GENET archive


6-Regulation: Is the new Swiss GE food labelling law WTO-consistent?

-------------------------------- GENET-news -------------------------------

TITLE:  GM Swiss cheese and wine in 2005?
SOURCE: Martin Trancik, Switzerland, posted by checkbiotech
DATE:   12 Jan 2005

------------------- archive: -------------------

GM Swiss cheese and wine in 2005?

BASEL - In Switzerland, new labelling regulations concerning GM food will
apply from February 2005. It seems to be doubtful, however, whether the
Swiss Government took into account existing international trade
agreements within the WTO.

In his Syngenta lecture, called "New Swiss Labelling Rules on GM Food:
Consumer Free Choice v. Business Interests", Daniel Wüger from the World
Trade Institute in Berne began by outlining the new regulations
introduced by an ordinance of the Swiss Federal Council entering into
force on February 1, 2005.

The general answer to the question what will have to be labelled as GM
food in Switzerland leads to the following answer. Labelling requirements
are triggered by genetically modified organisms themselves as well as by
food products obtained from such organisms. The latter will have to be
labelled as GM no matter whether they do contain GM-material or not. As
an example, genetically modified soy will have to be labelled as well as
oil or fat derived from GM-soy. Refined oil or fat, which do not contain
any DNA, also fall under the labelling requirement. One processing step
further down the line, a product such as soy lethicin, derived in turn
from soy oil or fat and equally free of any DNA as refined oil or fat,
will also be caught by the new labelling legislation.

On the other side, food not obtained from genetically modified organisms
(GMOs), but which may have been produced with the help of such organisms,
will not have to be labelled. Applied again to a real life example such
as corn, this regulation produces the following results. If the original
product is genetically modified corn, the whole processing chain from
corn starch turned into glucose syrup used for the production of sorbit
and invert sugar will have to be labelled as GM. If this process,
however, is initiated from conventional corn, no labelling whatsoever
will be required. The regulation, in other words, does not impose
labelling requirements although in both cases, that is even in the
processing of conventional corn, the techniques used to produce glucose
from starch and sorbit / inverted sugar from glucose may require the use
of GM-Amylase and of genetically modified enzymes. The application of
genetically modified enzymes to glucose obtained from conventional corn
in order to produce sorbit or inverted sugar does not turn the latter
products into GMOs.

Another decision, which in the current climate of scepticism towards
genetic modification will come as a relief to its proponents, deals with
the problem of food derived from animals fed with GMOs. This, of course,
solves the notorious problem whether milk from cows, who had been fed gm
crops should, in the name of consumer choice, be labelled as GM or not.
Here, the new regulation does not introduce any change: neither milk nor
cheese obtained from such milk will fall under the labelling
requirements. Milk from GM-fed cows will, therefore, be treated in the
same way as milk from conventionally fed cows.

A third area important to food processing is, of course, fermentation.
Experts will know that this concerns not only beer and wine, but also
vitamin production. Here, the Swiss regulation takes a rather strict line
above all when compared to the parallel legislation in the EC. In
Switzerland, GM-labelling will be required as soon as the substrate for
fermentation is so much as obtained from a GMO but also when the micro-
organism used for fermentation is genetically modified. In neither of
these two cases does the EC require labelling, although this may change
at least with regard to the use of GM micro-organisms. It is,
incidentally, interesting to note that the regulation seems to view
micro-organisms with greater distrust than macro-organisms. As seen, milk
given by GM-fed cows (macro-organism) will not have to be labelled
whereas fermentation products produced with the help of genetically
modified micro-organisms will be seen as a whole as genetically modified.

WTO Implications

In a country relying as heavily on international trade such as
Switzerland, regulations concerning food are bound to have some sort of
WTO-implications. These were the topic of the second part of Dr. Wüger's
lecture. Within the framework of the World Trade Organisation, two sub-
agreements might cause problems for the Swiss.

First of all, there is the agreement concerning so called SPS-measures,
that is, sanitary or phytosanitary measures. This agreement tries to
solve a basic conundrum which could be summarized as follows: How can a
country ensure that it is being supplied with food considered safe by
that country's own standards without, at the same time, using strict
health and safety regulations to protect domestic producers from
international competition?

At first sight, Swiss national regulation concerning GM-food does not
even seem to fall under the SPS-agreement, given that at the stage of
introduction to the market, health and safety measures will already have
been dealt with. After all, the way the regulation is worded, certain
important categories of final products will have to be labelled as GM
although they will be identical with the respective conventional
products. Therefore, the Swiss measure doesn't seem to have any health
implications whatsoever. As Dr. Wüger noted, "The objective of the Swiss
measure is really to provide information solely. A final product which is
the same as the conventional product doesn't have any special health

But it is, of course, a well established fact that consumer distrust
towards GM food is heavily inspired by health concerns, as reasonable or
unreasonable as they may be. GM-labelling laws may therefore disadvantage
foreign GM food imported to Switzerland, which in turn would be illicit
under the SPS-agreement: there are simply no health and safety issues
around GM-food which might justify a SPS-measure. Dr. Wüger pointed out
that as a dispute-settling body, in case of a conflict, the WTO would
have to decide whether to understand labelling regulations as health
measure or not. Dr. Wüger then posed the question, "How do you want to
deal with that question under WTO-law? Such a question has never arisen
with that regard."

Another WTO sub-agreement, however, may be closer related to the
introduction of GM-labelling laws: the agreement dealing with technical
barriers to trade, the so called TBT-agreement. The main purpose of the
TBT-agreement is to ensure that regulations or standards in the broadest
meaning of the term are not set arbitrarily in order to be used as an
excuse for protectionism.

Three criteria must be fulfilled for a TBT-measure to be legitimate.
First of all, the measure is required to have a legitimate objective.
This does not include consumer information. The prevention, however, of
deceptive practices is considered legitimate under the TBT-agreement.
With regard to this particular requirement, the new Swiss GM-labelling
regulation will probably not encounter serious problems: It could be
argued that preventing consumers from knowing how food has been produced
may be deceptive towards people objecting to genetic modification. On the
other side, Dr. Wüger pointed out that, "There is not much jurisdiction,
there haven't been many cases on that. What deceptive practices are, is
pretty much up for grabs."

The second requirement, however, is likely to cause more problems when
applied to the Swiss GM labelling regulations: A TBT-measure must not be
discriminatory. The criterion used to establish whether regulations or
standards discriminate against one product, while favoring another
equally competitive one, is likeness. As soon as two products are
considered to be alike, discrimination between them is forbidden. Some
sub-criteria by which "likeness" between products is measured could cause
some headache, such as the notion of "product characteristics". The
problem here, of course, lies in the fact that, say, starch derived from
conventional corn does not show any physical difference when compared to
starch produced from GM-corn. Even when additional sub-criteria for
"likeness" such as, "identity of end users" or "consumer preferences" are
thrown in, for those who are in favour of genetic modification in food
production, or who simply don't care, both varieties of starch will be
alike. Labelling GM-corn could, therefore, be considered discriminatory
argued Dr. Wüger, "I think there are good reasons to say that these
products are competitive and so you do have to treat them alike. And that
is what Switzerland is not doing, that's what the EC is not doing."

The third requirement, finally, by which the legitimacy of a TBT-measure
is established, invites classic lawyerly debating: the measure must, of
course, be proportionate. A balancing of interest has to be made between
the importance of the objective, the measure's impact on imports and
exports and the degree to which the measure actually contributes to the
objective. In case of a conflict, the WTO would therefore have to balance
answers to questions such as, how important it is to tell the Swiss that
parts of what they eat have been genetically modified, to what degree do
Swiss consumers actually read the labels on food packages and then allow
this information to inspire their choices and how the new regulation
affects other countries food exports to Switzerland. Dr. Wüger added,
"Proportionality, I think that's the principle under which the Swiss
measure will have most problems to be justified."

Dr. Wüger closed his analysis by pointing out that the Gene Technology
Law is based on promises in its article 17.6 that the Federal Council
would take into consideration international recommendations and external
commercial relations when passing more detailed legislation such as the
labelling ordinance. But it is doubtful, however, whether the Swiss
Government took into account existing international trade agreements
within the WTO. Dr. Wüger remarked, "I do think the Federal (Swiss)
Council is not doing that adequately. It is kind of strange that the
administrative statement to this ordinance...does not contain any word
about WTO-standards."

Labelling regulations are a highly complex and sometimes contradictory
affair, because they are an area where science, politics and prejudice
meet. No one can seriously doubt that it is legitimate to craft a law
defining what does and does not constitute genetically modified food. At
the same time, there can't be any doubt that labelling has become a tool
in the hands of those who oppose genetic modification on grounds of
principle. On the other side, the onus to inform clearly lies with the
food industry. Unfortunately, people often do not understand enough about
science, and thus governments will always have to take into account wide
spread misconceptions.

Martin Trancik is a studying Law at the University of Basel, Switzerland


European NGO Network on Genetic Engineering

Hartmut MEYER (Mr)
In den Steinäckern 13
D - 38116 Braunschweig

P: +49-531-5168746
F: +49-531-5168747
M: +49-162-1054755
E: coordination(*)
W: <>

   GENET-news mailing list