The U.S. Supreme Court's 1962 decision in Baker v. Carr
was a sharp departure from that
Court's long-standing policy of judicial nonintervention in redistricting cases.
Many
redistricting cases that reached the Supreme Court in the next several years were
challenges to situations in which differences in population among legislative districts, or
in the number of people represented by members of a single legislative body, were so
great that—viewed from the perspective of 1990—they are not only obviously
impermissible but also ludicrous. These situations had nearly all disappeared either before
or during the post-1970 round of redistricting. The purpose of this chapter is to discuss the
constitutional requirement of equal population among state legislative and congressional
districts as it has developed in federal cases decided since Baker v. Carr.
One Person, One Vote—Background
Although the history-making decision in Baker v. Carr held that state legislative (and, by
implication, congressional) districting cases are justiciable, and expressed confidence that
courts would prove able to “fashion relief” where constitutional violations might be
found,
the Supreme Court did not provide specific standards or criteria for judicial review
of state districting, or for judicial remedies.
Development by the Supreme Court of the substantive case law standards that govern
state legislative and congressional districting began the following year with Gray v.
Sanders,
in which the Court held that unit voting systems are unconstitutional per se.
That decision included the now familiar assertion by Justice Douglas that “[t]he
conception of political equality from the Declaration of Independence, to Lincoln's
Gettysburg Address, to the Fifteenth, Seventeenth, and Nineteenth Amendments can mean
only one thing—one person, one vote.”
Measuring Population Equality Among Districts
How is the degree of population equality (or inequality) among legislative or congressional districts measured? A clear understanding of the measures available and those used by the courts—and by the drafters of redistricting plans—is essential. The courts have not always been consistent or precise in their terms, and this has led to considerable misunderstanding and confusion. For example, courts have sometimes used terms with definite statistical meaning in a general, nonstatistical manner. A definition of terms, therefore, may be helpful at this point.
Ideal population. A logical starting point is the “ideal” district population. In a single-member district plan, the “ideal” district population is equal to the total state population divided by the total number of districts. (For example, if a state's population is 4 million and there are 40 legislative districts, the “ideal” district population is 100,000.) For purposes of this discussion, it will be assumed that a single-member districting plan is being considered. In districting plans that use multimember districts, the “ideal” population is more properly expressed as the “ideal” population per representative and is obtained by dividing the total state population by the total number of representatives. The number of representatives rather than the number of districts would thus be used in performing statistical calculations for districting plans that employ multimember districts.
There is, then, the need to express the degree to which: l) an individual district's population varies, or differs, from the “ideal;” and 2) all districts collectively vary, or differ, in population from the “ideal.”
Deviation. The degree by which a single district's population varies from the “ideal” may be stated in terms of “absolute deviation” or “relative deviation.” The “absolute deviation” is equal to the difference between its population and the “ideal” population, expressed as a plus (+) or minus (-) number, meaning that the district's population exceeds or falls short of the “ideal” by that number of people. (For example, if the “ideal” population is 100,000 and a given district has a population of 102,000, its “absolute deviation” is +2,000.) “Relative deviation” is the more commonly used measure and is attained by dividing the district's absolute deviation by the “ideal” population. The resulting quotient indicates the proportion by which the district's population exceeds or falls short of the “ideal” population, and usually is expressed as a percentage of the “ideal” population. (In the preceding example, the “relative deviation” is +2 percent.)
Several methods of measuring the extent to which populations of all the districts in a plan vary, or differ collectively from the “ideal,” are available.
Mean deviation. A frequently used measure is the “mean deviation,” expressed in absolute or relative terms. The “absolute mean deviation” of a set of districts from the “ideal” is equal to the sum of the absolute deviations of all the districts (disregarding “+” or “-“ signs) divided by the total number of districts. The “relative mean deviation” is equal to the sum of the individual district relative deviations (disregarding “+” or “-“ signs) divided by the total number of districts.
Overall range. Perhaps the most commonly used measure of population equality, or inequality, of all districts in a plan is overall range, which again may be expressed in absolute or relative terms. The “range” is a statement of the population deviations of the most populous district and the least populous district, expressed in either absolute or relative terms. (For example, if the ideal district population is 100,000, the largest district in the plan has a population of 102,000, and the smallest district has a population of 99,000, then the range is +2,000 and -1,000, or +2 percent and -1 percent.)
The “overall range” is the difference in population between the largest and the smallest
districts, expressed either as a percentage or as the number of people. (In the preceding
example, the “overall range” is 3 percent or 3,000 people.) Although the courts normally
measure a plan using the statistician’s “overall range,” they almost always call it
something else, such as “maximum deviation.”
None of the foregoing measures provides a full picture of the degree of population equality, or inequality, and perhaps several measures should be used in evaluating any set of districts. (For example, the overall range may be a large one because of the large deviation of only one district, but all the remaining districts may be clustered closely around the “ideal.” The use of “mean deviation” would reveal this.) For purposes of comparison and clarity, this book uses the measures of relative overall range and relative mean deviation expressed simply as overall range and mean deviation. Table 3 shows the various measures in mathematical form.
Table 3. Statistical Terminology for Districting
IDEAL DISTRICT POPULATION |
= |
State Population |
|
INDIVIDUAL DISTRICTS |
|
ABSOLUTE DEVIATION |
= |
District Population - |
RELATIVE DEVIATION |
= |
Absolute Deviation |
|
ALL DISTRICTS |
|
MEAN DEVIATION* |
= |
Sum of All Deviations |
RANGE* |
= |
Largest Positive Deviation and Largest Negative Deviation |
OVERALL RANGE* |
= |
Largest Positive Deviation +
Largest Negative Deviation |
*Can Be “Absolute” or “Relative” |
|
|
Source: NCSL, 1999.
Two Different Standards For Congressional And Legislative Districts
The equal population requirements for congressional districts and legislative districts do
not rest on the same stone in the constitutional foundation of the Republic. The equal
population standard for congressional districts, first enunciated by the Supreme Court in
Wesberry v. Sanders,
arises from Article I, Section 2, of the Constitution, “Representatives
. . . shall be apportioned among the several states . . . according to their respective
numbers. . . .” This standard has been strictly interpreted by the Court in Kirkpatrick v.
Preisler,
White v. Weiser
and Karcher v. Daggett.
By contrast, the Supreme Court held
in Reynolds v. Sims
that it is the Equal Protection Clause of the Fourteenth Amendment
that requires states to construct legislative districts that are substantially equal in
population. The degree of mathematical equality required by the Supreme Court has
varied between congressional districts and state legislative districts.
The Standard For Drawing Congressional Districts—Strict Equality
In Wesberry v. Sanders (1964), the Supreme Court held that the population of
congressional districts in the same state must be as nearly equal in population as
practicable.
In April 1969, the Supreme Court decided Kirkpatrick v. Preisler,
a case involving
congressional districts drawn by the Missouri General Assembly. The 10 districts had an
overall range of approximately 6 percent. Writing for a five-member majority, Justice
Brennan found that the plan failed to satisfy the “as nearly as practicable” standard of
population equality the Court had earlier enunciated in Wesberry v. Sanders. The
Kirkpatrick opinion specifically rejected the suggestion that there is a point at which
population differences among districts becomes de minimis and held that, insofar as a
state fails to achieve mathematical equality among districts, it must either show that the
variances are unavoidable or specifically justify the variances.
The opinion went on to
reject several purported justifications advanced by Missouri.
The justifications rejected included a desire to avoid fragmenting either political
subdivisions or areas with distinct economic and social interests, considerations of
practical politics, and even an asserted preference for geographically compact districts.
Also, the majority opinion held that Missouri had failed to show any systematic
relationship between its congressional district population disparities and either of two
other factors offered as justifications—varying proportions of eligible voters to total
population and projected future population shifts among districts.
(The Court did not
flatly rule out the latter consideration, but it said such projections must be well
documented and uniformly applied.)
In White v. Weiser, a 1973 case involving Texas congressional districts,
the U.S.
Supreme Court ruled that, although the overall range among Texas' 24 congressional
districts was smaller than that invalidated in Kirkpatrick v. Preisler in 1969, the Texas
districts were not as mathematically equal as reasonably possible and were therefore
unacceptable. The Court specifically rejected an argument that the variances resulted
from the Texas Legislature's attempt to avoid fragmenting political subdivisions.
Ten years later, in Karcher v. Daggett,
the U.S. Supreme Court reaffirmed its position in
Kirkpatrick v. Preisler that there is no level of population inequality among congressional
districts that is too small to worry about, as long as those challenging the plan can show
that the inequality could have been avoided. As Justice Brennan wrote for the 5-4
majority: “We thus reaffirm that there are no de minimis population variations, which
could practicably be avoided, but which nonetheless meet the standard of Art. I, Sec. 2,
without justification.”
The congressional redistricting plan drawn by the New Jersey Legislature had an overall
range of 3,674 people, or .6984 percent.
The plaintiffs showed that at least one other
plan before the legislature had a “maximum population difference” (overall range) of only
2,375 people or .4514 percent,
thus carrying their burden of proving that the population
differences could have been reduced or eliminated by a good faith effort to draw districts
of equal population. The Court also noted that the population differences could have
been reduced by the simple device of transferring entire political subdivisions of known
population between contiguous districts.
Once the plaintiffs had shown that the population differences could have been reduced,
the state had the burden of proving that each significant variation from the ideal was
necessary to achieve “some legitimate state objective.”
Any number of consistently applied legislative policies might justify some variance,
including, for instance, making districts compact, respecting municipal boundaries,
preserving the cores of prior districts, and avoiding contests between incumbent
Representatives ... . The State must, however, show with some specificity that a
particular objective required the specific deviations in its plan, rather than simply
relying on general assertions. The showing required to justify population deviations
is flexible, depending on the size of the deviations, the importance of the State's
interests, the consistency with which the plan as a whole reflects those interests, and
the availability of alternatives that might substantially vindicate those interests yet
approximate population equality more closely. By necessity, whether deviations are
justified requires case-by-case attention to these factors.
The New Jersey Legislature attempted to justify the population deviations as necessary to
preserve the voting strength of racial minority groups. But the evidence was directed at
only one of the 14 districts, so the Court found that New Jersey had failed to justify the
deviations in the other districts, and affirmed the lower court's rejection of the plan.
In 1997, Abrams v. Johnson
involved a challenge to a congressional plan on the basis of
violation of the guarantee of one person, one vote. This challenge was to a plan drawn
by the federal district court on remand after Miller v. Johnson
was decided, affirming
that Georgia’s Eleventh District was unconstitutional because race was a predominant
factor in drawing the district.
A subsequent amended complaint also challenged the
Second District on the same grounds and the trial court found that district also to be
unconstitutional on the same grounds. The court gave the Georgia legislature an
opportunity to redraw the plan, but a special session of the Georgia legislature adjourned
without adopting a new plan when the House of Representatives and the Senate could not
agree on the number of minority-majority districts to be adopted.
The court then drew
the plan at issue in this case.
The plan drawn by the district court had an “overall population deviation” (overall range)
of 0.35 percent and an “average deviation” of 0.11 percent.
No other plan submitted or
considered by the court that was otherwise judged to be a constitutional plan had a lower
overall range.
In a 5-4 decision, Justice Kennedy delivered the opinion affirming the
decision of the district court. (The dissent focused on grounds other than one person, one
vote.)
The Supreme Court affirmed the principle established in Karcher that “absolute population
equality [is] the paramount objective”
in congressional plans and the holdings in
Chapman v. Meir
and Connor v. Finch
that “[w]ith a court plan, any deviation from
approximate population equality must be supported by enunciation of historically
significant state policy or unique features.”
The Supreme Court found that the district
court did comply with the applicable standards for variance from absolute population
equality and effectively enunciated the specific policies and features that justified the
variances.
In addition to affirming the status quo regarding one person, one vote standards in
congressional redistricting, the court made two other important findings. The first was
that, even if the plan had failed on the basis of the one person, one vote challenge, “the
solution would not be adoption of the constitutionally infirm, because race based, plans
of appellants ... . Rather, we would require some very minor changes in the court’s
plan—a few shiftings of precincts—to even out districts with the greatest deviations.”
The second important finding is the acknowledgment by the court:
That exercise, however, and appellant’s objections to the court plan’s slight
population deviations, are increasingly futile. We are now more than six years from
the last census, on which appellant’s data is based. The difference between the
court plan’s average deviation (0.11%) and the Illustrative Plan’s (0.07%) is 0.04%,
which represents 328 people out of a perfect district population of 588,928. The
population of Georgia has not stood still. Georgia is one of the fastest growing
States, and continues to undergo population shifts and changes ... . In light of these
changes, the tinkerings appellants propose would not reflect Georgia’s true
population distribution in any event. The Karcher Court, in explaining the absolute
equality standard acknowledged that ‘census data are not perfect’ and that
‘population counts for particular localities are outdated long before they are
completed.’ 462 U.S., at 732. Karcher was written only two years from the previous
census, however, and we are now more than six years from one. The magnitude of
population shifts since the census is far greater here than was likely to be so in
Karcher. These equitable considerations disfavor requiring yet another
reapportionment to correct the deviation.
The Standard For Drawing Legislative Plans
Reynolds v. Sims
is the cornerstone in the development of the federal judiciary's
population variance standards for state legislative districting. The case is notable both for
the ruling that both houses of a bicameral state legislature must be districted on a
population basis, and for comments about what population-based districting requires. The
opinion by Chief Justice Warren includes the often-quoted comment that “mathematical
nicety is not a constitutional requisite,”
but nevertheless states that “the overriding
objective must be substantial equality of population among the various districts.”
The
Court declined at that time to express any view as to what degree of population equality
would or would not be held constitutional, observing that “what is marginally permissible
in one State may be unsatisfactory in another depending upon the particular
circumstances of the case.”
An especially significant comment—as matters later developed—differentiated between congressional and legislative districting. The Warren opinion said:
[S]ome distinctions may well be made between congressional and state legislative
representation. Since, almost invariably, there is a significantly larger number of
seats in state legislative bodies to be distributed within a state than congressional
seats, it may be feasible to use political subdivision lines to a greater extent in
establishing state legislative districts than in congressional districting while still
affording adequate representation to all parts of the State.
Mahan v. Howell—Congressional And Legislative Districting Differentiated
Although the Supreme Court recognized a distinction between congressional and legislative districting, it did not specify what differences in equality of population this might permit.
This uncertainty prevailed for nearly nine years, a period during which the 1970 census
was completed and the states undertook—and, in many cases completed—legislative
redistricting based on that census. Then, in February 1973, the U.S. Supreme Court
announced its decision in Mahan v. Howell,
a rather complicated challenge to Virginia's
legislative districting plan. Mahan involved issues of the constitutionality of multimember
districts and the treatment of certain naval personnel “home-ported” in Norfolk, Virginia,
as well as a challenge to the overall range of the plan enacted by the Virginia General
Assembly. A federal district court, concluding that the “maximum deviation” (overall
range) among house districts was 16.4 percent, declared the plan unconstitutional by
reason of that population disparity.
The Supreme Court majority opinion recounted some of the facts stated and conclusions reached in Reynolds, including those factors the Court had suggested might justify limited departure from strict population equality in legislative, as opposed to congressional, districting. The opinion, by Justice Rehnquist, stated:
Thus, whereas population alone has been the sole criterion of constitutionality in
congressional redistricting under Art. I, Sec. 2 [of the United States Constitution],
broader latitude has been afforded the States under the Equal Protection Clause in
state legislative redistricting ... . The dichotomy between the two lines of cases has
consistently been maintained.
The majority took note of the Virginia General Assembly's state constitutional authority to
enact local legislation dealing with particular political subdivisions. They found that this
legislative function was a significant and a substantial aspect of the Virginia legislature's
powers and practices and thus justified an attempt to preserve political subdivision
boundaries in drawing House of Delegates’ districts. The majority concluded that, while
the resulting overall range among house districts “may well approach tolerable limits, we
do not believe it exceeds them.”
Chief Justice Burger and justices Stewart, White and
Blackmun joined the majority opinion; Justice Powell took no part.
Dissenting justices Brennan, Douglas and Marshall sought, at some length, to refute the
contention that a distinction between standards for legislative and congressional
districting had been maintained by the Court.
They suggested that the “total deviation”
(overall range) in the Virginia House approached 25 percent, a figure they said placed the
plan in the same range as several others invalidated by the Supreme Court in the period
between 1967 and 1971.
(The differing conclusions as to the overall range of the
Virginia plan stem from alternative ways of treating the effect of floterial districts included
in the plan.)
The 10 Percent Standard
The distinction between the standard of population equality demanded in congressional
districting and that required in state legislative districting again was recognized and the
legislative districting standard somewhat clarified in June 1973 by the U.S. Supreme Court
decisions in Gaffney v. Cummings,
a Connecticut case, and White v. Regester,
a Texas
case. Each of these cases also arose from a state-drawn legislative districting plan that had
been challenged and struck down by a federal district court.
Gaffney v. Cummings involved a plan prepared by a bipartisan commission appointed
pursuant to Connecticut law. The plan's “total maximum deviation” (overall range) was
1.8 percent in the Senate and 7.8 percent in the House,
and one of its objectives was
described as “political fairness;” i.e., the political makeup of each house should roughly
reflect the proportion of the statewide total vote received by candidates of each major
party.
White v. Regester concerned the distribution of Texas House seats in a plan,
drawn by the state Legislative Redistricting Board, which had a “total variation” (overall
range) of 9.9 percent.
It was challenged both on that ground and on the complaint that
certain multimember districts invidiously discriminated against particular racial or ethnic
groups. (The latter complaint was found valid by the District Court and upheld by the
Supreme Court—that aspect of the case is discussed in chapter 5.)
The majority opinion in each of these cases was written by Justice White for himself, Chief Justice Burger, and justices Steward, Blackmun and Rehnquist, the same group that had formed the majority in Mahan v. Howell, as well as Justice Powell, who had taken no part in Mahan. In the Gaffney opinion, after again asserting that the Supreme Court had always maintained a distinction between congressional and state legislative districting cases, Justice White said:
Although requiring that the population variations among legislative districts in
Mahan be justified by substantial state considerations, we did not hold that in state
legislative cases any deviations from perfect population equality in the districts,
however small, make out prima facie equal protection violations and require that the
contested reapportionments be struck down absent adequate state justification.
The Gaffney opinion continued by holding that no prima facie violation of the Equal
Protection Clause had been shown, and that the “political fairness” objective of Gaffney
did not invalidate the plan.
Similarly, in the White opinion, the Supreme Court majority
declared: “Insofar as the District Court's judgment rested on the conclusion that the
population differential [i.e., overall range] of 9.9 percent ... made out a prima facie equal
protection violation under the Fourteenth Amendment, absent special justification, the
court was in error.”
The majority opinion observed: “Very likely, larger differences
between districts would not be tolerable without justification 'based on legitimate
considerations incident to the effectuation of a rational state policy.'”
justices Brennan, Douglas and Marshall, dissenting in both Gaffney and White with a
single opinion, asserted that the majority opinions in the two cases had, in effect,
established a 10 percent de minimis rule for state legislative districting, with states not
required even to try to justify “total deviations” (overall ranges) of that or a lesser degree.
Dicta in later Supreme Court decisions in Chapman v. Meier,
Connor v. Finch,
and
Brown v. Thomson
have confirmed that assertion.
Chapman v. Meier,
decided January 27, 1975, involved a redistricting of the North
Dakota Senate devised by a federal court, under which the “total variance” (overall range)
among districts was slightly more than 20 percent. Justice Blackmun, writing for the
unanimous Court, recalled that state-drawn redistricting plans having less than a 10
percent “deviation” (overall range), and where there was no showing of invidious
discrimination, were found valid in Gaffney and White, and that a “total population
variance” (overall range) of 16.4 percent was subject to court scrutiny but was found
justified in Mahan because it served to implement a rational state policy. He held that
none of the reasons advanced—absence of a particular racial or political group whose
voting power was minimized or canceled, sparse population of the state generally, and
desire both to preserve political subdivision boundaries and to continue an asserted
tradition of dividing the state along political subdivision lines and along the Missouri
River—was sufficient to justify the “variance” (overall range) of more than 20 percent.
In Connor v. Finch,
a case from Mississippi decided in May 1977, the Supreme Court's
majority opinion, by Justice Stewart, stated that the “maximum deviation” (overall range)
of the Mississippi redistricting plan at issue was computed by the federal district court
(which drew the plan) to be 16.5 percent for the Senate and 19.3 percent for the House.
The opinion noted that these figures “substantially exceed the 'under-10 percent'
deviations the Court has previously considered to be of prima facie constitutional validity
only in the context of legislatively enacted apportionments,” and concluded that the
district court failed to cite any unique feature of the Mississippi political structure that
would justify a “deviation” (overall range) of such magnitude.
The plan was therefore
invalidated. (The only dissenter was Justice Powell, who believed the plan should have
been remanded to the district court for such limited changes as were necessary to bring it
into conformity with Supreme Court guidelines.)
The only legislative reapportionment case involving population equality arising from the
1980 census and decided by the U.S. Supreme Court was Brown v. Thomson.
It
concerned the Wyoming House. The Wyoming constitution requires that each county
constitute a legislative district, to be apportioned at least one senator and one
representative. Wyoming had 23 counties, among which 64 representatives had been
apportioned in accordance with the 1980 census. Niobrara County, the least populous,
had 2,924 people, or 60 percent below the ideal of 7,337. The average deviation was 16
percent, and the “maximum deviation” (overall range) was 89 percent.
Justice Powell, in the majority opinion joined by Justices Burger, Rehnquist, O'Connor
and Stevens, put to rest any doubt that the Court intends to use a 10 percent standard to
judge legislative apportionment plans by saying that the Court's decisions had established,
as a general matter, that a legislative apportionment plan with a “maximum population
deviation” (overall range) under 10 percent is insufficient to make out a prima facie case
of invidious discrimination under the Fourteenth Amendment so as to require justification
by the state.
A plan with larger disparities in population, however, creates a prima facie case of
discrimination and therefore must be justified by the State ... . The ultimate inquiry,
therefore, is whether the legislature's plan “may reasonably be said to advance [a]
rational state policy” and, if so, “whether the population disparities among the
districts that have resulted from the pursuit of this plan exceed constitutional
limits.”
Justice Powell stated that consideration must be given to the character as well as the
degree of deviations when analyzing a state redistricting plan: “The consistency of
application and the neutrality of effect of the nonpopulation criteria must be considered
along with the size of the population disparities in determining whether a state legislative
apportionment plan contravenes the Equal Protection Clause.”
Justice Powell concluded that Wyoming's constitutional policy—followed since
statehood—of using counties as representative districts and ensuring that each county had
at least one representative, was supported by substantial and legitimate state concerns,
and had been applied in a manner free from any taint of arbitrariness or discrimination.
He also found that the population deviations were no greater than necessary to preserve
counties as representative districts, and that there was no evidence of a built-in bias
tending to favor particular interests or geographical areas.
Proving Discrimination Within The 10 Percent Range
States should not assume that any legislative districting plan having less than a 10 percent
overall range is safe from successful challenge. Even if the Court is prepared to allow the
states some leeway from redistricting perfection, now that the basic law of population
equality is well established, it is unlikely that the justices would be unduly hesitant to
strike down a plan having an overall range of less than 10 percent if a challenger were to
succeed in raising a suspicion that the plan was not a good faith effort overall or that there
was something suspect about the districts involved.
However, the decisions in Gaffney,
White and Brown indicate that the challenger of such a plan has the initial burden of
showing that the plan violates the Equal Protection Clause.
The Supreme Court said in
the White case that it could not “glean an equal protection violation from the single fact
that two legislative districts in Texas differ from one another by as much as 9.9 percent ...
.”
And it indicated in Gaffney that a showing by the plaintiff that an alternative plan with
a lower “variation” (overall range) could be devised is not in itself sufficient to require a
federal court to invalidate a plan adopted by a state legislature.
A relatively high mean deviation, even within the context of an overall range of less than
10 percent, may make it easier for a challenger to meet the burden of establishing an
equal protection violation. In Gaffney, the majority opinion pointed out that, although the
“total maximum deviations” (overall ranges) were 7.8 percent in the House and 1.8
percent in the Senate,
the respective mean deviations were only 1.9 percent and .45
percent.
Similarly, the White opinion contrasts the 9.9 percent “total variance” of the
Texas House districting plan with its mean deviation of 1.8 percent.
Finally, it should be noted that there is nothing in the U.S. Constitution or case law to
prevent state courts from imposing stricter standards of population equality, under state
constitutions, than the federal courts demand. State legislatures may also impose stricter
population standards of equality in their redistricting law.
“Rational State Policies” That Justify Exceeding The 10 Percent Standard
If a state enacts or adopts a plan with an overall population range of more than 10 percent
in either house and the plan is challenged in federal court, the state will have the burden
of showing both that the more than 10 percent overall range is necessary to implement a
“rational state policy”
and that it does not dilute or take away the voting strength of any
particular group of citizens. The obvious question, then, is: What are the criteria of a
“rational state policy” that are constitutionally relevant to legislative districting?
Affording representation to political subdivisions. The majority opinion in Reynolds v.
Sims stated: “So long as the divergences from a strict population standard are based on
legitimate considerations incident to the effectuation of a rational state policy, some
deviations from the equal-population principle are constitutionally permissible”
in
legislative districting. That opinion continued: “Considerations of area alone provide an
insufficient justification for deviations from the equal-population principle.”
It also
observed:
A consideration that appears to be of more substance in justifying some deviations
from population-based representation in state legislatures is that of insuring some
voice to political subdivisions, as political subdivisions. Several factors make more
than insubstantial claims that a State can rationally consider according political
subdivisions some independent representation in at least one body of the state
legislature, as long as the basic standard of equality of population among districts is
maintained. Local governmental entities are frequently charged with various
responsibilities incident to the operation of state government. In many States much
of the legislature's activity involves the enactment of so-called local legislation,
directed only to the concerns of particular political subdivisions. And a state may
legitimately desire to construct districts along political subdivision lines to deter the
possibilities of gerrymandering.
In Mahan v. Howell the majority reaffirmed the foregoing position and stated:
We are not prepared to say that the decision of the people of Virginia to grant the
General Assembly the power to enact local legislation dealing with the political
subdivisions is irrational. And if that be so, the decision of the General Assembly to
provide representation to subdivisions qua subdivisions in order to implement that
constitutional power is likewise valid when measured against the Equal Protection
Clause of the Fourteenth Amendment.
The majority opinion went on to hold that Virginia's “plan for apportionment of the
House of Delegates may reasonably be said to advance the rational state policy of
respecting the boundaries of political subdivisions.”
In Brown v. Thomson,
the Supreme Court showed that it was willing to go a long way to
support a state's constitutional policy of using counties as legislative districts and ensuring
that each county had at least one representative, even when that meant upholding a plan
with a “maximum deviation” (overall range) of 89 percent. Writing for the majority,
Justice Powell found that the policy, followed since statehood, was supported by
substantial and legitimate state concerns and had been applied in a manner free from any
taint of arbitrariness or discrimination. He also found that the population deviations were
not greater than necessary to preserve counties as representative districts, and that there
was no evidence of a built-in bias tending to favor particular interests or geographical
areas.
In 1994, the district court in Quilter v. Voinovich
relied on Mahan v. Howell and Brown
v. Thomson in ruling that Ohio's legislative district plan with a “total deviation” (overall
range) of 13.81 percent for House districts and 10.54 percent for Senate districts did not
violate the one person, one vote guarantee because the deviation was justified by the
rational state policy of preserving county lines. The Court stated, in distinguishing Quilter
from other cases where a rational state policy argument was rejected, that Ohio had a
clearly stated constitutional policy, the plan advanced that policy, and that the deviations
resulting from the plan were not constitutionally excessive.
The Court also pointed out
that the plan was not advanced arbitrarily as evidenced by the fact that the plan did in
fact preserve whole counties within the applicable population limits.
Affording representation to political subdivisions was, as of 1998, the only “rational state
policy” that had actually been accepted by the Supreme Court as justification for a
legislative districting plan that had an overall range greater than 10 percent. The record
since 1973 suggests that the Supreme Court is not easily persuaded to accept even this
justification. It declined to do so in Chapman v. Meier, Connor v. Finch and Langsdon v.
Millsaps.
In its unanimous decision in Chapman the Court found: “It is far from apparent that North
Dakota policy currently requires or favors strict adherence to political lines.”
The
opinion also noted that it would have been possible to follow such a policy in North
Dakota and still achieve a significantly lower overall range.
Similarly, in a concurring
opinion in Connor, Justice Blackmun wrote:
I do not understand the [Supreme] Court to disapprove the District Court's decision
to use county lines as districting boundaries wherever possible, even though this
policy may cause a greater variation in district population than would otherwise be
appropriate for a court-ordered plan. The final plan adopted [by the District Court,
and subsequently appealed] appears to produce even greater population disparities
than necessary to effectuate the county boundary policy.
In Langsdon v. Millsaps, Tennessee's apportionment plan for its House of Representatives
had a “maximum deviation” (overall range) of 13.9 percent and divided 30 counties.
The state argued that the “variance” of 13.9 percent was necessary in order to comply
with the state constitutional prohibition on splitting counties, but the plaintiffs presented a
plan with a “total population variance” (overall range) of 9.847 percent that split only 27
counties.
The district court held, and the Supreme Court affirmed, that, although the
“constitutional provision against splitting counties is a rational state policy to be
considered in apportionment legislation,” in this case it was “patently unreasonable to
justify a 14% variance on the basis of not splitting counties” because, as plaintiffs had
shown, fewer counties may be split while decreasing the variance below the goal of 10
percent.
Mahan v. Howell
and Brown v. Thomson
are the only cases in which the Court has
found that affording representation to political subdivisions is a “rational state policy” that
justifies exceeding the 10 percent overall range. And in Brown v. Thomson, Wyoming's
policy of affording representation to political subdivisions may have been less important
to the result than was the peculiar posture in which the case was presented to the Court.
The appellants chose not to challenge the 89 percent overall range of the plan, but rather
to challenge only the effect of giving Niobrara County its own representative.
The
Legislature had provided that, if this representation for Niobrara County were held
unconstitutional, it would be combined with a neighboring county in a single
representative district; the house then would consist of 63 representatives.
In that event,
the overall range would be reduced to 66 percent and the average deviation to 13
percent.
Rather than find the whole plan unconstitutional and require the state to be
redistricted without respecting county boundaries (as it had done in Whitcomb v.
Chavis
for the Indiana General Assembly), the Court chose to confine itself to the
marginal effect of giving Niobrara County a representative and found that it was de
minimis: “These statistics make clear that the grant of a representative to Niobrara County
is not a significant cause of the population deviations that exist in Wyoming.”
Justice O'Connor, joined by Justice Stevens, concurred in the result but emphasized that it was only because the challenge was so narrowly drawn that she had voted to reject it:
I have the gravest doubts that a statewide legislative plan with an 89 percent
maximum deviation could survive constitutional scrutiny despite the presence of the
State's strong interest in preserving county boundaries. I join the Court's opinion on
the understanding that nothing in it suggests that this Court would uphold such a
scheme.
Justice Brennan, writing for himself and justices Marshall, Blackmun and White, dissented from the Court's holding, stressing:
[H]ow extraordinarily narrow [the Court's holding] is, and how empty of likely
precedential value ... . [I]t is unlikely that any future plaintiffs challenging a state
reapportionment scheme as unconstitutional will be so unwise as to limit their
challenge to the scheme's single most objectionable feature ... . [P]laintiffs
henceforth will know better than to exercise moderation or restraint in mounting
constitutional attacks on state apportionment statutes, lest they forfeit their small
claim by omitting to assert a big one.
The Court reaffirmed this narrow view of its holding in Brown by later citing it as authority
for the statement that “no case of ours has indicated that a deviation of some 78% could
ever be justified.”
Overall ranges in excess of 10 percent appear most likely to be upheld in cases of apportionment. In apportionment, members are apportioned among political subdivisions rather than a district being drawn for each member. However, such schemes have been upheld only when the number of members greatly exceeds the number of political subdivisions among which they are apportioned. In Reynolds v. Sims, the Alabama Legislature apportioned 106 seats among 67 counties, with each county being assured one seat. This resulted in an overall range of 16:1. In overturning the apportionment scheme, the Court stated:
[A] State may legitimately desire to construct districts along political subdivision
lines to deter the possibilities of gerrymandering. However, permitting deviations
from population-based representation does not mean that each local governmental
unit or political subdivision can be given separate representation, regardless of
population. Carried too far, a scheme of giving at least one seat in one house to
each political subdivision (for example, to each county) could easily result, in many
States, in a total subversion of the equal-population principle in that legislative
body. This would be especially true in a State where the number of counties is large
and many of them are sparsely populated, and the number of seats in the legislative
body being apportioned does not significantly exceed the number of counties. Such
a result, we conclude, would be constitutionally impermissible.
In Abate v. Mundt, 18 members of county government were apportioned among five cities, resulting in an overall range of 11.9 percent. The Court noted that:
[A] desire to preserve the integrity of political subdivisions may justify an
apportionment plan which departs from numerical equality ... . [T]he facts that local
legislative bodies frequently have fewer representatives than do their state and
national counterparts and that some local legislative districts may have a much
smaller population than do congressional and state legislative districts, lend support
to the argument that slightly greater percentage deviations may be tolerable for local
government apportionment schemes.
In Brown v. Thomson, the Wyoming Legislature was apportioning 64 representatives
among 23 counties. The Court noted with approval Schaefer v. Thomson,
where a
three-judge district court held that the apportionment of the Wyoming Senate of 25
senators among 23 counties, with the two largest counties each having two senators, so
far departed from the principle of population equality that it was unconstitutional.
But
the Court went on to state: “The Wyoming House of Representatives presents a different
case because the number of representatives is substantially larger than the number of
counties.”
In Board of Estimate v. Morris,
the charter of the city of New York had created a Board
of Estimate composed of eight ex-officio members, each of whom was an elected official.
The mayor, comptroller and city council president were elected city-wide and had two
votes each. The five borough presidents were elected by their boroughs and had one vote
each. The overall range of population among the five boroughs was 132 percent. The
overall range of population per vote on the board, including the three at-large members
with two votes each, was 78 percent. The Supreme Court held that the city had failed to
carry its burden of justifying “such a substantial departure from the one-person, one-vote
ideal.”
Thus, the use of apportionment among political subdivisions may afford an acceptable
scheme where the number of seats apportioned is substantially larger than the number of
political subdivisions among which they are apportioned. The apportionment of 435 seats
in the federal House of Representatives among the 50 states results in an overall range of
76.2 percent.
The Supreme Court has maintained the status quo regarding population equity arising
from the 1990 census in legislative redistricting cases. In Voinovich v. Quilter,
the
Court in a unanimous opinion delivered by Justice O’Connor reversed a federal district
court opinion holding that “total deviations in excess of 10% cannot be justified by a
policy of preserving the boundaries of political subdivisions.”
Justice O’Connor validated the 10 percent de minimis standard for state legislative districts established in Gaffney and White, quoting Brown that:
[M]inor deviations from mathematical equality among state legislative districts are
insufficient to make out a prima facie case of invidious discrimination under the
Fourteenth Amendment so as to require justification by the State. Our decisions
have established as a general matter, that an apportionment plan with a maximum
population deviation under 10% falls within this category of minor deviations. A
plan with larger disparities in population, however, creates a prima facie case of
discrimination, and therefore must be justified by the State.
Once the plan’s deviation exceeds this threshold, a prime facie case of discrimination has been established and the court must then undertake to determine whether the plan advances a rational state policy and whether the deviation exceeds constitutional limits in accordance with Mahan and Brown.
Other state policies. The Supreme Court has, at least in dicta in Karcher v. Daggett, said that other state policies besides affording representation to political subdivisions may be used to justify a variance from equal population.
Any number of consistently applied legislative policies might justify some variance,
including, for instance, making districts compact, respecting municipal boundaries,
preserving the cores of prior districts, and avoiding contests between incumbent
Representatives ... . The State must, however, show with some specificity that a
particular objective required the specific deviations in its plan, rather than simply
relying on general assertions. The showing required to justify population deviations
is flexible, depending on the size of the deviations, the importance of the State's
interests, the consistency with which the plan as a whole reflects those interests, and
the availability of alternatives that might substantially vindicate those interests yet
approximate population equality more closely. By necessity, whether deviations are
justified requires case-by-case attention to these factors.
Since Karcher v. Daggett was a congressional redistricting case, where strict equality of population is required, these additional “rational state policies” would presumably apply with even greater force in a legislative redistricting case.
Legislative Plans Drawn By A Court
An interesting feature of Chapman v. Meier and Connor v. Finch is the Supreme Court's indication that, where it becomes necessary for a federal court to draw a state legislative districting plan, the 10 percent standard does not apply:
A court-ordered plan ... must be held to higher standards than a State's own plan. With a court plan, any deviation from approximate population equality must be supported by enunciation of historically significant state policy or unique features.
…
[U]nless there are persuasive justifications, a court-ordered reapportionment plan of
a state legislature ... must ordinarily achieve the goal of population equality with
little more than de minimis variation. Where important and significant state
considerations rationally mandate departure from these standards, it is the
reapportioning court's responsibility to articulate precisely why a plan ... with
minimal population variance cannot be adopted.
Although the 10 percent standard does not apply, “[t]his is not to say ... that court-ordered reapportionment of a state legislature must attain the mathematical preciseness
required for congressional redistricting.”
The U.S. Supreme Court has determined that the Apportionment Clause of Article I,
Section 2, of the U.S. Constitution requires that the population of all the congressional
districts in a state be as nearly equal in population as practicable.
There are no de
minimis variations, which could practicably be avoided, without justification.
Justification might include making districts compact, respecting municipal boundaries,
preserving the cores of prior districts, or avoiding contests between incumbent
representatives.
However, the state must show with some specificity that a particular
objective required the specific deviations in its plan.
The Court has held that, under the Equal Protection Clause of the Fourteenth Amendment
to the U.S. Constitution, both houses of a state legislature must have districts that are
substantially equal in population.
However, the Court has distinguished between
congressional plans and legislative plans, saying that a legislative apportionment plan is
not prima facie invalid because of population inequality as long as its overall range is less
than 10 percent.
Even if the overall range is more than 10 percent, a state may be able
to justify the inequality of population by showing that it was necessary to provide
representation to political subdivisions, as political subdivisions,
or to avoid splitting
political subdivisions.
However, this may be possible only where the number of
representatives is apportioned among a substantially smaller number of political
subdivisions.
Table 4 shows the degree of population equality attained by redistricting plans based on the 1990 census.
Table 4. 1990s District Population Equality
|
Congressional |
State House |
State Senate |
|||
State |
Ideal District Size |
Percent Overall Range |
Ideal District Size |
Percent Overall Range |
Ideal District Size |
Percent Overall Range |
Alaska |
577,227 |
0.00 |
38,482 |
0.10 |
115,445 |
0.09 |
Arkansas |
551,947 |
N/A |
13,751 |
15.50 |
27,502 |
11.70 |
Arizona |
610,871 |
0.00 |
122,174 |
9.85 |
122,174 |
9.85 |
Arkansas |
587,681 |
0.73 |
23,507 |
9.52 |
67,164 |
9.27 |
California |
572,308 |
0.49 |
372,991 |
1.80 |
745,981 |
1.60 |
Colorado |
549,066 |
0.00 |
50,683 |
4.96 |
94,126 |
4.90 |
Connecticut |
547,853 |
0.05 |
21,769 |
8.78 |
91,309 |
7.98 |
Delaware |
668,696 |
N/A |
16,248 |
9.58 |
31,722 |
10.18 |
Florida |
562,519 |
0.00 |
107,816 |
4.99 |
323,448 |
0.86 |
Georgia |
588,929 |
0.93 |
35,990 |
9.95 |
115,682 |
9.95 |
Hawaii |
554,114 |
0.00 |
19,488 |
9.76 |
39,756 |
9.86 |
Idaho |
503,375 |
0.00 |
28,764 |
9.88 |
28,764 |
9.88 |
Illinois |
571,530 |
0.00 |
96,870 |
0.00 |
193,740 |
0.00 |
Indiana |
554,416 |
0.00 |
55,441 |
3.36 |
110,883 |
2.19 |
Iowa |
555,351 |
0.05 |
27,768 |
1.97 |
55,535 |
1.45 |
Kansas |
619,370 |
0.01 |
19,563 |
9.72 |
61,135 |
6.89 |
Kentucky |
614,216 |
3.57 |
36,853 |
9.91 |
96,981 |
9.66 |
Louisiana |
602,853 |
0.04 |
40,190 |
9.97 |
108,204 |
9.78 |
Maine |
613,967 |
0.00 |
8,132 |
43.74 |
35,084 |
4.16 |
Maryland |
597,683 |
0.00 |
33,911 |
10.42 |
101,733 |
9.94 |
Massachusetts |
601,642 |
0.00 |
37,603 |
9.93 |
150,410 |
4.75 |
Michigan |
580,956 |
0.00 |
84,503 |
16.13 |
244,613 |
15.81 |
Minnesota |
546,887 |
0.00 |
32,650 |
5.90 |
65,300 |
3.42 |
Mississippi |
514,643 |
0.02 |
21,042 |
9.97 |
49,485 |
8.96 |
Missouri |
568,563 |
0.20 |
31,393 |
8.96 |
150,502 |
8.42 |
Montana |
799,065 |
N/A |
7,991 |
9.97 |
15,981 |
9.51 |
Nebraska |
526,128 |
0.20 |
N/A-unicameral |
|
32,212 |
3.81 |
Nevada |
600,916 |
0.00 |
28,615 |
4.55 |
57,230 |
2.28 |
New Hampshire |
554,626 |
0.07 |
2,773 |
14.53 |
46,219 |
12.36 |
New Jersey |
594,630 |
0.00 |
193,255 |
4.60 |
193,255 |
4.60 |
New Mexico |
505,023 |
0.16 |
21,640 |
9.89 |
36,073 |
9.58 |
New York |
580,337 |
0.00 |
119,936 |
9.43 |
294,925 |
4.29 |
North Carolina |
552,386 |
0.00 |
55,239 |
9.97 |
132,573 |
9.94 |
North Dakota |
641,364 |
N/A |
13,037 |
8.71 |
13,037 |
8.71 |
Ohio |
570,901 |
0.00 |
109,563 |
13.60 |
109,563 |
13.60 |
Oklahoma |
524,264 |
0.00 |
31,144 |
6.13 |
65,533 |
3.93 |
Oregon |
568,464 |
0.00 |
47,372 |
1.89 |
94,744 |
1.69 |
Pennsylvania |
565,793 |
0.01 |
58,530 |
4.94 |
237,632 |
1.86 |
Rhode Island |
501,732 |
0.02 |
10,035 |
14.66 |
20,059 |
13.00 |
South Carolina |
581,117 |
0.00 |
28,119 |
5.20 |
75,798 |
1.00 |
South Dakota |
699,999 |
N/A |
19,886 |
9.47 |
19,886 |
9.47 |
Tennessee |
544,071 |
0.02 |
49,461 |
9.96 |
148,383 |
13.92 |
Texas |
566,217 |
0.00 |
113,243 |
9.99 |
547,952 |
9.98 |
Utah |
574,283 |
0.02 |
22,971 |
7.94 |
59,409 |
7.60 |
Vermont |
562,758 |
N/A |
3,752 |
17.62 |
18,759 |
16.36 |
Virginia |
562,487 |
0.00 |
61,874 |
9.67 |
154,684 |
8.53 |
Washington |
540,744 |
0.00 |
99,320 |
0.00 |
99,320 |
0.00 |
West Virginia |
597,826 |
0.09 |
17,935 |
9.96 |
105,499 |
9.98 |
Wisconsin |
543,530 |
0.00 |
49,412 |
0.92 |
148,235 |
0.52 |
Wyoming |
455,975 |
N/A |
7,560 |
9.97 |
15,120 |
9.60 |
Source: NCSL, 1996.
Note: Alaska, Delaware, Montana, North Dakota, South Dakota, Vermont and Wyoming received only one seat in the U.S. House of Representatives, so their congressional plans did not have an overall range.
Go to NCSL Redistricting home page
This page is maintained by the Redistricting Task Force for the National Conference of State Legislatures
Update: 10/31/03 (psw)
Comments: peter.wattson@senate.leg.state.mn.us