Priorities for the Economics of International Agricultural Trade Presentation at the CFARE Workshop

1
Priorities for the Economics of International
Agricultural TradePresentation at the C-FARE
WorkshopSan Antonio, TexasNovember 6, 2003

• Daniel A. Sumner
• University of California Agricultural Issues
  Center and Department of Agricultural and
  Resource Economics, UC Davis

2
Stakeholder priorities for competitiveness and
profits

• Farmers and others have the priority to become
  more competitive and profitable.
• Competitiveness and profit involves technology,
  management, marketing and a host of issues to
  which economists can contribute research,
  teaching and outreach.
• International trade economics has some
  contribution to make on this goal, but is
  relevant also to other stakeholder goals.
• Thus, my title is broader than that assigned.

3
Priorities for the Economics of International
Agricultural Trade

• Broader than contributing to competitiveness and
  profit in US agriculture.
• Competitiveness and profit are positive for
  farmers and other businesses and generally
  positive for the economy, but parts of US
  agriculture may be best served by learning to
  exit rather than attempting to compete.
• Furthermore some firms and industries may be
  maintained in profitability only through public
  investments or policies that are not profitable
  for the economy as a whole.

4
Public economic priority must remain on broad
social benefits

• Most often using economics to contribute to
  competitiveness is consistent with general social
  benefits.
• In fact, economics can illuminate where this
  consistency holds and where it may not.
• From a broader perspective we can provide
  information that will help farmers and others
  shift resources to areas that are likely to
  competitive such that agriculture as a whole
  gains profitability and competitiveness.
• This may not satisfy each commodity constituency.

5
International issues are pervasive and interact
with most other topics for which economics is
important

• International dimensions apply to any economics
  that also applies in other countries, which is
  most economics.
• Food safety and nutrition issues,
• Technology development and transfer,
• Rural development, and
• Environmental and resource issues
• all have international components.

6
Priority topics for the economics of agricultural
trade

• International trade policy (this has long been
  the core topic for trade economists)
• Analysis for trade negotiations
• Analysis for trade dispute settlement
• Analysis for policy reform consistent with trade
• Projections of agricultural competitive positions
  for different locations
• Cost comparisons
• Basis for economic advantages in different
  regions
• Global competition policy
• Foreign direct investment and trade flows

7
A Trade policy economics has been used more
fully every year

• Economics is used because it is useful for
  several phases of trade negotiations and trade
  disputes.
• Nonetheless, many economists are frustrated that
  economics is not more influential in the
  resolution of disputes and negotiations.
• But, is our analysis is ready to play an even
  larger role?
• How can we be more relevant and accurate to be
  more useful?

8
Uses of trade policy economics in negotiations

• Set the general understanding of the benefits and
  costs of more open markets
• Project outcomes of specific options and
  proposals
• Provide ex post projections of actual negotiated
  agreements to guide understanding of what to
  expect
• Assess the outcomes and effects of previous
  agreements or dispute settlement
• Each of these requires different approaches and
  must not be confounded.
• All are difficult

9
General quantitative sense of impacts of reform

• 1. The idea is to give the flavor and rough idea
  of implications of various policy reforms
• Stylized policies and proposals
• Estimates using retrospective counterfactuals
• Aggregate summary results
• Improvement on the gross misrepresentations of
  non-economic approaches and rhetoric
• Not detailed or specific enough to guide
  negotiations, but can motivate policy directions
• CGE models and most academic studies

10
Detailed projections of specific options

• 2. With sufficient policy detail, projections can
  guide choice of negotiating options.
• Often requires short turn-around and rapid
  communication to negotiators.
• Requires comprehensive understanding of policies,
  markets and alternatives
• Requires underlying research and models that can
  be adapted quickly.
• CGE models are not yet adaptable for this work
• Academic-style research supplies data, parameters
  and model development, usually not detailed
  projections relevant to negotiation process

11
Projections of impacts of negotiated policy reform

• 3. Can guide expectations and plans of market
  participants and may influence implementation
• Requires comprehensive understanding of policies,
  markets and alternatives
• Requires underlying research and models that can
  be adapted to the detailed specification of the
  actual agreement.
• Most academic-style research models do not have
  sufficient policy detail and are not adapted to
  specify complex policy changes.

12
Estimation of outcomes attributable to specific
policy reform

• 4. Most reforms are partial and normal market
  flux makes isolating of policy reform
  complicated.
• Requires careful specification of expectations of
  real policy changes real changes
• Most expected market shifts are gradual and small
  relative to agricultural market variability
• Requires marshalling as much data as can be found
• Econometrics must be developed carefully and
  creatively
• Can be useful for understanding of future policy
  change

13
Empirical economics may be more used and useful
in trade dispute settlement

• Trade dispute resolution applies detailed
  economic analysis for proceeding at USITC, WTO
  and related forums
• A high proportion deal with agriculture or NR
• This will spread globally from North America
• Already there is more economics in the WTO
  dispute settlement process than a few year ago
• We need more development of standard tools and
  standards for how to evaluate the economics
  applied by staff and parties

14
Elements to improve applicability of empirical
trade research

• Model approximations of policies and alternatives
  carefully
• All models rely on approximations, but this is
  not the place to cut corners for useful analysis
• Gross approximations of policy are helpful
  pedagogic tools, but may be more misleading than
  helpful for empirical trade policy research
• Seemingly small differences can have significant
  impacts
• Domestic support schemes in agriculture provide
  many examples

15
Elements to improved applicability of empirical
trade research

• Understand details of commodity markets and
  relationships
• Even definitions of commodities may not be
  obvious
• Is wheat wheat? (Is durum a different product?
  How closely related are HRS and HRW?
• When can indica and japonic rice be aggregated?
• Similar aggregation questions apply to markets,
  how integrated is India in the world dairy
  market?
• These are research questions that can only be
  approached with a thorough empirical research
• Econometric evidence is often lacking and
  difficult to develop

16
Elements to improve applicability of empirical
trade research

• Parameter values drive model results
• Model complexities may hide reliance of results
  on a few supply or demand elasticities
• Yet, we continue to have problems conceptualizing
  appropriate parameters and how they might be
  defined and obtained
• Parameters depend on the specific market
  situation and policy question. E.g.. it makes no
  sense to ask what is the supply elasticity for
  wheat.
• Much econometric effort has not helped
• These are basic questions to which we must devote
  more systematic and careful thinking

17
Econometrics alone will not yield estimates of
policy-relevant supply response

• Government programs and expectations about the
  programs affect the relevant supply response
  parameters. As policies change or when we
  investigate new policies, we must use parameter
  estimates that are structural and not functions
  of the policy in place.
• Econometric estimates from the literature will
  seldom have this property.

18
Defining the counter-factual baseline projection
may be crucial

• Even comparative results are not invariant to
  baseline projections under the alternatives
• Effects of TRQs and deficiency payments depend on
  projections of world and domestic prices
• This means investing more in baseline projections
  to understand impact of policy reforms
• ERS, FAPRI and very few others make these
  investments and few academic resources are
  devoted to the critique

19
We may only attempt to project Pb - Pa, rather
than Pb itself. Still, relative impact of
removing a, for example, deficiency payments
(marketing loan or C-C payment) depends on the
baseline. The same applies to the impact of
removing a TRQ. When comparing different
internal prices this implies that long term
exchange rate projections are required!
Pb
Projection
Pa
Current policy
T1
T0
20
B Projections of regional production patterns
and competitive positions

• Which industries are on the competitive edge and
  what will keep them there?
• This requires considering opportunity costs of
  resources in each location.
• The peach industry in California competes first
  with tomatoes and almonds in California and then
  with the peach industry in Greece or China
• Assessments of cost of production in various
  regions can help, but these are dangerous and
  very difficult to get right.
• Budget measures are fraught with error and a
  problematic track record.

21
Projections of regional production patterns and
competitive positions

• It is easy to miss advantages or linkages that
  give an industry an edge
• Quality
• Reputation
• Existing business relationships
• Economists have a difficult time analyzing these
  factors and others.
• Must be careful not to oversell our capabilities
  to compare costs and assess competitiveness

22
C Global competition policy

• The joint research in international trade and
  industrial organization is well underway
• Practical findings and useful empirical or
  simulations are more difficult to master
• Using countries as the unit of analysis is less
  interesting than considering firms
• Traditional questions of monopoly power and cross
  border analysis of mergers combines these
  sub-fields
• With consolidation globally in both farm supply
  and marketing these issues become more relevant

23
D Firm relationships driving trade rather than
governments driving trade

• When firms cross borders this may have important
  implications for choice of suppliers
• Thus, foreign direct investment may affect trade
  in goods and service
• Perhaps a Wal-Mart produce supplier in the U.S.
  will also have an advantage supplying produce to
  Wal-Mart in Mexico.
• Perhaps firm-based quality standards or
  firm-based rules will dominate government-based
  standards
• COOL may become irrelevant, but not yet

24
Concluding comments

• Services provided by economists will continue to
  be demanded in international trade policy
  analysis
• We have a long way to go to improve our
  performance
• Economics can inform parties of projected
  economic impacts, decisions are based on
  interests and economics can make these more
  transparent
• Projections of competitiveness require detailed
  specific information and remain troublesome, with
  much capacity to mislead
• Research linking trade with industrial
  organization may be growing in importance
• The core demands for economics remain trade
  policy