Operation of the SPS Agreement

1
Operation of the SPS Agreement

• Lecture 38
• Economics of Food Markets
• Alan Matthews

2
Key questions

• What does the resolution of SPS disputes tell
  about the interpretation of the SPS Agreement?
• How are developing countries affected by rising
  food safety standards in developed countries
• Are rising food safety standards a competitive
  burden on EU food producers and farmers?

3
Some Cases

• Cases which went to Dispute Settlement
• EU beef hormones
• Tasmania (Australia) salmon
• Japan apples
• EU GMOs
• Cases settled through the SPS Committee
• EU aflatoxins

4
Primer on WTO dispute settlement

• Issues first raised in relevant Committee for the
  agreement (Committee on Agriculture, SPS
  Committee)
• Country can call for consultations
• Country can lodge complaint
• Heard by Panel
• Appellate Body
• Country can request arbitration in the event of
  partial compliance.

5
EU hormones dispute

• One of the first to be taken under the new SPS
  Agreement
• Roberts (1998) provides a comprehensive summary
• National treatment was not an issue
• EU did not claim defence under Art 5.7
  (precautionary principle)

6
EU hormones dispute

• Panel found that EU ban was inconsistent with
• Art 5.1 (not based on risk assessment)
• Art 3.1 (not based on international standards) or
  Art 3.3 (no scientific justification for claim
  that ban led to higher protection)
• Unjustified and arbitrary protection which varied
  from protection provided by other EU measures
• Burden of proof issue?

7
EU regulatory framework for GMOs

• Put in place in the early 1990s to protect
  citizens health and the environment while
  creating a unified market for biotechnology
• Authorisation requires a risk assessment
• Since entry into force of Directive 90/220 18
  authorisation were approved for commercial
  release, but none since October 1998. Five MS
  said they would refuse approval until new
  regulations on labelling and traceability were
  introduced.
• Some member states invoked the safeguard clause
  in the Directive to temporarily ban the
  marketing of GM maize and rapeseed.

8
The EU and GMOs

• If trade restrictive measures are motivated by
  concerns over super-weeds or food safety, then
  the SPS Agreement applies
• have the risks been assessed, does scientific
  evidence justify the restriction, is the
  appropriate level of protection consistently
  applied, is it minimally trade distorting?
• If mandatory labelling is justified by the
  consumers right to know, then the TBT Agreement
  applies
• the US contests the need for this

9
The WTO Panel

• Proceedings began in 2003 Panel report
  unofficially released in March 2006
• US, Argentina and Canada have complained that
• A de facto moratorium on GM approvals, since
  1998, had no scientific justification
• Four Member States (Austria, France, Greece and
  Italy) banned GM products that had been approved
  by the EU

10
The WTO Panel

• The measures at issue
• The general moratorium, i.e. suspension of
  approvals
• Product-specific moratoria or marketing bans
• Member states national measures prohibiting the
  marketing of GMOs

11
WTO panel findings

• EUs moratorium violated WTO rules because it led
  to undue delay in assessing marketing
  applications for GMOs, contrary to Art. 8 of the
  SPS Agreement
• Similarly for the product-specific measures
• Member State bans violated WTO rules because they
  were not based on a risk assessment
• Panel did not question parties right to conduct
  pre-market risk assessment of GMOs
• Panel did not consider whether GMO products are
  like non-GMO products and can be treated
  differently

12
Developing country issues

• Significance of SPS standards as barriers to
  trade
• Cost of meeting SPS standards
• Limited input into the design of standards
• The cost of the standards themselves
• Certification
• Transparency and technical assistance

13
EU Aflatoxins

• January 1998 notified the SPS Committee of its
  plans to introduce new legislation
• No Codex standard at the time
• Impact on developing countries potentially severe
• e.g. Ghana pointed out that 80 of its exports
  were of groundnuts, and that the impact on trade
  could be severe
• World Bank suggested it could halve imports of
  nuts and cereals from Africa for a trivial gain
  in EU food safety
• EU made some changes, but many developing
  countries still profoundly unhappy

14
SPS measures and consumer protection

• Traditional trade measures were taken to protect
  producers easy to show under standard
  assumption that trade measures reduce welfare
• SPS measures often take in response to consumer
  concerns the welfare effects can be very
  different
• Consider case of ban on GMFs (genetically
  modified foods) where consumers have preference
  for non-GMF product (Gaisford and Chui-Ha, 2000).

15
The model (Gaisford and Chui-Ha)

• Two country world, Europe and North America
• Free trade prior to introduction of new GMF
• New GMF developed in North America
• Europe prohibits domestic production of the GMF
  and continues to produce only non-GMF
• Assume Europe small relative to North America
• Assume that European welfare only depends on
  quantities of GMF and non-GMF directly consumed
  as private goods (i.e. no externalities)

16
The model

• GMF is perceived in Europe as a low-quality
  substitute for the non-GMF
• In the absence of credible labelling, individual
  consumer cannot determine whether food is GM or
  not we have a pooling equilibrium
• GM technology reduces cost of production in
  supplying country, resulting in fall in world
  price

17
Price
Snon-GMF
Initial equilibrium before the introduction of
the GMF variety
Initial non-GMF world price
Dnon-GMF
Quantity
18
Price
Snon-GMF
New equilibrium following introduction of GMF
product
Initial non-GMF world price
Pw
Final GMF world price
Pf
Dnon-GMF
Dpooled
Quantity
Domestic output ?
Domestic consumption ?
19
Price
Snon-GMF
Welfare changes following introduction of GMF
product
T
Pe
V
U
Initial non-GMF world price
X
W
Pw
Y
Z
Final GMF world price
Pf
Dnon-GMF
Dpooled
Quantity
Qe
20
Welfare impact of introduction of GMF

• Demand curve shifts downward because of decline
  in average quality
• Loss of consumer surplus (TVX) (adverse
  quality effect)
• Increase in consumer surplus (YZ) loss of
  producer surplus Y (net price effect)
• If adverse quality effect dominates, European
  welfare falls.

21
Can EU improve its welfare with an import ban?

• Only non-GMFs remain available and no adverse
  quality effect arises
• However, a harmful price effect arises
• Non-GMF imports are non-available, price rises
  from Pi to Pe
• Producer surplus rises UV, consumer surplus
  falls (UVWX)
• Fall in EU welfare (WX)
• But fall may be less than allowing unlabelled GMF
  imports Z-(TVX)
• Embargo is superior is TV exceeds WZ

22
Is mandatory labelling a superior option?
Price
Snon-GMF
Pe
A
B
C
Ps
G
E
F
Initial non-GMF world price
Final GMF world price inc. labelling cost
H
Pf
Dseparate non-GMF
Ddemand separate GMF
Dnon-GMF
Quantity
Qe
Qs
There are now two separate markets for
conventional and GMF products
23
Is mandatory labelling superior?

• Start with embargo on GMFs welfare loss is
  CFG
• Mandatory labelling gives rise to a separating
  equilibrium EU consumers now have a choice
• Advent of GMFs will create a second market
• Availability of GMF will shift the demand curve
  for non-GMFs because of availability of
  substitute product

24
Is mandatory labelling superior?

• Start with non-GMF market 1. Raise price to Ps
  assuming GMF price is infinite (i.e. prohibited).
  Relevant demand curve is Dnon-GMF.
• Welfare change CS (EFG) PS (E)
• Gain from new product H
• Overall gain is H (FG)

25
Compare with import embargo

• Adverse price effect is smaller with mandatory
  labelling -gt smaller welfare loss on non-GMF
  market by C
• Also gain on GMF market of H
• Mandatory labelling unambiguously better than
  embargo on GMF

26
Conclusions

• Mandatory labelling may still be challenged under
  WTO because it imposes large costs on exporters
  to develop Identity Preservation Systems
• Could evidence of consumer preferences be
  used/required as defence of labelling?