Could Maryland justify unequal population sizes among Baltimore City's state legislative districts because of a census undercount or a desire to keep the city's boundaries intact?

Plain-English summary

The 1990 census showed Baltimore City's population had fallen to 736,014, down from 786,775 in 1980, meaning the city's delegation in the General Assembly would necessarily shrink. If the city kept eight senatorial districts entirely within its boundaries, those districts would run about 9.6% below the statewide "ideal" district population, and once combined with normal variance elsewhere in the state, the overall statewide maximum deviation would likely exceed the usual 10% threshold courts use to presume a population-equality violation. A state senator asked whether Maryland could justify that kind of deviation either by pointing to the Census Bureau's own finding of a 4.9% undercount in Baltimore City, or by relying on a policy of preserving the city's boundaries as a single legislative unit.

The opinion concluded that the census undercount alone would not justify deviations above 10%. It read the U.S. Supreme Court's decision in Karcher v. Daggett as a real obstacle to relying on undercount data, since the Court there held that population deviations could not be excused by "flaws in census data" absent a level of "precision" the state could not achieve given how uncertain and unevenly distributed undercount estimates are. On the second question, the opinion reached a more favorable but still qualified answer: preserving the integrity of political subdivision lines, including Baltimore City's identity as a discrete jurisdiction, was a recognized justification for state legislative population deviations, and the Maryland Court of Appeals had itself described the city's boundaries as deserving of "a long practice of preserving the City's integrity as a discrete and insular jurisdiction." The opinion found this reasoning could plausibly support keeping eight senatorial districts within Baltimore City even though that produced a deviation above 10%, but flagged that any such plan would need a well-documented legislative record, would need to keep population variances in the rest of the state as low as possible, and would face real doubt since the Supreme Court had never blessed a deviation nearly as large as some hypothetical alternatives (like a 20% statewide maximum) regardless of the justification offered.

Currency note

This opinion was issued in 1991, addressing the specific facts of the 1990 census cycle and the 1992 legislative redistricting plan then being prepared. Subsequent statutory amendments, court decisions, or later AG opinions may have changed the analysis, particularly given later U.S. Supreme Court and Maryland Court of Appeals redistricting decisions and subsequent decennial census cycles. Treat this page as historical context, not current legal advice. Verify the current legal standard for population deviations in state legislative redistricting, and the current text of Article III, §4 of the Maryland Constitution, before relying on any specific rule described here.

Common questions

Could Baltimore City's census undercount justify giving it extra legislative districts beyond what its official population supported?
According to this 1991 opinion, no, not on its own. The opinion read Karcher v. Daggett as requiring a level of precision about undercount data that Maryland could not show, since undercount estimates were themselves imprecise and varied unpredictably from place to place.

Could Maryland instead justify a bigger population deviation just to keep Baltimore City's legislative districts entirely within its own boundaries?
Possibly, according to the opinion, though it called the question "not free from doubt." It found that preserving a political subdivision's boundaries, including Baltimore City's identity as a historically distinct jurisdiction, was a recognized justification courts had accepted for state legislative deviations, but said any resulting plan needed a well-documented legislative record and needed to keep deviations elsewhere in the state as low as possible.

Was there an upper limit on how large a population deviation could be, no matter the justification?
The opinion suggested yes. It noted the Supreme Court had never blessed a deviation as large as 20% statewide, and cited a case where the Court found a 16.4% maximum deviation "may well approach tolerable limits," suggesting that even a strong justification like preserving city boundaries had practical limits.

Background and statutory framework

The 1990 census recorded Baltimore City's population at 736,014, down from 786,775 in 1980, a decline that would necessarily shrink the city's General Assembly delegation from its then-existing nine senatorial districts to roughly 7.23 senators' worth of population if all districts were made exactly equal. If the city retained eight senatorial districts entirely within its boundaries, those districts would run about 9.565% below the statewide ideal district population, a shortfall that, combined with ordinary variance elsewhere in the state, would likely push the statewide maximum deviation above 10%. A national Post-Enumeration Survey released in June 1991 found a 4.9% (36,000-person) census undercount in Baltimore City, but the Commerce Department had declined to adjust the official census on that basis, a decision the office had separately analyzed in its August 12, 1991 opinion on adjusted census data, 76 Opinions of the Attorney General 206 (1991).

The opinion began from the Fourteenth Amendment's Equal Protection Clause requirement that both houses of the General Assembly be apportioned on a population basis, Reynolds v. Sims, 377 U.S. 533, 568 (1964), noting the office had always viewed Article III, §4 of the Maryland Constitution's population equality standard as no more stringent than the federal requirement, an issue the Court of Appeals had reserved, In re Legislative Districting, 299 Md. 658, 683 n. 17, 475 A.2d 428 (1982). Unlike the "precise mathematical equality" demanded for congressional districts, state legislative districts with a maximum deviation under roughly 10% do not by themselves establish a prima facie equal-protection violation, Gaffney v. Cummings, 412 U.S. 735, 745 (1973), and the Supreme Court has never fixed an absolute ceiling on deviation, though it suggested in Mahan v. Howell, 410 U.S. 315, 329 (1973), that a 16.4% maximum deviation "may well approach tolerable limits." The Court has recognized that legitimate state interests, such as maintaining the integrity of political subdivisions, providing compact and contiguous districts, Reynolds v. Sims, 377 U.S. at 578, preserving the cores of prior districts, or avoiding contests between incumbents, Karcher v. Daggett, 462 U.S. 725, 740 (1983), can justify deviations above the minor-variance threshold, but the required showing is "flexible" and scales with the size of the deviation, the importance of the state's interest, and the consistency with which the plan as a whole reflects that interest, Karcher, 462 U.S. at 741.

On the undercount question, the opinion treated Karcher v. Daggett as the leading and largely unfavorable precedent: there, the Supreme Court rejected New Jersey's argument that a census undercount rendered small district variances the functional equivalent of equality, reasoning that the undercount rate "is not known precisely," that statistical imprecision does not make small deviations meaningless, and that census data remains "the only reliable, albeit less than perfect, indication" of relative district population, so any attempt to explain deviations by census flaws "must be supported with a precision not achieved" in that case, 462 U.S. at 738. Given the acknowledged unreliability of the undercount data for Maryland and Baltimore specifically, the opinion concluded it would be inadvisable to justify deviations above 10% on the basis of the undercount, in Baltimore City or statewide.

On the boundary-preservation question, the opinion found more room. While a census undercount had not been recognized as justifying small population variances, preserving the integrity of political subdivision lines had been accepted as a legitimate state interest, see, e.g., Brown v. Thompson, 462 U.S. 835 (1983) (upholding an 89% deviation tied to Wyoming's decision to grant a representative to its least populous county, on unusual facts). The Court of Appeals had itself described maintenance of Baltimore City's boundaries as "a continuation of a long practice of preserving the City's integrity as a discrete and insular jurisdiction," In re Legislative Districting, 299 Md. 658, 691 n. 22, 475 A.2d 428 (1982), language the opinion read, though dicta, as suggestive of a sufficient justification for some deviation above 10% if the city retained eight senatorial districts. The opinion flagged three cautions: the Supreme Court's emphasis on consistent application of a state's stated policy, though it found the General Assembly's history of recognizing Baltimore City's uniqueness supported consistency here; the possibility that a seven-district alternative could equally serve boundary-preservation interests, which the opinion said would need to be addressed and documented in the legislative record if eight districts were retained instead; and the sheer size of any resulting deviation, since a very large deviation, such as 20% statewide, would probably not be defensible under any justification.

Citations and references

Statutes:

• Article III, §4 of the Maryland Constitution, the state population equality standard for legislative redistricting

Cases:

• Reynolds v. Sims, 377 U.S. 533, 568, 578 (1964), U.S. Supreme Court decision requiring population-based apportionment and recognizing subdivision-integrity and compactness as legitimate deviation justifications
• Gaffney v. Cummings, 412 U.S. 735, 745 (1973), U.S. Supreme Court decision on the under-10%-deviation threshold for a prima facie equal-protection violation
• Mahan v. Howell, 410 U.S. 315, 329 (1973), U.S. Supreme Court decision suggesting a 16.4% maximum deviation approaches tolerable limits
• Karcher v. Daggett, 462 U.S. 725, 738, 740-41 (1983), U.S. Supreme Court decision rejecting reliance on census undercount data absent precise proof, and describing the flexible showing needed to justify deviations
• Brown v. Thompson, 462 U.S. 835, 849-50 (1983), U.S. Supreme Court decision upholding an 89% deviation on unusual facts tied to preserving a county's own representation
• In re Legislative Districting, 299 Md. 658, 683 n. 17, 691 n. 22, 475 A.2d 428 (1982), Maryland Court of Appeals decision describing the long practice of preserving Baltimore City's integrity as a distinct jurisdiction

Source

• Landing page: https://oag.maryland.gov/resources-info/Pages/attorney-general%E2%80%99s-opinions.aspx
• Original PDF: https://oag.maryland.gov/resources-info/Documents/pdfs/Opinions/1991/Volume76_1991.pdf (this opinion appears at printed pages 214-219 of the bound annual volume; Maryland's site does not publish a standalone PDF of this opinion)

Original opinion text

Best-effort transcription from a scanned PDF. Minor errors may remain — the linked PDF is authoritative.

Elections - Reapportionment and Redistricting - General
Assembly - Justification For Population Deviations in
Baltimore City Districts

                         August 28, 1991

The Honorable John A. Pica, Jr.
Maryland Senate

 You have requested our opinion whether the State may justify a

maximum population deviation in excess of 10% in State legislative
districts on the basis of (1) the census undercount in Baltimore City or
(2) a policy decision to maintain the integrity of the City's boundaries.

  For the reasons discussed below, we conclude as follows:

  1. The census undercount in Baltimore City would not alone

justify deviations from population equality in excess of 10%. In
addition, any attempt to explain population deviations on the basis of
statewide flaws in census data must be supported with a precision that
might be found lacking in existing undercount data.

 2. Although the issue is not free from doubt, the policy of

preserving the integrity of the City's boundaries could be found to
justify population deviations somewhat in excess of 10%.

                                   I

                            Background

 The 1990 U.S. Census shows that Baltimore City's population has

declined to 736,014, down from the 1980 count of 786,775. This drop
in population will necessarily result in a reduction in the City's
delegation in the General Assembly.1 If the City were to retain eight
senatorial districts within municipal boundaries, and their population
were distributed evenly, the districts would have an average population
of 92,002, or a deviation of minus 9.565% from the "ideal" district of
101,733 persons.2 If such a plan were adopted, it is highly unlikely that
population variances in the remainder of the State could be kept
sufficiently low so as to prevent the maximum percentage deviation
statewide from exceeding 10%.

    1
       Baltimore City presently has nine senatorial districts. Based on 1990

census figures, the City would have 7.23 senators within its boundaries if all
legislative districts were absolutely equal in population.

 A national Post-Enumeration Survey conducted by the U.S. Census

Bureau and released in June 1991 disclosed that Baltimore City had a
census undercount of 4.9%, or 36,000 people.3 However, the
Commerce Department has refused to adjust the census on the basis of
the Post-Enumeration Survey, contending that in many cases, including
those of Maryland and Baltimore City, adjusted figures were less
accurate than decennial census data. See generally 76 Opinions of the
Attorney General 206 (1991).

                                    II

                                Analysis

A. General Principles

The Equal Protection Clause of the Fourteenth Amendment to the

U.S. Constitution requires that the seats in both houses of the Maryland
General Assembly be apportioned on a population basis. Reynolds v.
Sims, 377 U.S. 533, 568 (1964).4 However, unlike a requirement of
"precise mathematical equality" for congressional districts, minor
deviations from population equality among state legislative districts -
generally speaking, a maximum deviation of less than 10% - are
insufficient to set forth a prima facie case of violation of the one-
person/one-vote requirement of the Fourteenth Amendment. Gaffney v.
Cummings, 412 U.S. 735, 745 (1973). In short, the State is not

    2
        If the City were to retain all of its existing nine senatorial districts

within its boundaries, the deviation from the ideal would be minus 19.6% - a
malapportionment that very probably would be constitutionally indefensible.
3
If these figures were added to the City's decennial census count,
"ideally" Baltimore would be entitled to 7.53 senators. If the undercount
statewide is factored into the equation, this figure would be lower.
4
The Attorney General's Office has always viewed the population
equality standard of Article III, §4 of the Maryland Constitution as no more
stringent than the requirements of the Fourteenth Amendment. See Letter from
Assistant Attorney General Robert A. Zarnoch to the Honorable Benjamin L.
Brown (June 4, 1991). The Court of Appeals has reserved the issue. See In re
Legislative Districting, 299 Md. 658, 683 n. 17, 475 A.2d 428 (1982).

required to justify such minor variances. However, deviations in excess
of 10% must be justified by consistent adherence to a legitimate State
policy.

The Supreme Court has never specified the absolute upper limit for

maximum population deviation in a state plan. However, in Mahan v.
Howell, 410 U.S. 315 (1973), the Court suggested that a 16.4%
maximum percentage deviation "may well approach tolerable limits."
410 U.S. at 329.5

 In a series of opinions, the Supreme Court has identified the state

interests that might be deemed sufficient to justify more than minor
deviations from the Fourteenth Amendment's one-person/one-vote
requirement. For example, in Reynolds v. Sims, the Court recognized
that deviation from population equality may be necessary to "maintain
the integrity of various political subdivisions" and to "provide for
compact districts of contiguous territory." 377 U.S. at 578. And in
Karcher v. Daggett, a congressional redistricting case, the justices
indicated that "preserving the cores of prior districts" and "avoiding
contests between incumbent[s]" "might justify some variance." 462 U.S.
725, 740 (1983).6

 However, the mere articulation of these interests or their singular

or arbitrary application is insufficient to justify sizeable population
variances. In Karcher v. Daggett, the Court noted that "[t]he 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." 462 U.S.
at 741.

    5
        In Brown v. Thompson, 462 U.S. 835 (1983), a majority of justices

rejected a constitutional attack on a Wyoming apportionment plan with a
maximum deviation of 89% from population equality, caused by the State's
decision to grant a representative to its least populous county. However, Brown
involved a "strange set of facts," Tribe, American Constitutional Law 1072 n. 10
(2d ed. 1988), viz. the fact that the state apportionment plan was not challenged
as a whole. The concurrences in this narrow decision indicated that an interest
in preserving county boundaries would not justify an 89% maximum deviation.
462 U.S. at 849-50.
6
We do not mean to suggest that these considerations exhaust the list of
permissible justifications for variances. Others are discussed in the letter of
advice cited in note 4 above.

B. Undercount as Justification

  Karcher v. Daggett is the leading Supreme Court case on a state's

attempt to use census undercounts or inaccuracies as a justification for
population variances. In that case, New Jersey argued that, because of
the census undercount, the variances at issue should have been
considered the functional equivalent of zero. 462 U.S. at 735. The
Court responded that "[t]here are two problems with this approach.
First, appellants concentrate on the extent to which the census
systematically undercounts actual population - a figure which is not
known precisely and which, even if it were known, would not be
relevant to this case. Second, the mere existence of statistical
imprecision does not make small deviations among districts the
functional equivalent of equality." Id. The Justices went on to state
that:

          The census may systematically undercount
     population, and the rate of undercounting may vary
     from place to place. Those facts, however, do not
     render meaningless the differences in population
     between congressional districts, as determined by
     uncorrected census counts. To the contrary, the
     census data provide the only reliable - albeit less than
     perfect - indication of the districts' "real" relative
     population levels. Even if one cannot say with
     certainty that one district is larger than another merely
     because it has a higher census count, one can say with
     certainty that the district with a larger census count is
     more likely to be larger than the other district than it
     is to be smaller or the same size. That certainty is
     sufficient for decisionmaking.... Furthermore,
     because the census count represents the "best
     population data available," ... it is the only basis for
     good-faith attempts to achieve population equality.
     Attempts to explain population deviations on the basis
     of flaws in census data must be supported with a
     precision not achieved here.

462 U.S. at 738 (emphasis in original).

 Karcher v. Daggett stands as an obstacle to ready reliance on

undercount data, even though it may not shut the door entirely on the
use of the census undercount to justify population variances. At the very
least, it requires "precision" in any state attack on census data. And, as
our August 12, 1991 opinion on the use of adjusted census data for
legislative redistricting emphasizes, it may not be possible to show that
the undercount data is superior to decennial census figures. In light of
the uncertain reliability of census undercount data and the holding in
Karcher v. Daggett, we think that it would be inadvisable to justify
variances in excess of 10% on the basis of the census undercount, either
in Baltimore City or statewide.

C. Integrity of Subdivision Lines

  Although a census undercount has yet to be recognized as a

justification for small variances from one-person/one-vote requirements,
maintaining the integrity of political subdivision lines has been deemed
a permissible state policy that would support such deviations. See, e.g.,
Brown v. Thompson, 462 U.S. 835 (1983).

 Significantly, in the 1982 State reapportionment litigation, the Court

of Appeals wrote that the "maintenance of the City's boundaries
represents a continuation of a long practice of preserving the City's
integrity as a discrete and insular jurisdiction." In re Legislative
Districting, 299 Md. 658, 691 n. 22, 475 A.2d 428 (1982).7 To be
sure, the Court's remarks were made in the context of rejecting a
"compactness" challenge to certain City districts, and the judges went on
to observe that the goal of maintaining City lines was achieved without
sacrifice of substantial equality of population. 299 Md. at 691-91 n. 22.
Nevertheless, this acknowledgement by the Court of Appeals of the
General Assembly's recognition of Baltimore City's uniqueness for
redistricting purposes, although dicta, is suggestive of a justification
sufficient to support some population variances in excess of 10%, which
would necessarily result from the retention of eight senatorial districts
within the City.

  Three factors give us pause, however. First, the Supreme Court has

emphasized consistent application of state policies asserted in
justification of nonexcessive population variances. And, with respect to
the 1992 legislative redistricting plan, it is virtually impossible to
maintain the boundary lines of all of the City's counterparts, the
counties. Nonetheless, the General Assembly has consistently
recognized the City's uniqueness and the value in preserving its integrity
as a singular jurisdiction. Thus, we believe a state interest in
maintaining City lines would not fail on consistency grounds.

    7
      The Court was also sympathetic to the State's "rational goal of avoiding

additional loss of senior legislators" caused by the decline in Baltimore City
population reported in the 1980 census. 299 Md. at 691.

  Second, it might be argued that an interest in preserving City lines

could also be served by including just seven senatorial districts within
municipal boundaries. However, it may be that other legitimate State
interests - for example, preserving the cores of prior districts and
avoiding contests between incumbents - would be ill-served by limiting
the City to seven senatorial districts within its boundaries. If the
Governor and the General Assembly were to maintain eight senatorial
districts in Baltimore City, the problems associated with a seven-district
alternative should be thoroughly documented in the legislative record.

 Finally, a vital factor in justifying population variances in excess of

10% is the degree of variance at issue. For example, a 20% maximum
percentage deviation statewide would probably not be defensible,
regardless of the circumstances. See Mahan v. Howell, 410 U.S. at 329
(16.4% maximum percentage deviation "may well approach tolerable
limits"). Thus, population variances around the remainder of the State
would have to be kept as low as possible to increase the chances that a
court would accept the State's interest in preserving City lines.

                                III

                           Conclusion

 In summary, it is our opinion that, although reliance on the census

undercount may not justify deviations from population equality in excess
of 10%, minor deviations higher than that threshold might be justified
by a State interest in preserving the City's integrity as a discrete and
insular jurisdiction.

                                            J. Joseph Curran, Jr.
                                            Attorney General

                                            Robert A. Zarnoch
                                            Assistant Attorney General

Jack Schwartz
Chief Counsel
Opinions and Advice