The authors studied food losses in four food service institutions in Stockholm, Sweden. The results show that about one-fifth of the food produced is lost. Plate waste is the single largest source of loss, at 11-13% of the amount of food served. Losses in food service institutions can be of significant economic value, and arable land equivalent to 1.5% of the area under cultivation in Sweden may be used to produce food eventually lost in food service institutions. The results indicate that the economic and environmental consequences of current levels of food losses may be substantial.

This figure is comparable to the figure calculated by Rebecca White for domestic food waste. See Reductions in the energy intensity of the UK food chain: where should they be made, and would consumers support them? Rebecca White, 3 September 2004, Environmental Change and Management, University of Oxford, Masters Dissertation.

The study looks at a selection of foods: pork, rice, dried peas, tomatoes, carrots and potatoes. It finds that pork and rice produced the highest GHG emissions, followed by tomatoes. Dried peas were the least GHG-intensive of all the foods. It also notes that CO2 and energy use is an inadequate measure of food chain emissions since methane and refrigerant emissions are very considerable. It notes that a 'critical' life cycle stage for one crop may not be so critical, relatively speaking, for another. The study also points out that for food, the functional unit of measurement is very important. It compares four sample diets made up of various combinations of the foods studied to produce diets classed as: exotic non-vegetarian (pork, rice tomatoes), domestic non-vegetarian (pork, potatoes, carrots, peas) exotic vegetarian (rice, tomatoes, peas) and domestic vegetarian (peas, carrots and potatoes). Carlsson-Kanyama then measures GHG emissions in terms of kg of protein, beta carotene and calories that the four different diets provide. The study concludesthat the domestic vegetarian diet produces the lowest level of emissions for the highest level of nutrients, followed by the domestic non vegetarian diet. Seasonality is highlighted as an important indicator of GHG sustainability.

The report covers the following:

1. The Swedish supply chain and the role of different actors - this is fairly similar to the UK situation.
   
2. The perceptions of and use of environmental information in the food supply chain by purchasers
   
   • the report distinguishes between commercial and public food service institutions (eg. local authorities)
     
   • it looks at four different approaches to environmental info - finance oriented, regulation oriented, customer oriented and quality control oriented - and shows how these limit environmentally-preferable choices
     
   • it suggests that negative info (ie this product is bad for the environment) can be more effective in altering purchasing behaviour than positive info (this is good for the environment) although in most cases price dominates decision making
     
   • it highlights the difficulties of environmental decision making given the fragmented nature of the food supply chain (ie. everyone is responsible for his/her patch and no more) and the various price/regulatory/quality etc pressures some of which can work against the environment
     
3. It highlights LCAs of six different products - tomatoes, carrots, onions, broccoli, pulses and frozen chicken - and finds that there is no simple correlation between distance and environmental impact
   
4. It explores three different scenarios for the way the food system could go – 'Pricehunter', 'Green and global' and the 'Eco-friendly farmers' market' and looks at the environmental implications of these.

The study compared differences in electricity consumption both by the different food types, and by differences in portion size (cooking for one or four portions) and cooking appliance (oven, hob, microwave, kettle). The study also estimates the energy use in the life cycle stages before cooking. It concludes that energy use for cooking can be a large or dominating part in the energy use for some foods, usually those of vegetable origin with a low to medium degree of processing. Electricity use is greatly influenced by the choice of cooking appliance and the number of portions cooked. For example a microwave oven can be up to ten times more energy efficient than a conventional oven for baking potatoes. It is always more energy efficient to cook several portions at the same time. The study concludes that there is great potential for improving the energy efficiency of diets by adjusting the choice of food ingredient and the cooking method.