Competitive Liberalization:

1

Competitive Liberalization A Tournament
Theory-Based Interpretation Simon J. Evenett,
www.evenett.com
2
Competitive Liberalization

• Stimulating a competition for access to the U.S.
  market.
• Greater role for foreign and security policy in
  U.S. trade policy-making.
• Inclusion of non-market access provisions in
  trade agreements.
• Some preliminary comments.

3
Rank-order tournament models

• Examine incentives created by relative
  performance systems are used to allocate scarce
  top slots.
• Basic 2-period tournament model (Lazear 1995)
• Employer announces wages W1 and W2.
• Employees choose unobserved effort levels. Then,
  employer assigns employee to jobs according to
  relative performance.

4
Adapting tournament models

• USA runs a contest for 2 trading partners,
  solicits market access offers from them, and
  awards 2 PTAs.
• Compensation is in form of extent of access to
  the USA marketnot financial compensation.
• All three countries make politically painful
  market access offers.
• As contestants market access offers are
  observed, what is the source of uncertainty?
• Market access offers can be financed in 3 ways.

5
The model

• USA announces its intention to offer 2 PTAs at
  end of period 2, giving the most valuable PTA
  (worth W1) to contestant with highest
  foreign/security policy-adjusted market access
  offer. Loser gets a different PTA worth W2.
• In period 1 USA does not know what foreign policy
  concerns will be important in period 2 but
  commits to take them into account.
• Contestants choose market access offer before
  foreign policy outcomes are known. Once the
  latter are known, USA awards W1 and W2.

6
Solving the model...backwards
1 receives W1 if
FOC for 1
7
Symmetric Nash Eq in period 2
IC
PC
IC, PC ?
8
Solving the model in period 1
FOC
9
Properties of the equilibrium

• Equilibrium market access offers (m) unaffected
  by foreign policy considerations.
• Moreover, the equilibrium market access offers
  (m) are the same as in a deterministic benchmark
  case where no foreign policy considerations were
  present.
• But maintaining discretion for foreign policy
  matters as
• g(0)??W1-W2?
• W2lt0 if

10
Designing PTAs to reduce W2

• Think of W2 as being the sum of two components,
  ?Y ?Z
• The USAs agreement with the losing contestant
  (?Y)
• More US sectors exempted from its market access
  offer.
• Longer transition periods before market access
  improves.
• Delay implementation of this agreement after that
  of PTA with winning contestant (that is,
  sequencing).
• Include controversial non-market access
  provisions.
• Caveat Possible limits to reducing ?Y namely,
  ?Ygt0.
• The USAs agreement with the winning contestant
  (?Z)
• Ensure losers exports fall with this agreement.

11
Those findings again...this time in English.

• This tournament enables the USA to include
  foreign policy considerations at no cost in terms
  of market access elicited from the trading
  partners.
• Not every tournament has this feature.
• Foreign policy considerations matter, though, for
  the design of the PTAs offered and can account
  for sequencing, exemptions, and inclusion of
  deliberately controversial non-market access
  provisions.
• If retaining discretion for foreign policy
  considerations becomes large enough, the losing
  contestants politicians utility may actually go
  down, even though its PTA with the USA raises its
  politicians utility.

12
Comparative statics
13
Does a tournament perspective add value?

• Emphasises relative as well as absolute treatment
  of contestants encourages analysts to compare
  PTAs, not just evaluate single PTAs.
• Points to the role that non-trade policy
  objectives can play in determining relative
  treatment of contestants.
• Highlights the importance of the extent of
  domestic pain is needed to finance a market
  access offerwith the suggestion that there is a
  clear incentive to exploit opportunities for
  trade diversion.
• Provisions of PTAs and the sources of financing
  market access are not explicitly modelled.
• Departures from symmetry and risk-neutrality are
  feasible but messy.