’ markets,
• expert estimates.

Panels are selected to be representative of the total food market. Even in the most well developed organic markets, such as that of Germany, the organic food market covers less than 10% of the total food market. Because of the small size of the organic market relative to the total market, the number of households recording organic purchases in a specific product category can be very small (depending on the size of the panel). Small errors, for example in product classification, could potentially have a large distorting impact on the total value of the organic food market when panel data are extrapolated to the total organic sales data. Furthermore, it is known that a relatively small proportion of households account for a relatively large proportion of total spending in the organic food market (e.g. 20% of households account for 80% of spending in the UK (Buder et al., 2010; SA/WDA/OCW, 2004). This distribution can only partially be explained through demographic variables (such as education, age, income). If, in the composition of such a panel, the number of high spending organic households is over- or underrepresented, significant errors in estimating the total market value could occur. 

A number of studies have been carried out in Germany to further explore the use of household panel data in estimating the organic market. Bien and Michels (2007) carried out a study of 500 households from a German panel, who were asked to keep additional records of their organic purchases in addition to the panel returns. The authors, on behalf of ZMP (Central Market and Price Report Office) and the panel company GfK, used these data to develop a specific methodology for fresh products to minimise misclassifications, based on checking price differences. For purchases of fresh products, a data check is used such that they are only accepted as organic if their price is higher than the specific non-organic threshold price, set on the basis of the monthly price reporting published in Germany.

A study conducted by Niessen and Hamm in 2006 analysed the mistakes made by German consumers when classifying conventional products as organic, and found that older householders without children tend to make more mistakes in this area than other respondents. The researchers found that consumers made most mistakes (around 35%) when buying the products through direct sales, on-farm sales, or markets. More than 70% of all classifying mistakes made in egg sales occurred when eggs were bought through direct marketing, with the percentage being even higher for strawberries and for potatoes. Around 5-15% of the products bought in supermarkets or discounters were classified wrongly.

Retail panels can be a valuable data source in estimating the total organic sales value. They have the advantage of covering nearly 100% of EAN-coded products and participating supermarkets. They also give a direct comparison of organic and non-organic market values, but they do not cover direct sales and may have gaps in coverage of the independent sector. Furthermore, they do not cover sales of loose fresh products that are not bar-coded. 

The case studies of Germany and France in the OrganicDataNetwork report (Gerrard et al., 2004) showed that specialist retail panels for organic shops provide data on this sector in Germany and in France.     

• Example
  In Germany, the Nielsen retail panel covers only packaged products and not all supermarkets take part. It therefore covers only approximately 30% of the German organic market. In countries with a higher share of supermarkets, such as Denmark, Sweden, the UK or Switzerland, this share would be higher but some sales channels outside the supermarkets such as natural food stores, direct sales, online sales are not covered.

A good coverage of the organic market can be achieved via a survey of retailers in countries where supermarkets and other retail outlets play a major role in the sale of organic products. Compulsory surveys will get a good coverage rate, but it may also be possible to get information from retailers on a voluntary basis: although coverage should be carefully considered in this case. Surveys may also be a good instrument to cover sales through independent shops that are not covered by retail panels.  Another option may be to study the annual reports of the supermarket chains, which give information on the total retail sales of organic products in some countries.

• Examples
  Such a survey, which includes a survey on organic products, is carried out in Denmark by the Danish Statistical Office and in Sweden by Statistics Sweden (SBC), and covers a large part of the market in these countries. The surveys are compulsory but excluding catering sales and direct marketing. In Denmark, once the surveys have been evaluated, different forms of quality checks are performed and, respondents are sometimes contacted for further questions (Statistics Denmark, 2014).
  
  In the UK, a survey of most major multiple retailers, restricted to the total turnover in the organic market (but not broken down by product categories), is carried out annually by a trusted third party and the results are one of several sources used to estimate the total value of retail sales the UK Organic Market Report, which is published annually by the Soil Association. 

Data on sales through farm shops and farmers’ markets are more commonly collected by surveys of producers, because they tend to be not covered by any other data source. Brown (2002) reviewed research on farmers’ markets in the USA from 1940 to 2000 and identified some methodological issues in carrying out research into direct sales; including the possibility of over-looking small producers. Furthermore, censuses and surveys often rely on farmers’ memory and farmers’ openness and so may be open to error (she quoted other researchers who found that farmers tend to understate rather than overstate sales). She also reviewed the various attempts that had been made to assess the economic value of direct sales, and demonstrated the large discrepancies that can arise as different methods are used to extrapolate data from a sample to state level (Brown, 2002). All of these issues apply to assessing the organic sales through farmers’ markets and farm shops with the additional problem of having to separate out organic sales from non-organic sales.

Expert estimates of the organic market, or a specific sector, can provide additional information for the purpose of cross checking of data and trends and thus can be one source for obtaining an estimate of the total organic market. However, , they should only be used as the only data source if no other data collection method (panel data or surveys) exists or can be carried out. If using experts, it is necessary to carefully select experts for specific market segments (e.g. meat, dairy, arable) and cross-check estimates against other data sources (e.g. production volume, see above). The bases of expert estimates are usually surveys of retail chains or wholesalers to get an idea of the market.

• Example
  In Estonia, the market for locally produced products is estimated based on a rather detailed survey of specialised organic shops and other shops selling organic products and organic producers/processors. The overall organic domestic market value is based on very rough estimation based on information collected from the shops and is probably not very accurate. Domestic market value data are published very late, more than a year later. As market data appear so late, the Estonian Centre for Ecological Engineering makes its own small surveys on estimated changes in retail sales from one year to another.