Instututions & Policy
Preferences In -- Policies Out
Preferences Government Policies
(interests) (legal constraints
on economic or
• Issues emerge,
• Interests (preferences) are formed, and
• Information is transmitted to the
• Institutions of government, where policy may or may not change.
Pivotal Politics Game
Two Strategies to Pass Laws:
Get two-thirds of each House; or
Get a majority in each House and president.
Policy change requires large majorities;
There are numerous points along the process where
legislation can be held up;
Policy will therefore be a compromise among key
Policy will change incrementally, if at all,
and will satisfy many competing interests.
Policy Making Via Delegation
But a lot of policy is made not by Congress
Rather, Congress delegates authority to the
president or regulatory agencies.
Environmental Protection Agency
Food and Drug Administration
The same mechanisms that make it hard to
pass policy, make it hard to check agencies.
Why does Congress ever delegate
When it does, how and under what
conditions will Congress delegate
What are the implications for policy?
– Save time/Reduce workload
– Take advantage of agency expertise
– Protect special interests
– Shift the blame
Two factors that affect congressional
Legislation is complex
– But legislators have only limited time and
Bureaucrats are experts
– Legislators can solve this problem by
delegating authority to regulatory agencies.
– So Congress delegates to take advantage of
– Policy will be well informed and reflect the
technical expertise of agencies.
Process creates legislative logrolls with lots of pork.
For example, the 1988 OTCA.
Inefficiencies arise because:
Each individual equates the marginal benefits and costs for his
While ignoring negative externalities imposed on other
If these externalities are small Congress may not care.
If large, imposing high costs on taxpayers and consumers, then
legislators may want to reduce costs.
– Example: 1930 Smoot-Hawley Tariff Act
Delegate authority to a central agent who
internalizes both costs and benefits.
Trade policy (first cut)
Congress solves collective dilemma by delegating to
President has a national constituency and therefore
trades off costs and benefits across all districts.
Authority is subject to constraints.
Give agent incentive to take actions;
Then provide checks.
Logic of Delegation
Two alternative modes of policy making
– Regulatory Agencies:
Delegation to Executive
Legislators decide where policy is made.
Since legislators’ primary goal is reelection,
policy will be made so as to maximize
legislators’ reelection chances.
Transaction Cost Politics
Each alternative mode of policy making has
its own set of costs:
– Legislative Policy Making (Committees)
Logrolling, delay, informational problems
– Agency Policy Making (Delegation)
Principal-agent problems of oversight
So when deciding where policy will be
made, legislators trade off these internal and
external costs of policy production.
Congress’sdecision to delegate is like a
firm’s make-or-buy decision.
– Legislators can either produce policy internally,
– Legislators can subcontract out (delegate) to the
Low Discretion High Discretion
Trade Off: Legislative vs. Agency
Amount of discretion delegated to the
executive balances these costs at the
Legislators seek reelection
Two choices of how to make policy
– Committees or Agencies (make-or-buy)
Each has costs
Legislatorstrade off these two options when
deciding what to delegate and what to do
Floor sets policy pF
Don’t F x=p F+w
N F d
F P A x =SQ+pA+w
Floor sets President sets Agency learns
discretion (d) Agency ideal exact value of w ,
status quo (SQ) point A sets policy pA s.t.
Institutional Choice Policy Making Process Outcomes
Policy (p) = Policy chosen Floor Voter (f) = Median Legislator
Status quo (SQ) = Policy in effect President (p) = President’s ideal point
State of the World (w )= Uncertainty Agency (A) = Agency’s ideal point
Outcomes (x) = Combination of p and w . Discretion (d) = Amount of authority delegated
-R w 1 [w1+d]
xF=SQ w 2 xA= xP w3 R
Range Outcomes Example
w < xA - SQ -d w+ SQ + d w1
xA -SQ- d < w < x A- SQ + d xA w2
xA -SQ+d < w w+ SQ - d w3
Depending on the value of w, the president will set
policy as close to his ideal point as possible,
subject to limits on discretion set by Congress.
Discretion & Inter-branch Conflict
Discretion declines as President’s
ideal point moves away from the
median legislative voter.
Median R President’s
Legislative Voter Ideal Point
Summary of Results
Congress benefits from delegating (through
informational gains or the elimination of
inefficient logrolls) at some distributive cost.
the president moves farther away,
Congress delegates less authority.
– If the president is too far away (P>R), Congress
does not delegate at all.
When Congress does delegate, it constrains
the president by setting d<R.
Tariffs over Time
Since the 1934 RTAA, steep decline in tariffs.
Delegation Is Not Monolithic
The terms of delegation have changed
Sometime Congress delegates broad authority
Sometimes Congress limits this authority
(e.g., fast track procedures).
Delegation is constrained by administrative
Congressional override procedures
Changes in criteria
Gives rise to procedural protectionism
The politics is in the procedures
Does Divided Government
Impact Trade Outcomes?
Theory: Divided government leads to less
has less protectionist preferences than median
member of Congress
So less delegation leads to higher tariffs
Major Postwar Trade Acts
Major U.S. Trade Legislation, Divided Government, & Delegation, 1948-1992
Legislation Administration Divided Delegation
1948 Extension of the RTAA Truman Divided Decreased
1949 Extension of the RTAA Truman Unified Increased
1951 Extension of the RTAA Truman Unified Decreased*
1953 Extension of the RTAA Eisenhower Unified Increased
1954 Extension of the RTAA Eisenhower Unified Increased
1955 Extension of the RTAA Eisenhower Divided Decreased
1958 Extension of the RTAA Eisenhower Divided Decreased
1962 Trade Expansion Act Kennedy Unified Increased
1974 Trade Reform Act Nixon Divided Increased*
1979 Trade Agreements Act Carter Unified Increased
1984 Trade and Tariff Act Reagan Divided Decreased
1988 Omnibus Trade and Reagan Divided Decreased
Effect of Delegation on Tariffs
Least Squares Estimates of the Effect of Delegation on the Tariff
Dependent Variable: log(TARIFF)
Independent Variable Model 1 Model 2
CONSTANT 0.023 0.032
log(GNP)t-1 -0.67 -0.85
log(UNEMPLOY)t-1 0.045 0.040
log(PPI)t-1 -0.44 -0.49
Number of Observations 42 42
R2 0.17 0.19
Note: t-statistics in parentheses. * a < .10. ** a < .05.
Tariff & Non-Tariff Barriers, 1950-86
Astariffs have declined, Non-tariff barriers
have steadily increased.
Structure of US Trade Policy
Escape Clause National Security Retaliation Countervailing Anti-Dumping
U.S. Law: Section 201 Section 232 Section 301 Section 303 Section 731
1974 Trade Act 1962 Trade Act 1974 Trade Act 1930 Tariff Act 1930 Tariff Act
Modified: 1979, 1984, 1988 1974, 1979, 1988 1984, 1988 1974, 1984, 1988 1974, 1979, 1984
Rule: Increased imports Imports threaten to Barriers restrict US Export subsidy Price below “fair
cause or threaten to impair national commerce causes or threatens to market value” causes
cause substantial security by weakening cause material injury or threatens to cause
injury vital domestic industry No injury test material injury
Penalty: Duty, quota, OMA or At president’s Determined by USTR Tariff which offsets Tariff which raised
trade adjustment discretion subject to direction of subsidy or negotiated price to fair market
assistance or other president settlement value or negotiated
Investigating Agency: ITC Commerce (ITA) USTR Commerce (ITA) and Commerce (ITA) and
Recommendation 6 months 270 days2 12-18 months1 160-300 days 2235-420 days
Decision-maker President, based on President USTR, subject to the Commerce (ITA)
ITC’s proposed direction of the
Decision Due 60 days No deadline Included in Upon recommendation
Congressional Yes, within 90 days if No No No No
Override president rejects
In 1983, penalty was determined by the President, who was the decision maker in 301 cases. The USTR’s recommendation was due between 9 and 14 months,
depending on the nature of the unfair practice. The decision was due 21 days after the USTR’s report to the president. The OTCA of 1988 gave this authority to
In 1983, recommendation was due in 12 months.
Petitions Under US Trade Law
Pending as of
Type of Action Years included # of Petitions Successful Success Rate Unsuccessful1 Most Recent
201 (1958-1977) 87 15 .17 72
Escape Clause (1975-1984) 53 18 .34 35 1
232 (1962-1983) 14 3 .21 11 0
301 (1975-1987) 43 32 .74 6 4
303 (1979-1984) 235 27 .11 189 19
731 (1979-1984) 175 48 .27 127 60
Cases that were withdrawn by petitioners after the investigation began were coded as unsuccessful; these included 51 anti-dumping cases, 64 CVD cases, 3
301 cases, and 1 national security case.
Twenty-six agricultural cases were excluded.
In 20 cases, a negotiated settlement was reached; in 12 cases the president retaliated.
The administrative procedures associated with
each trade remedy influence who uses them
and the probability of success.
In theearly 1980s, west coast cement
producers were facing serious competition
from foreign imports, especially from
Mexican cement producers received oil from
the government at discounted prices.
– May be seen as an unfair advantage
What are US firms’ options?
US Firms’ Strategies
US trade laws (AD,
Joint law suit
Options Congress President Agency Courts Likelihood of Success
US trade laws (AD,
Joint law suit
Calculating Antidumping Duties
Sale price $85 $80
Transportation 12 20
Terminal & trans. 10 17
U.S. Customs -- 2
Mill Net Price
US plant Dumping margin =
(52 - 31) / 31 = .68
terminal $12 .
58 percent duties levied on Mexican exports
All other firms withdraw from US market
Appeals in US courts (Loses)
Lobbies legislators (Little impact)
Files GATT petition (Wins, US overrides)
Lobbies Mexican Government (NAFTA)
NAFTA Dispute Settlement (Loses)
Cemex plays US domestic politics game
Setsup headquarters in Houston
Purchases a plant in Texas, employing 1,800
Builds coalitions with consumers and representatives
Joint law suit
Session 13: Cemex and
– To see an recent and ongoing instance of
integrated strategy in an international setting
» multiple institutions
US trade law
» shifting interests
» complex information
– To step back and consider the nature of trade
institutions more generally
Cemex’s Initial Market Strategy
Objective: to become a major international player
Actions (conventional, market strategy)
– Dominate in Mexico via acquisitions and investment in
– Export to the US when economy is sluggish
– Obtain capacity in other countries
Spain, Venezuela, Panama
Nonmarket consequence of the export component
of the market strategy
– Antidumping petition
Was this anticipated? Could it have been?
Cemex’s Revised Strategy
Comply with all requests to maximize chance of
Outcome: lost. (58% duty)
Seek to overturn the decision
– In the courts
Outcome: min. influence: no rent chain; little US presence
Try to reduce the duty in annual reviews (integrated)
– Withdraw from low-price markets in US
– Maintain the bulk (low-price) market in Mexico
Outcome: (e.g.) duty lowered to 42.74% in 1994
Shift to a potentially neutral institutional arena
– File a GATT petition
Outcome: won the battle, lost the war
Revised Strategy (cont.)
Market components consistent with nonmarket
– Export to US from Spain, which is not subject to the
– Export from Mexico to Japan to utilize Mexican
– Acquire capacity in the US
Establishes a US presence
Signals long-term commitment
Raises prospects for coalition building
The strategy is integrated
– Market components lessen the nonmarket pain in the
– Nonmarket components reflect long-term international
Cemex Update: Problems
Cemex’s inability to overturn antidumping
decision created three problems in the
– Persistent complaining by US rivals at periodic
Rivalsclaim that dumping margin should be based
on bagged (no bulk) cement, thus 111%
Does Cemex have to fight just to keep the situation
from getting worse?
– Stranded Cemex terminal and distribution
facilities in US
– Excess capacity in Mexico
Responses & Consequences: Nonmarket
Argued that the US should abide by GATT
decision – no effect
Refused to supply data on bagged cement
– ITA uses default procedure which takes the maximum
(not weighed average) of margins; margin goes to
61.85% in May, 1995
Cemex files for NAFTA dispute resolution in July
– Untried, but relatively good expectations
Cemex is winless within the US system
US Department of Commerce indicates that it thinks Cemex
stands a good chance of winning
– Unfounded, though: summer of 1996
At approximately the same time...
Responses & Consequences: Market
Demand in US is high, supply is low, prices rise
Cemex acquires facilities in Venezuela and Panama
Cemex exports from Mexican plants to Japan,
Indonesia, the Philippines, Malasia, and Taiwan
Cemex exports to the US market from its facilities in
Spain and Venezuela
– escapes antidumping margin at least in the short run
Cemex purchases a plant in Texas, moves regional
headquarters from Monterey to Houston, employs
1800 in US
– building a rent chain; US domestic presence
New Objectives, Strategy, Implementation
– Change the way US calculates dumping margins
– Negotiate a more favorable settlement
Strategy and implementation
– Coalition building with consumers
National Association of Homebuilders (180,000 firms; better
coverage than Cemex)
– Enlist allies to be policy entrepreneurs
Tom Delay (Republican whip from district with a Cemex
plant); Gene Green (Democrat, also with 3 Cemex plants)
Hired Randy Delay & a DC public relations firm
– Public advocacy
op ed in Wall Street Journal
ads in Roll Call
Counteractive lobbying by US
– Joe Barton (Republican of Texas) writes to all
House members asking them to oppose
– Gene Green gets 34 co-signers on a letter to
Department of Commerce and USTR asking for
a negotiated settlement
Ongoing international struggle has evolved
into an ongoing domestic political struggle,
A Concluding Perspective
Nation A (comparative advantage) Nation B
Nation C Nation D
International political economy
Generalizations from Cemex
Market rivals in international settings have rights to
initiate antidumping proceedings against imports.
It is difficult to win in administrative processes that are
governed by mandates, due process, openness, and court
reviews. Legislation or strong legislative oversight are
usually required for nonmarket success.
On the surface, trade policy lies in the domain of
governments and international relations.
Beneath the surface, trade policy is set and implemented
in an environment of intense domestic politics.
In 1990 Cemex had little influence on trade policy. As
of 1996, its integrated strategy has increased its