Relation of Representation and the Party System to the Number of Seats Apportioned to a Legislative District

• Author: Ruth C. Silva
• DOI: 10.1177/106591296401700410
• Date: 01 December 1964
• Published date: 01 December 1964
• Subject Matter: Articles

RELATION OF REPRESENTATION AND THE
PARTY SYSTEM TO THE NUMBER OF SEATS
APPORTIONED TO A LEGISLATIVE DISTRICT
RUTH C. SILVA
Pennsylvania State University
HE
NUMBER OF SEATS apportioned to a legislative district has been
alleged to yield nine results. The single-member district, for example, has
been said to produce localism, less able candidates, shorter legislative tenure,
greater opportunity for gerrymandering, weak and decentralized parties, an empha-
sis on candidates rather than on parties or issues, a particular kind of legislative re-
sponsibility to the electorate, a less representative legislature, and a two-party rather
than a multi-party system. The multi-member district allegedly produces nine oppo-
site results. Both facts and logic show, however, that the first seven of these &dquo;results&dquo;
are spurious in the case of either the single-member or the multi-member district.1
Although the number of seats apportioned to a district does influence both the party
system and the representative character of the legislature, the connecting variable
is the electoral system used in the district.
Analysis of the multi-member and the single-member constituency necessarily
raises the question of the electoral system, which in turn raises a normative and more
subjective question concerning the legislature’s proper function. One electoral sys-
tem may be suitable for a legislature that plays one role but may also impair the legis-
lature’s ability to perform other important duties. In the United States, for example,
the legislature is presumed to be a body that should not only formulate public policies
but also reflect various interests in the community. Formulating policies requires
a legislature with a working majority. A voting system that produces legislative
majorities may do so, however, precisely because it does not produce a representative
legislature.2
2
BLOC ELECTORAL SYSTEMS
Unlike a proportionate electoral system, which can be used only in multi-mem-
ber districts, a bloc electoral system can be used either in single-member or in multi-
member districts. In fact, while both single-member and multi-member constitu-
encies are used for the election of state legislators, one of these bloc systems -
the
single-ballot-plurality vote -
is used for the election of 98 of the 99 state legislative
chambers.,3 This system allows each voter to ballot for a number of candidates equal
1
See "Compared Values of the Single- and the Multi-Member Legislative District," Western
Political Quarterly, 17 (September 1964), pp. 504-16.
’ Viewing the legislature as a body whose primary function is to formulate policy is largely an
American conception. While Europeans generally call their representative assemblies
"parliaments" or talking bodies, Americans call theirs "legislatures" or law-proposing
bodies. For discussions of these two concepts of representation and their historical de-
velopment, see Charles A. Beard and John D. Lewis, "Representative Government in Evo-
lution," American Political Science Review, 26 (April 1932), 223-40; Alfred DeGrazia,
Public and Republic (New York: Knopf, 1951); Harold F. Gosnell, Democracy: The
Threshold of Freedom (New York: Ronald, 1948), p. 149n; Robert Luce, Legislative
Principles ( Boston : Houghton Mifflin, 1930).
3
The lower house of the Illinois Legislature is elected by cumulative voting.
742

---

743
to the number of seats apportioned to his constituency, and the candidates with a
plurality in the district are elected. Allegedly, this arrangement tends to produce
working legislative majorities, enables a political party to salvage a legislative ma-
jority from defeat at the polls, gives little or no representation to more than two
parties, and elects legislators who are not the choice of an electoral majority in their
respective districts.4
4
Because of the Liberal party’s persistence in New York politics, for example, no
party has polled a majority of the statewide popular vote in half of the state’s legis-
lative elections during the past decade. In every case when no electoral majority
existed, however, the single-ballot-plurality vote and the single-member district pro-
duced a legislative majority. In five of New York’s ten most recent elections, the
system translated a slim popular margin into a safe legislative majority. (See Table
I. ) This is not surprising since a bloc electoral system usually magnifies the legislative
strength of the party with the statewide electoral plurality. In five other elections,
however, the Republicans captured a legislative majority despite their defeat at the
polls. This phenomenon was caused largely by two factors. First, the average Demo-
cratic district is more populous than the average Republican district -
a result of
misapportionment that generally favors Republican areas.5 Second, district lines
are drawn so that Democratic electoral strength is concentrated in certain districts.
Consequently, the Democrats roll up large electoral majorities in these districts while
they lose other districts by a narrow margin&dquo; -
a situation that obviously suggests
gerrymandering.7
7
It would be incorrect to imply, however, that the disparity between a party’s
proportion of the statewide popular vote and its proportion of legislative seats is
caused wholly by misapportionment and gerrymandering and not by the single-
ballot-plurality vote. Even if an equal number of votes were cast in each district and
even if district lines were
drawn according to a set and impartial standard, the single-
member district and the single-ballot-plurality vote would rarely, if ever, produce a
legislature which accurately reflects the various parties’ statewide electoral strength.
Only in the unlikely event that each party carried a percentage of districts equal to
4
See Enid Lakeman and James D. Lambert, Voting in Democracies (London: Faber & Faber,
1955), pp. 25-49; W. J. M. MacKenzie, Free Elections (New York: Rinehart, 1958),
pp. 50-53.
’ Gus Tyler and David I. Wells, "New York ’Constitutionally Republican,’ " in Malcolm E.
Jewell (ed.), The Politics of Reapportionment (New York: Atherton, 1962), pp. 221,
232. This is particularly true of the Assembly. Ruth C. Silva, "Apportionment of the New
York Assembly," 31 Ford. L. Rev. 1, 13-28, 44-47 (October 1962) ; and "Apportionment
of the New York Senate," 30 Ford. L. Rev. 595, 603-38 (April 1962). But see Ruth C.
Silva, "The Population Base for Apportionment of the New York Legislature," 32 Ford.
L. Rev. 1, 21-30 (October 1963) ; "Making Votes Count," National Civic Review, 52
(October 1963), 489-92.
6
Ruth C. Silva, "Legislative Representation — With Special Reference to New York," Law and
Contemporary Problems, 27 (Summer 1962), 408, 420-26; "Compared Values ... ,"
pp. 512-14.
7
The Legislature draws the boundaries of senatorial districts. N.Y. Const. Art. III, § 4 (1894).
In any county entitled to more than one assemblyman, the local authority divides the
county into single-member assembly districts. Ibid., Art. III § 5 (1894). This arrange-
ment means that Republicans draw the boundaries of senatorial districts and that Demo-
crats draw the boundaries of assembly districts in New York City and Albany County
while Republicans draw the boundaries of assembly districts in most of the other multi-
member counties.

---

744
TABLE I
COMPARISON OF LEGISLATIVE AND ELECTORAL STRENGTH IN NEW YORK
(in percentages)
* All votes cast for a candidate are included in his party’s vote even though he may have
polled some of these votes on another ticket. Liberal votes cast for a Democrat, for example,
are included as Democratic votes.
f The difference equals the percentage of seats won minus the percentage of the statewide
vote -
e.g., 56.67-48.87 =
-f-7.80.
t The distortion equals the difference divided by the percentage of the statewide vote -
e.g., 7.80-* 48.87 =
15.96. Thus, in being translated into Assembly seats, the Republicans’ 1962
2
Assembly vote was magnified 15.96 per cent.
SouRCE : Data tabulated from N.Y. Legislative Manual, 1955-1963, and data on the 1962
legislative elections supplied by the N.Y. Secretary of State.
its percentage of the statewide vote, would the legislature reflect the popular vote
with precision. Misapportionment and gerrymandering merely exaggerate or invert
the legislature’s so-called unrepresentative character, which the electoral system
would produce in some measure without misapportionment and gerrymandering.
Nor would it be correct to imply that the so-called unrepresentative character
of New York’s legislature is caused by the single-member district rather than by the
single-ballot-plurality vote. Use of this electoral system in multi-member districts
generally produces an even less representative legislature. Since party is usually the
primary factor in determining how most votes are cast, the majority party’s weakest
candidate will usually be elected over the opposition’s strongest candidate in a multi-
member constituency. Two parties will win seats in such a constituency only if they

---

745
are so evenly matched that the independent or floating vote is sufficient to elect the
minority party’s strongest candidate or candidates.’ Such marginal districts are rare,
however, in most states where the single-ballot-plurality vote has been combined
with the multi-member district. In Pennsylvania’s legislative elections of 1962, for
example, only two of the state’s 41 multi-member districts split their representation
...