Description
answer the two questions from the reading: the total of the slides are 8
Please no intro or Conclusion, the students already did the reading?
You will create 4 slides for the first question
for this question: Why countries have the political cleavages they do, and why they change?
so first you will mention why this question is important then you will answer the question then you will mention What is the strength and the weakness of the outcome and then What the authors are missing in the reading
And then 4 slide slides the second question
for this question: Why Do Some Coalitions Win Over Others?
for this one you will first mention why this question is important then you will answer the question then you will mention What is the strength and the weakness of the outcome and then What the authors are missing in the reading
================
here is the reading:
– Commerce and Coalitions: How Trade Affects Domestic Political Alignments Ronald Rogowski: the introduction chapter and the chapters 6,7
– Hiscox , Michael international trade and politics conflict: chapter 1-3
-The Determinants of Tariff and Nontariff Trade Restrictions in the United States
-Lohmann , divided government and US trade policy
Unformatted Attachment Preview
Divided Government and U.S. Trade Policy: Theory and Evidence
Author(s): Susanne Lohmann and Sharyn O’Halloran
Source: International Organization , Autumn, 1994, Vol. 48, No. 4 (Autumn, 1994), pp.
595-632
Published by: The MIT Press
Stable URL: https://www.jstor.org/stable/2706897
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
, The MIT Press , Cambridge University Press and University of Wisconsin Press are
collaborating with JSTOR to digitize, preserve and extend access to International Organization
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
Divided government and U.S. trade
policy: theory and evidence
Susanne Lohmann and Sharyn O’Halloran
Researchers have proposed a variety of explanations to account for U.S. trade
policymaking. The international political economy literature emphasizes the
interaction between international power relations and commercial exchange,
the effect of global regimes in fostering cooperation among trading states, and
the impact of international market conditions on the domestic economy.’ The
scholars associated with this literature assert that trade policy is viewed most
appropriately within the constraints imposed by international relations and
market structures. Absent from their work, however, is a clear recognition of
the effect of domestic conflicts and political institutions on trade policy. Even
those scholars who emphasize the importance of domestic politics miss
important recent insights in the American politics literature according to which
institutions are themselves an object of political choice.2
This work was supported by a James and Doris McNamara Faculty Fellowship at the Graduate
School of Business, Stanford University (Lohmann) and a Social Science Research Council
Fellowship in Foreign Policy Studies (O’Halloran). We thank David Baron; Jonathan Bendor;
David Brady; Linda Cohen; David Epstein; Keith Krehbiel; John Londregan; Ronald Rogowski;
participants at the annual meeting of the American Political Science Association, Washington,
D.C., September 1991; the National Bureau of Economic Research Conference, Cambridge,
Mass., November 1991; the Midwestern Political Science Association Meetings, Chicago, April
1992; and the seminar participants at Columbia, Cornell, Princeton, Rice, Stanford, Washington
(St. Louis), and Yale Universities for their comments on an earlier draft.
1. See Robert Gilpin, The Political Economy of Intemational Relations (Princeton, N.J.:
Princeton University Press, 1987); Stephen Krasner, ed., Intemational Regimes (Ithaca, N.Y.:
Cornell University Press, 1983); John A. C. Conybeare, Trade Wars: The Theory and Practice of
Intemational Commercial Rivalry (New York: Columbia University Press, 1987); and David Lake,
Power, Protection, and Free Trade: Intemational Sources of U.S. Commercial Strategy, 1887-1939
(Ithaca, N.Y.: Cornell University Press, 1988). For an overview of this literature, see Edward D.
Mansfield, “The Concentration of Capabilities and International Trade,” Intemational Organization 46 (Summer 1992), pp. 731-63.
2. See Helen Milner, Resisting Protectionism (Princeton, N.J.: Princeton University Press, 1988);
Joanne Gowa, Closing the Gold Window: Domestic Politics and the End of Bretton Woods (Ithaca,
N.Y.: Cornell University Press, 1983); Beth Yarbrough and Robert Yarbrough, Cooperation and
Govemance in Intemational Trade (Princeton, N.J.: Princeton University Press, 1988); Peter
Gourevitch, Politics in Hard Times: Comparative Responses to Intemational Economic Crises (Ithaca,
Intemational Organization 48, 4, Autumn 1994, pp. 595-632
? 1994 by The 10 Foundation and the Massachusetts Institute of Technology
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
596 International Organization
This article develops and tests a distributive politics model of U.S. trade
policy that captures both the roles of domestic conflict and institutional design.
In this model, legislators find themselves trapped into inefficient logrolling
when passing trade legislation. We derive conditions under which Congress
grants trade policymaking authority to the President in order to achieve more
efficient outcomes. We also show that in the presence of partisan conflict and
divided government, the members of the majority party in Congress may have
incentives to constrain the President’s use of delegated authority, thereby
forcing the President to accommodate partially their protectionist pressures.
As a consequence, divided government may be associated with higher levels of
protection.
Two theories examine the role of institutions in forming U.S. trade policy.
Presidential dominance and congressional dominance theories both link the
policy preferences of the President and Congress to political outcomes.3 In
their extreme form, these theories are diametrically opposed in their view of
congressional-executive relations and consequently in the outcomes they
predict.
Proponents of the presidential dominance hypothesis argue that Congress
has delegated much of its authority to set trade policy to the President.4 Even if
legislators retain some constraints over executive action, they are unwilling or
unable to use their power. This hypothesis has been formulated in various ways.
For example, I. M. Destler argues that legislators realize that they are unable to
resist the protectionist pressures of special interest groups. Thus, they choose
to bind their hands by delegating policymaking authority to the President (or an
executive broker), who is less susceptible to protectionist pressures. This
delegation of authority insulates Congress from interest group demands and
allows legislators to shift the blame for the negative side-effects of trade
liberalization.5 The policy predictions of this model are consistent with the
observation that U.S. tariff rates peaked when Congress set tariffs item by item
N.Y.: Cornell University Press, 1986); and Susanne Lohmann, “Electoral Cycles and International
Policy Cooperation,” European Economic Review 37 (1993), pp. 1373-91.
3. For an overview of this literature, see Terry Moe, “The New Economics of Organization,”
American Joumal of Political Science 28 (November 1984), pp. 739-77. For an application to trade
policy, see Sharyn O’Halloran, Politics, Process, andAmerican Trade Policy (Ann Arbor: University
of Michigan Press, 1994).
4. See Arthur M. Schlesinger, The Imperial Presidency (Boston: Houghton Mifflin, 1973); Robert
Pastor, Congress and the Politics of U.S. Foreign Economic Policy, 1926-1976 (Berkeley: University
of California Press, 1980); Lawrence Margolis, Executive Agreements and Presidential Power in
Foreign Policy (New York: Praeger Publishers, 1986); I. M. Destler, American Trade Politics
(Washington, D.C.: Institute for International Economics, 1992); Stephan Haggard, “The
Institutional Foundations of Hegemony,” International Organization 42 (Winter 1988), pp. 91-119;
Judith Goldstein, “Ideas, Institutions and American Trade Policy,” International Organization 42
(Winter 1988), pp. 179-217; and I. M. Destler, “U.S. Trade Policy-making in the Eighties,” in
Alberto Alesina and Geoffrey Carliner, eds., Politics and Economics in the Eighties (Chicago:
University of Chicago Press, 1991).
5. For a similar argument, see Morris P. Fiorina, “Legislative Choice of Regulatory Forms:
Legal Process or Administrative Process,” Public Choice, vol. 39, no. 1, 1982, pp. 33-66.
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
U.S. trade policy 597
in the 1930 Smoot-Hawley Tariff Act, whereas tariff rates declined precipitously once Congress delegated trade policymaking authority to the President
in the 1934 Reciprocal Trade Agreements Act.
Proponents of the congressional dominance hypothesis, on the other hand,
argue that the delegation of authority to the President does not imply a
relinquishing of power.6 Administrative procedures, “fire-alarm” or “policepatrol” oversight, and the credible threat of sanctions effectively constrain the
President’s leeway to set policy. As a consequence, executive decision making
will mirror congressional interests.7 Applied to trade policy, the congressional
dominance hypothesis in its extreme form suggests that even if Congress
delegates power to the executive, the President’s discretionary powers are
limited due to procedural constraints. The resulting trade policies, then, are
identical to those that would be implemented by Congress if its members
passed trade legislation without recourse to delegation. At first blush, the
congressional dominance hypothesis is consistent with the observation that
Congress repeatedly has constrained the President’s discretion by requiring
presidential trade proposals to fulfill numerous consultation requirements and
pass a variety of veto points, as in the 1974 Trade Reform Act and its 1984
amendments as well as the 1988 Omnibus Trade Act.
In our view, both the presidential and congressional dominance hypotheses
provide an incomplete picture of trade policymaking. According to the
presidential dominance hypothesis, the delegation of trade policymaking
authority is a matter of political choice; but this approach fails to acknowledge
that under certain circumstances, Congress may choose not to cede to the
President complete discretion. On the other hand, the congressional dominance hypothesis allows Congress to set the terms of delegation; but this
approach does not capture the possibility that legislators may deliberately
design a delegation regime to implement policy outcomes different from those
that would result if Congress enacted legislation on its own. We synthesize
these different strands of literature in a more general model of delegation and
accommodation in which both the presidential and congressional dominance
6. See Barry R. Weingast and Mark Moran, “Bureaucratic Discretion or Congressional Control:
Regulatory Policymaking by the Federal Trade Commission,” Journal of Political Economy 91
(October 1983), pp. 765-800; Mathew McCubbins and Thomas Schwartz, “Congressional
Oversight Overlooked: Police Patrols and Fire Alarms,” American Joumal of Political Science 28
(February 1984), pp. 165-79; and Mathew McCubbins, Roger Noll, and Barry Weingast,
“Administrative Procedures as Instruments of Political Control,” Joumal of Law, Economics, and
Organization 3 (Fall 1987), pp. 243-78.
7. The literature on the “new institutionalism” is often, and perhaps narrowly so, identified with
the congressional dominance hypothesis. Recently, a number of scholars have modified the
congressional dominance hypothesis to take into account the President’s veto powers. See, for
example, Mathew McCubbins, Roger Noll, and Barry Weingast, “Structure and Process; Politics
and Policy: Administrative Arrangements and the Political Control of Agencies,” Virginia Law
Review 75 (March 1989), pp. 431-82; and D. Roderick Kiewiet and Mathew D. McCubbins, The
Logic of Delegation: Congressional Parties and the Appropriations Process (Chicago: University of
Chicago Press, 1991).
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
598 International Organization
hypotheses are obtained as special cases. Such a synthesis in particular explains
why Congress partially constrained presidential policymaking by granting the
executive fast-track authority in the 1974 Trade Reform Act: the President may
engage in international trade negotiations on behalf of Congress but presidential trade proposals must pass an up-or-down congressional vote. Subsequent
acts and amendments have strengthened or weakened the implied constraint
while retaining its partial nature.
The next section first presents an informal exposition of our theory and then
formally develops the model. We subsequently provide empirical support for
our theory through a brief historical overview of institutional changes in
postwar U.S. trade policy and an econometric examination of changes in
average tariff rates. The last section summarizes our conclusions. The appendix
contains a formal derivation of the results.
A distributive politics model of delegation
and accommodation
Collective dilemmas and delegation
Our model explains why a majority in Congress would delegate power to the
President. We build on an extensive literature that examines how players
engaged in a noncooperative game with inefficient outcomes might be able to
achieve more desirable results by choosing delegates to play on their behalf.8 In
these models, a principal delegates discretionary authority to an agent who is
deliberately chosen to have different preferences than the principal. In contrast
to standard principal-agent models, these models do not consider moral
hazard or adverse selection. Nor are they driven by the desire of a political
principal to shift the blame for unpopular decisions.9
In our model, individual members of Congress and the President have
different constituencies and consequently differ in their trade policy goals.
Each legislator cares only about the benefits and costs of protectionism to his or
her own district. Thus, in passing trade legislation, Congress is trapped into
distributive logrolling that results in inefficiently high levels of protection. The
total marginal costs imposed on all sectors by each district-specific measure
exceed the marginal benefits to the protected district. This inefficiency arises
because each individual member weighs the marginal benefits and costs for his
or her own district when proposing a trade policy measure, but ignores the
8. This idea has been developed in the context of central banking and taxation: see Kenneth
Rogoff, “The Optimal Degree of Commitment to an Intermediate Monetary Target,” Quarterly
Joumal of Economics 100 (1985), pp. 1169-89; Susanne Lohmann, “Optimal Commitment in
Monetary Policy: Credibility Versus Flexibility,” American Economic Review 82 (March 1992), pp.
273-86; and Nahum D. Melamud and Dilip Mookherjee, “Delegation as Commitments: The Case
of Income Tax Audits,” RAND Joumal of Economics 20 (1989), pp. 139-63.
9. For an example of the latter, see Fiorina, “Legislative Choice of Regulatory Forms.”
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
U.S. trade policy 599
negative externalities that might be imposed on other districts.10 (We define a
vector of trade policy measures as efficient if it minimizes the sum of the
districts’ losses. The efficient frontier is given by the set of such vectors that are
obtained by varying the weights assigned to the districts. This concept of
efficiency is more general than the widespread notion that an outcome is
efficient only if the total benefits exceed the total costs.)
Inefficient logrolling procedures, together with a norm of universalism that
assigns benefits to each district, are supported by the following logic.11 Each
legislator proposes to protect his or her own district, and the legislature votes
on each of the proposals by majority rule. If any one legislator were to vote
against another legislator’s proposal, the former legislator would be punished:
some other legislator would propose an amendment that assigns no protection
to the defector’s district but maintains the proposed levels of protection to all
other districts. Given that all other legislators vote in favor of each others’
proposals, no legislator’s vote is decisive. Each district-specific measure passes,
and it is individually rational for each legislator to vote for all proposals. Thus,
the universalistic logrolling pattern is supported as an equilibrium by a
particular set of beliefs.12
The President, on the other hand, has a national constituency and cares
about the losses incurred by all districts. If given discretionary powers to set
trade policy, the President would implement measures that trade off the
marginal benefits derived from protecting industries in one district against the
marginal costs imposed on all other districts. As a consequence, efficient
outcomes would be achieved. Each member of Congress would be better off
under this outcome than under the inefficient logrolling outcome. In this
setting, the legislators would never choose to constrain the President’s powers,
since such constraints could only lead to less efficient policy.
In our view, institutional constraints might arise in the presence of partisan
conflict between Congress and the President. We develop a constituency-based
notion of partisan conflict. Consider the situation where the President weighs
the marginal costs and benefits of protecting various constituencies according
to partisan reelection considerations: a Democratic President places a relatively higher weight on Democratic constituencies, while a Republican President cares relatively more about Republican constituencies. If different parties
are in control of the executive and legislative branches, the President’s
10. See also Barry Weingast, Kenneth Shepsle, and Christopher Johnsen, “The Political
Economy of Benefits and Costs: A Neoclassical Approach to Distributive Politics,” Joumal of
Political Economy 89 (August 1981), pp. 642-64.
11. This concept of universalism does not exclude the possibility that the district-specific benefits
enjoyed by any one district are outweighed by the total costs incurred by that district.
12. The Folk Theorem applies to a situation in which the members of Congress can vote
indefinitely on a sequence of proposals. This theorem implies that the set of proposal and voting
strategies described above is sustainable as a Nash equilibrium for some set of beliefs about
punishment strategies. Clearly, the equilibrium characterized above is only one out of multiple
belief-driven equilibria.
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
600 International Organization
response to district-specific economic conditions will differ ex ante from the
response desired by the congressional majority party.13 In this situation,
members of the majority party may choose to delegate authority to the
President but constrain executive powers by making presidential proposals
subject to an up-or-down vote. This constraint forces the President to
accommodate partially congressional demands for more protectionist trade
policy measures. The President will choose to accommodate those members
who can be “bought off” at least expense and consequently will assemble a
bipartisan majority to vote in favor of the proposal. In general, this majority will
tend to include a higher proportion of members from the President’s party.
Nevertheless, the President will exclude members of his or her party from
districts with very unfavorable economic conditions and will include members
of the other party from districts with favorable economic conditions.
Thus, our model suggests that voting on presidential trade bills may appear
to be nonpartisan even in the presence of severe partisan conflict between
Congress and the President. In contrast, the legislators’ party affiliations are
crucial to votes on the institutional arrangements that underlie trade policy,
since those votes are cast before the district-specific economic conditions are
revealed.
Our model implies that protection levels are higher when economic
conditions are unfavorable. Moreover, protection levels will be lower when the
President alone sets trade policy than when Congress passes trade legislation,
and a constrained President will tend to implement more protectionist trade
policy measures than an unconstrained one. Finally, the President’s discretionary powers will be more constrained under divided than under unified
government. As a consequence, divided government may be associated with
higher levels of protection than unified government.
Policy preferences of the players
In this section, abstracting from the bicameral structure of Congress, we
formally develop a model of the strategic interaction between Congress and the
President that supports the implications discussed above. We first specify the
preferences of the players. The country consists of n legislative districts, where
n is odd. The political representation of the people in this country is
determined geographically. In the legislature, each district is represented by
one legislator, who is motivated to maximize that district’s utility function:
Ui( pl . Pn) = PsiPi -Pi w E P3J (1
13. While we do not explicitly model the role of party in Congress, our main results are robust
with regard to the possibility that members of the majority party partially or fully exclude members
of the minority party from distributive logrolling.
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
U.S. trade policy 601
where i E N = {1, . . ., n}; pi is a district-specific trade policy measure that is
restricted to be greater than or equal to zero; P3i (the weight on the
district-specific benefits of protection) is the absolute value of a draw from a
normal distributionf (pi) with mean zero and strictly positive but finite variance
u2; and y is a strictly positive weight on the costs district i incurs due to the trade
policy measures that afford protection to other districts.
The formulation of the utility function captures the notion that the benefits
of protectionism are concentrated, while the costs are dispersed. The benefits
that the producers in district i derive from being protected by the district-
specific trade policy measure pi are given in the first term in equation 1, ,3ipi.
The second term, pi2, reflects the costs incurred by the consumers in district i
due to district-specific protection. The linear formulation of the benefit term
and the quadratic formulation of the cost term will allow us to derive a unique
closed-form solution to the maximization problem.
The weight P3i determines the district’s optimal trade-off between the
benefits and costs generated by district-specific protection. We will interpret
the realization of this weight as being a function of district-specific economic
conditions. If the economic conditions in a district are favorable, then the
district has a low realization of weight Pi on the benefits derived from
district-specific protection; unfavorable economic conditions are associated
with a high realization of Pi. The assumption that P3i is the absolute value of a
draw from a normal distribution implies that the level of protection desired by
the representative of a district is strictly positive, as is the efficient level of
protection for that district. In addition, this assumption reflects the notion that
extremely unfavorable economic conditions (high Pi) occur more rarely than do
fairly favorable (low fi) or fairly unfavorable (intermediate Pi) conditions.
The third term in equation 1, l]ENIi Pj, captures the negative cross-district
external effects that producers and consumers in district i experience as a result
of the trade policy measures that afford protection to other districts. The
severity of these external effects increases as y increases, reflecting an increase
in the degree to which local economies are interdependent. For simplicity, we
assume that the districts’ utility functions have identical weights y and differ
only in the weight Pi.
The President has a national constituency and maximizes the sum of the n
districts’ utility functions:
n
i=1
where
We
J
)
2
(2)
jENi
wi
will
President
conflict,
is
the
focus
on
weigh
two
weighs
the
c
all
President
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
602 International Organization
TABLE 1. Nonpartisan and partisan weights (we) for majority and minority p
districts under unified and divided govemment, where n = the number of
districts, m = the number of majority party members, and d = the severity
ofpartisan conflict
Partisan case
Nonpartisan case Unified govemment Divided govemment
Majority party districts 1 1 + d(n – m) 1 – d(n – m)
Minority party districts 1 1 – dm 1 + dm
or her party than on the other party’s districts. Let m be the number of majority
party members in Congress. The relative weights placed on majority and
minority party districts under unified and divided government are given in
Table 1. In our model, the severity of the partisan conflict is represented by the
parameter d (see Table 1). For d = 0, the President weighs the districts’ utilities
in the aggregate utility function equally (equation 2). As d increases, the
partisan conflict becomes more severe, as is reflected in the increasing
difference between the weights assigned to majority and minority party
districts.
Time sequence
The formal model consists of two stages. At the institutional design stage,
members of Congress choose by majority rule whether to retain or delegate
their trade policymaking powers. If the members of Congress delegate power,
they decide by majority rule whether to constrain partially the President by
subjecting presidential trade proposals to an up-or-down (closed-rule) simple
majority vote.
After the institution is set up, a one-shot game is played. For the duration of
that game, the legislators are fully committed to respect the institutional
constraints agreed upon at the institutional design stage.14 A simplified game
tree is shown in Figure 1.
Once the institutional arrangement is set up, a vector of relative weights on
the benefits and costs of district-specific protection, [pi, … ., n], is realized and
14. An assumption that politicians can “solve” collective dilemmas by credibly committing to
institutions is standard in institutional design models. This assumption is critically reviewed in
Jonathan Bendor and Susanne Lohmann, “Institutions and Credible Commitment,” mimeograph,
Graduate School of Business, Stanford University, February 1993.
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
U.S. trade policy 603
CONGRESS
CZ
.-
Does
not
Delegates
to
4- = delegate to and constrains
‘ * President /Delegates to but President
does not constrain
73~ ~ President
NATURE NATURE NATURE
chooses chooses chooses
district-specific district-specific district-specific
shocks
shocks
shocks
4 CONGRESS 4 PRESIDENT 4 CONGRESS
4-
makes
makes
makes
proposals proposals proposals
CONGRESS
l
ONGRESS
Votes yes Votes no Votes yes Votes no
PRESIDENTIAL CONGRESSIONAL DELEGATION AND
DOMINANCE DOMINANCE ACCOMMODATION
GAME
GAME
GAME
FIGURE 1. Time sequence of events
commonly observed by all players. This is the only source of uncertainty in the
model.
The specific set of institutional constraints determine the sequence in which
players can make or accept proposals. If the legislators do not delegate trade
policymaking authority, they play the congressional dominance game. Each
legislator proposes a district-specific trade policy measure, and each proposal is
voted on by simple majority vote under an open rule. If the President is given
unlimited powers, the presidential dominance game is played. The President
unilaterally sets the district-specific trade policy measures. Finally, if the
President’s powers are constrained, then Congress and the President play the
delegation and accommodation game. In this case, the President’s trade policy
proposal is subject to an up-or-down vote in Congress. If the President’s
proposal receives a majority vote, it is implemented. If the proposal fails,
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
604 International Organization
Congress reverts to the congressional dominance game. Thus, the outcome of
the congressional dominance game plays the role of a threat point.15
Trade policy outcomes will depend on the institutional constraints, the
players’ trade policy preferences, and the realization of the district-specific
economic conditions. Rationally anticipating the play of the game once
institutional constraints are specified, legislators form an expectation of the
resulting trade policy outcomes. As a consequence, they have induced
preferences over institutional arrangements. At the institutional design stage, a
majority of legislators choose whether to delegate authority to the President
and whether to constrain the President’s authority by making presidential
proposals subject to an up-or-down vote.
The legislature operates behind a “veil of ignorance” in choosing the
institutional design that will affect trade policy outcomes. Ex ante, members of
Congress do not know whether economic conditions in their districts will be
favorable or unfavorable. Thus, all members of the majority party have
identical preferences over institutional arrangements, as do all members of the
minority party. The assumption that the members of the majority and minority
parties have homogeneous preferences is not crucial. For instance, legislators
who represent districts dominated by the automobile industry may know in
advance that they are likely to favor a higher level of protection for automobiles
in the future, while representatives of districts dominated by the computer
industry expect to favor a low level of protection for computers. (Formally, this
possibility could be represented by having different variances for the districtspecific shocks. While our qualitative results would continue to hold, their
derivation would become more complicated.)
The structure of the game, including the order of play and the preferences of
all players, is common knowledge. (Of course, this assumption is not identical
to or implied by the assumptions of perfect or complete information. Indeed,
the game described above is a game of imperfect information since the
district-specific shocks are unknown at the outset.) The equilibrium concept is
perfect Bayesian equilibrium. Moreover, the players are constrained not to use
weakly dominated strategies.16 The model is solved by backward induction. We
first derive the equilibrium proposal and acceptance strategies for the game
stage. Then we derive the institutional choice made by a majority in Congress at
15. We find this assumption more plausible than the alternative assumption that the status quo
is maintained if the President’s proposal is voted down. Moreover, this assumption allows us to
view the congressional dominance hypothesis as a special case of our model. If Congress
constrained the President’s powers by making presidential proposals subject to an open rule vote,
the President’s discretionary powers would fully unravel due to the subsequent amendment
activity. The resulting outcome would correspond to the one that is implemented in the
congressional dominance game.
16. This constraint eliminates undesirable voting equilibria of the following type. Suppose all
legislators believe that a proposal will pass by more than one vote so that no legislator can affect the
outcome by changing his vote. As a result, each legislator is indifferent between voting for or
against any proposal, and one proposal may in fact pass although a majority would be better off if it
failed, and vice versa.
This content downloaded from
104.181.75.154 on Sat, 09 Mar 2024 09:39:45 +00:00
All use subject to https://about.jstor.org/terms
U.S. trade policy 605
TABLE 2. Game stage implicationsfor levels ofprotection
Delegation and
accommodation game
Congressional Presidential
dominance dominance Voting constraint Voting constraint
game game not binding binding
Level of High and Low and Low and High and
protection inefficient efficient efficient inefficient
the institutional design stage. In each case, we examine the implications of
partisan conflict and unified or divided government. The equations to which we
refer in the proofs of the propositions are provided in the appendix.
Game stage
An overview of the im