By Judith Friedlander, University of Technology, Sydney
Are soy milk’s environmental attributes based on substance or froth? Is soy a sustainable solution in the dairy debate?
Comparative environmental analysis of different food groups is like comparing, well, apples and oranges.
Reports such as the Food and Agriculture Organization’s (FAO) Livestock’s Long Shadow document the negative environmental impacts of cattle and dairy production and consumption.
If cattle are major emitters of the intensive greenhouse gas methane and use large amounts of water, are protein alternatives such as soy less harmful to the environment?
Soy, of course, is much more than soy milk. According to the CSIRO, “60 per cent of all products at the supermarket already contain soybean”. (Look at the ingredients lists on breads, flours, oils, pet food and sausages, for example.)
The Federal Department of Agriculture, Fisheries & Forestry’s (DAFF’s) Australian Food Statistics (2011) report states that Australia produced 47kt of soybeans in 2010-11 and imported $36M worth of oilseeds in 2010-11 (soybean is classified as an oilseed rather than a pulse).
According to Dr Andrew James, who leads CSIRO’s soybean breeding program, the vast majority of whole-bean soy products Australians consume – such as tofu, soy milk and many baking products – derive from Australian grown beans. However, the soy components used in processed foods generally come from imported soy protein powder, isolate or lecithin.
“We also import 500 to 800,000 tonnes of soybean meal. A crude guess is that would cost about $600 a tonne. That’s $500 million or more,” says Dr James.
Australians have been developing a taste for soy milk over the last decade. A Soy Australia report showed in 2009 Australians drank three litres each of soy milk a year. This is a 50% increase on 1998.
The report also claimed that most of the soy milk production in Australia is based on imported soy protein or soy protein isolate, a refined form of soy protein made from defatted soy flour.
Says Dr James: “If the milk is made from whole beans it will be Australian-grown beans that are used. If made from protein powder or isolate then that will be imported.”
In 2010-11, Australians drank 2296 ML of “dairy milk”, and 2061 ML in 2005-06. The annual average 2005-07 milk consumption per Australian was 230kg (approximately 223 litres).
Soy milk, soy-based drinks, soy dairy-free products and energy bars showed the strongest growth. Traditional foods like tofu have seen a decline in the past couple of years.
Australian data on the environmental impact of the dairy and soy industries is not as comprehensive as from other countries such as Sweden, Denmark and the United Kingdom. While there are international differences in production, life cycle analysis research is informative.
A recent Swedish study showed a span of 0.4 to 30kg of CO₂ equivalents produced per kilogram for different food items. The lowest emissions per kilogram were from legumes, poultry, and eggs. The highest were from beef, cheese and pork.
Cornell University scientist, David Pimentel, has found it takes about 14 kilo-calories (kcal) of fossil-fuel energy to produce 1kcal of milk protein using conventional milk production. Organically produced milk might require a little less than 10kcal of fossil-fuel energy per kcal.
In comparison, Pimentel’s data suggests that in a conventional soybean production system, one kcal of fossil energy invested produces about 3.2kcal of soybean. For 1kcal of fossil energy invested in organic soybean production, you get an average of 3.8kcal of soybeans. This means it takes between .26 and .31kcal of fossil fuel to make 1kcal of soybeans (contrasted with 10-14kcal to make 1kcal of dairy milk protein).
Pimentel states that soy protein accounts for about 35% of those kilocalories, so it appears that making soy protein is more energy-efficient than dairy protein.
Soy milk, of course, is the sum of its parts and whether using ground soy beans or soy isolate, other ingredients are added to make the liquid that consumers use in their coffees and cereals. These extras include calcium and oil. So the production process and its energy and water components need to be considered.
A 2010 “current and possible futures” study into greenhouse gas emissions across the top 45 food commodities in the UK recommended dairy milk and products be replaced by soy-based milk products.
A recent Dutch study comparing the water footprints of soybean and equivalent animal products found that soy milk and the soy burger have much smaller water footprints than cow milk and the average beef burger. The water footprint of the soy milk products analysed in this study was 28% of the water footprint of the global average cow milk. The water footprint of the soy burger examined was 7% of the water footprint of the average beef burger in the world.
Another important environmental parameter to consider is how much phosphorus is used to produce food. Modern agriculture is dependent on phosphorus derived from phosphate rock. It’s a non-renewable resource, and current global reserves may be depleted in 50 to 100 years. Meat and livestock production are associated with high phosphorus use and a vegetarian diet demands significantly less phosphate fertilizer than a meat-based diet.
The genetic modification (GM) debate is a series on its own, but it is relevant to note that all soybeans grown in Australia are GM-free, according to CSIRO’s Dr James. He compares this to the US, Canada, Argentina and Brazil, where over 90% of the soybeans grown are genetically modified.
Dr James says that if the soy milk bought in Australia states it is made from whole soybeans, it is “most likely” made from non-GM Australian soybeans. If the label says it is made from soybean isolate, it is “most probably imported from the US”.
Judith Friedlander does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.