Bargaining and Bicameralism

• DOI: http://doi.org/10.1111/lsq.12179
• Date: 01 February 2018
• Published date: 01 February 2018
• Author: Giri Parameswaran

GIRI PARAMESWARAN

Haverford College

Bargaining and Bicameralism

In bicameral legislatures, the protection of small states often motivates the malap-

portionment of the upper house. Using a legislative bargaining model, I show that

malapportionment may produce the opposite effect. Under unicameralism, same-state

legislators are shown to not inherently be coordinated to cooperate, diminishing the fear

of a big-state conspiracy. By contrast, under bicameralism, preference complementarities

enable upper-house legislators to effectively coordinate their state delegations, and

this skews the expected allocation in favor of big states. Hence, unless bicameralism

significantly increases their agenda power, small states will fare even worse under

bicameralism whenever they are disadvantaged under unicameralism.

The equality of representation in the Senate is ...evidently the result of a compromise

between the opposite pretensions of the large and small States. ... A government

founded on principles more consonant tothe wishes of the larger States, is not likely to

be obtained from the smallerStates.

— James Madison, Federalist No. 62

Bicameralism is an institutional feature common to many legisla-

tures around the world. In almost all bicameral systems, legislators in the

lower house represent districts of roughly equal size by population (Tsebe-

lis and Money 1997), ref‌lecting the principle of equal representation. By

contrast, representation in the upper house is often intentionally malappor-

tioned to benef‌it particular constituencies.

1

For example, a malapportioned

upper house that gives equal representation to states or provinces has the

obvious effect of overrepresenting regions with smaller populations.

In the United States, the overrepresentation of small states in the

upper house resulted from a compromise with the small states, who

feared that their interests would be ignored in the popular chamber where

a few large states could command majorities between th emselves. A sim-

ilar compromise, to entice smaller states to join the federation, informed

the design of the Australian Senate. The logic seems straightforward:

overrepresentation of small states in the upper house protects them

against usurpation of the policy agenda by the larger states. Implicit in

the argument’s logic is the expectation that state delegations in a

LEGISLATIVE STUDIES QUARTERLY, 43, 1, February 2018 101

DOI: 10.1111/lsq.12179

V

C2017 Washington University in St. Louis

unicameral legislature will likely vote as a coordinated unit. This, in

turn, requires that the interests of legislators from the same state suff‌i-

ciently overlap, so that a policy that is held to be desirable by some

legislators from a given state will likely be held desirable by most or all

legislators from that state.

But this is a strong assumption. Policies may, after all, benef‌it vot-

ers in some parts of a state but not others. A policy to f‌inance public

transit infrastructure, for example, will likely be supported by legislators

representing districts in New York City but is unlikely to be of much

value to legislators representing districts upstate. The interests of people

in different electoral districts need not coincide simply because those dis-

tricts are located within the same state. (Thorpe [2010] f‌inds, for

example, that geographic disparities in the allocation of defense contracts

are tied to local, rather than statewide, economic conditions.) Over a

variety of policy issues, local interests may well be most salient in deter-

mining legislator behavior.

This article investigates the claim that bicameralism provides an

advantage to small states. I assume that legislators are concerned with

the interests of their immediate constituency, so that parochial and state

interests do not necessarily coincide. I show that, contrary to conven-

tional wisdom, bicameralism does not necessarily improve the welfare

of small states and may have the unintended consequence of harming

them. Moreover, bicameralism will tend to worsen outcomes for small

states precisely when those states would already have been disadvan-

taged under unicameralism. Hence, the case for bicameralism as a check

against policy usurpation by big states may be mistaken and warrants

further examination.

To study this question, I adapt the bargaining framework in Baron

and Ferejohn (1989) to model decision making in a bicameral legisla-

ture. The legislature must allocate a f‌ixed surplus among various

districts. Each district is contained within a state, and small states contain

fewer districts than big states. Representation in the lower house is by

district, while representation in the upper house is by state. Lower-house

legislators seek to maximize the allocation to their district alone, while

upper-house legislators seek to maximize the total allocation to the dis-

tricts within their state. If recognized, a legislator will propose an

allocation that maximizes funding to her constituency while earning the

support of majorities in both chambers. In equilibrium, the proposal will

target districts whose representatives are “cheap”—that is, who will

demand fewer concessions before supporting a proposal. The price of

legislators in turn depends on the size of the allocation they expect to

receive. Legislators who rationally expect to receive more will be more

102 Giri Parameswaran

demanding than those who expect to receive less. Since allocating funds

to a given district benef‌its both the legislator representing that district

and the legislator representing the associated state, there are complemen-

tarities between the preferences of lower- and upper-house legislators

from the same state. This complementarity will be crucial to the analysis

that follows.

The main insight of this article is as follows: Under unicameralism,

while it is possible for big-state legislators to privilege one another when

forming coalitions (to the detriment of small states), such behavior

would often not be in their best interest. Indeed, a big-state legislator

may f‌ind it more rewarding to collaborate with small-state legislators,

especially if they are cheaper. Hence, under unicameralism, legislators

from the same state are not inherently coordinated to work together; they

are not natural coalition partners. Additionally, if the concern of policy

usurpation by big states is well founded, then it must be that small-state

legislators are cheaper coalition partners, since they would be expected

to gain less in the bargaining process. Hence, the fear that big states will

conspire against small states is least compelling precisely in those cases

when small states are expected to fare poorly.

Now, consider a bicameral legislature. The incentives for lower-

house legislators are analogous to the unicameral case. However, the

incentives for upper-house legislators differ markedly, since they are

concerned with the total allocation across their state. (This is consistent

with the empirical f‌indings in Lee [2004] and Lazarus and Steigerwalt

[2009] that the statewide interests of upper-house legislators differ from

the parochial interests of lower-house legislators from the same state.)

When recognized, an upper-house legislator will allocate funds across

each of the districts within his state. Given the complementarity in pref-

erences, this naturally earns the support of the lower-house legislators

from his state, resulting in coordinated support for policies among state

delegations. Furthermore, this coordinated support from the proposer’s

delegation reduces the number of out-of-state legislators upon whom the

proposer must expend resources to build a winning coalition. Since big-

state delegations are larger than small-state delegations, they are then

able to retain more resources within their own state, thus skewing the

distribution in their favor. I refer to this as the reduced-requirement

effect. By introducing state-level representatives, bicameralism has the

unintended consequence of creating incentives for precisely the sort of

coordinated behavior that generated the concern of policy usurpation by

big states under unicameralism.

Of course, by requiring that proposals receive majority support in

the upper house, where small-state legislators are more numerous,

103Bargaining and Bicameralism