House Apportionment: Could Census Corrections Shift a House Seat?

CRS Report for Congress
House Apportionment: Could Census
1
Corrections Shift a House Seat?
Royce Crocker
Specialist in American National Government
Government and Finance Division
Summary
On September 30, 2003, the Census Bureau briefed selected congressional staff on
changes that will be made to the 2000 census counts through the bureau’s “Count
Question Resolution Program,” (CQR). This report uses the incomplete figures released
to staff at the September briefings to assess the potential impact on states’ claims to
Representatives if these figures were to be used to apportion seats in the House instead
of the official apportionment figures released by the Census Bureau on December 28,

2000. Although no seats would shift among states based on these numbers, the already-^th

narrow margin between North Carolina and Utah for the last (435) seat in the House
is narrowed by a factor of 10. This report will be updated when the Census Bureau
releases final numbers from the CQR program.
Background
One of the fundamental issues before the framers at the Constitutional Convention
in 1787 was how power was to be allocated in the Congress among the smaller and larger
states. The solution ultimately adopted, known as the Great (or Connecticut)
Compromise, resolved the controversy by creating a bicameral Congress with states
represented equally in the Senate, but in proportion to population in the House. The
Constitution provided the first apportionment of House seats: 65 Representatives were
allocated to the states based on the framers’ estimates of how seats might be apportioned
after a census. House apportionments thereafter were to be based on Article 1, section 2,
as modified by the Fourteenth Amendment:
Amendment XIV, section 2. Representatives shall be apportioned among the
several States ... according to their respective numbers....
Article 1, section 2. The number of Representatives shall not exceed one for
every thirty Thousand, but each State shall have at least one Representative....

¹ This report originally was authored by David Huckabee, who has retired from CRS.
Congressional Research Service ˜ The Library of Congress

The process of apportioning seats in the House is constrained both constitutionally
and statutorily. The Constitution defines both the maximum and the minimum size of the
House. There can be no fewer than one Representative per state, and no more than one
for every 30,000 persons.²
The current apportionment formula was adopted in 1941. In addition to specifying
the apportionment method, the 1941 Act sets the House size at 435 and mandates
administrative procedures for apportionment. The President is required to transmit to
Congress “a statement showing the whole number of persons in each state” and the
resulting seat allocation within one week after the opening of the first regular session of
Congress following the census.³
The Census Bureau has been assigned the responsibility of computing the
apportionment. As matter of practice, the director of the bureau reports the results of the
apportionment by December 31st of the census year. Once the results are received by
Congress, the Clerk of the House is charged with the duty of sending to the governor of
each state a “certificate of the number of Representatives to which such state is entitled”
within 15 days of receiving notice from the President.⁴
There are no provisions for altering the apportionment because of errors that may
later be found in census counts. The Census Bureau’s Federal Register program notice
pertaining to the “Count Question Resolution Program,” states that while “corrections
made to the population and housing unit counts by this program will result in the issuance
of new official Census 2000 counts ... and may be used by governmental entities for future
programs requiring official Census 2000 data,”⁵ the “Census Bureau will not change the
apportionment or redistricting counts to reflect corrections resulting from the CQR
process.”⁶
The CQR Program
The Count Question Resolution Program provides a mechanism for corrections to
the official census counts. Three types of challenges to the census counts are considered
by this program, which could result in: (1) correcting inaccurate boundaries of
jurisdictions; (2) reassigning inaccurately geocoded living quarters and their associated

² The actual language in Article 1, section 2, pertaining to this minimum size reads as follows:
“The number of Representatives shall not exceed one for every thirty Thousand, but each State
shall have at least one Representative.” This clause is sometime misread to be a requirement that
districts can be no larger than 30,000 persons. It should be read as a minimum-size population
requirement, i.e. districts cannot have fewer than 30,000 persons.
³ 55 Stat. 761 (1941) Sec. 22 (a). [Codified in 2 U.S.C. 2(a).] In other words, after the 2000
Census, this report was due in January 2001.
⁴ Ibid., Sec. 22 (b).
⁵ U.S. Dept. of Commerce, Bureau of the Census, “The Census 2000 Count Question Resolution
Program,” Federal Register, vol. 66, no. 77, July 6, 2001, p. 35589.
⁶ Ibid., p. 35593.

population to correct boundaries; and (3) adding or removing specific living quarters (and
the persons residing in them) that were erroneously included in the counts.
The CQR process began on June 30, 2001, and jurisdictions had until September 30,
2003, to file challenges under the program. The Census Bureau released tables to
Congress showing the current tally of changes to the official counts identified as of
September 30, 2003, but not all challenges to census counts had yet been processed. The
most significant changes affect North Carolina, whose 13^th Representative was the last
(435^th) seat assigned in the official reapportionment of House seats following the 2000
census.
As of September 30, 2003, North Carolina would have its official 2000 census count
reduced by 2,673 persons because, according to the Raleigh News & Observer, the Census
Bureau “counted 2,696 students living in UNC-CH [University of North Carolina, Chapel
Hill] dormitories — twice.”⁷ Students in dormitories are supposed to be enumerated by
the “group quarters” procedure where forms are sent to the institution instead of
individuals in the group quarters. In the case of UNC-Chapel Hill, forms were sent to
some students as well as to the university. The discrepancy between the 2,673 and 2,696
figures is accounted for by other CQR activities in North Carolina. As of this writing, the
Census Bureau has not completed evaluating all challenges to the North Carolina figures,
so these figures may change; but the number of persons North Carolina would need to
lose for its 13^th district to drop to the 436^th position in the apportionment rankings would
be reduced from 3,084 to 309 persons if these new numbers were used to reapportion the
House (if Congress were to choose to do so).
On a nationwide basis, 7,183 persons are either added to, or subtracted from, state
populations, resulting in a net change of 1,427 persons from the September 30, 2003,
CQR program data. No seats would shift among states based on the revised
apportionment populations derived by these figures (see Table 1).
Table 1. Revisions of 2000 Census Apportionment Population
Based on Census Bureau’s September 30, 2003,
Count Question Resolution Program Numbers
^{2000 C}^eⁿ^s^u^s^Co^unt^quest^ioⁿ^Revⁱ^sed
^St^a^t^e^a^ppo^rt^io^nm^ent^a^res^olution^b^a^ppo^rt^io^nm^ent^c^Sea^t^s
^po^pula^t^ioⁿ^ch^an^ges^po^pula^t^ioⁿ
^A^L^4,461,130^4,461,130⁷
^A^K^628,933^-¹^628,932¹
^A^Z^5,140,683^5,140,683⁸
^A^R^2,679,733^2,679,733⁴
^C^A^33,930,798⁵^33,930,803⁵³
^C^O^4,311,882⁷³⁶^4,312,618⁷
^C^T^3,409,535¹⁹^3,409,554⁵
^D^E^785,068^785,068¹
^F^L^16,028,890⁴⁴²^16,029,332²⁵
^G^A^8,206,975⁶¹^8,207,036¹³
^H^I^1,216,642^1,216,642²

⁷ Rob Christensen, “Miller Vows to Stay in Seat Utah Covets,” Raleigh News & Observer, Oct.
2, 2003, [http://www.newsobserver.com/politics/miller/story/2915960p-2680128c.html], visited
Oct. 6, 2003.

^{2000 C}^eⁿ^s^u^s^Co^unt^quest^ioⁿ^Revⁱ^sed
^St^a^t^e^a^ppo^rt^io^nm^ent^a^res^olution^b^a^ppo^rt^io^nm^ent^c^Sea^t^s
^po^pula^t^ioⁿ^ch^an^ges^po^pula^t^ioⁿ
^ID^1,297,274³¹^,297,277²
^IL^12,439,042²⁷⁴^12,439,316¹⁹
^IN^6,090,782²¹^6,090,803⁹
^IA^2,931,923⁵⁸^2,931,981⁵
^K^S^2,693,824³⁹⁶^2,694,220⁴
^K^Y^4,049,431⁴⁴⁰^4,049,871⁶
^L^A^4,480,271^-¹⁸⁴^,480,253⁷
^ME^1,277,731^1,277,731²
^MD^5,307,886^5,307,886⁸
^MA^6,355,568^6,355,568¹⁰
^MI^9,955,829³⁶^9,955,865¹⁵
^MN^4,925,670¹³^4,925,683⁸
^MS^2,852,927^-²^2,852,925⁴
^MO^5,606,260¹⁴⁷²^5,607,732⁹
^MT^905,316^905,316¹
^N^E^1,715,369²¹^,715,371³
^N^V^2,002,032^2,002,032³
^N^H^1,238,415^1,238,415²
^N^J^8,424,354^-³^8,424,351¹³
^N^M^1,823,821^1,823,821³
^N^Y^19,004,973²⁰⁷^19,005,180²⁹
^N^C^8,067,673^-²⁶⁷³^8,065,000¹³
^N^D^643,756^643,756¹
^O^H^11,374,540³^11,374,543¹⁸
^O^K^3,458,819^-²^3,458,817⁵
^O^R^3,428,543³³^3,428,576⁵
^PA^12,300,670^12,300,670¹⁹
^R^I^1,049,662^1,049,662²
^S^C^4,025,061^-¹¹²^4,024,949⁶
^S^D^756,874^756,874¹
^T^N^5,700,037^-¹⁶⁵^,700,021⁹
^T^X^20,903,994^-³⁰^20,903,964³²
^U^T^2,236,714²⁹^2,236,743³
^V^T^609,890^609,890¹
^V^A^7,100,702^-²¹⁷^,100,681¹¹
^WA^5,908,684²⁰^5,908,704⁹
^WV^1,813,077⁶¹^,813,083³
^WI^5,371,210²⁹^5,371,239⁸
^WY^495,304^495,304¹
^281,424,177^1,427^281,425,604⁴³⁵
^a^{U.S. Dept. of}^C^o^m^m^e^{rce, B}^u^reau^of^t^h^e^C^eⁿ^s^u^s^,^C^e^ns^us^{2000 Show}^s^Resⁱ^dent^Popul^atⁱ^{on of}
^{281,421,906; Appor}^tⁱ^onm^ent^C^ount^s^D^e^lⁱ^ver^{ed t}^o^Pr^esⁱ^d^ent^,^P^r^es^s^R^e^leas^{e C}^B^00-^C^N^.64^(W^as^hⁱⁿ^g^t^on^:
^{Dec. 28,}^2000),^table^1.^(P^leas^{e n}^o^{te th}^{at th}^{e apportion}^m^en^{t popu}^lationⁱⁿ^clu^d^es^th^{e f}^o^reigⁿ^-^bas^e^{d m}^ilitary
^an^{d ot}^h^e^{r f}^e^deral^em^pl^oy^ees^lⁱ^vⁱⁿ^g^ov^ers^eas^.)^b
^U^.^S^.^D^e^pt^{. of}^Co^m^m^e^r^c^e^,^Bu^reau^of^t^h^{e C}^eⁿ^s^u^s^,^Popul^atⁱ^ons^of^St^at^es^Incl^udi^{ng C}^ount^Q^u^es^tⁱ^oⁿ
^Res^o^l^u^tⁱ^{on (}^C^Q^R⁾^C^o^r^r^ectⁱ^ons^T^h^r^{ough Sept}^em^ber^{30, 2003}^{, table prov}^{ided at a brief}ⁱⁿ^g^{to con}^g^r^es^sⁱ^on^al
^s^t^af^f^on^S^e^pt^{. 30, 2003.}^c
^Fig^u^r^es^calcu^{lated by}^C^R^S^.
As noted above, the already narrow margin between the claims of North Carolina
and Utah for the 435^th seat in the House would become even narrower if the CQR
program numbers were to be substituted for the official apportionment numbers. Utah’s
margin to gain a seat drops from 855 to 87 persons (see Table 2). The margin between

the 435^th and 436^th seat was similarly close in the reapportionment following the 1970
Census. If Oklahoma had had 283 fewer persons, or Oregon had had 229 more persons,
Oklahoma would have lost a seat to Oregon. Unlike the present case, the margin would
have increased if corrected totals had been used for these states to 1,314 persons for
Oklahoma and 1,071 more persons for Oregon.
Table 2. Population Needed to Gain or Lose a Seat Using Census
Bureau September 30, 2003, Count Question Resolution Program
Numbers to Adjust the 2000 Census Apportionment Population
^Revⁱ^sed^Priority^Pop^.ⁿ^eed^ed^{to gai}ⁿ^{or l}^o^s^e^s^e^at^Using^U^sⁱⁿ^{g offi}^ci^al
^Ra^nk^St^a^t^e^Sea^t^a^ppo^rt^io^nm^en^val^u^e^a^revis^ed^a^ppo^rt^io^nm^ent
^t^po^pula^t^ioⁿ^po^pula^t^ioⁿ^b^po^pula^t^ioⁿ^c
⁴²⁰^C^A⁵¹^33,930,803^671,930.02^-¹^,324,952^-¹^,325,368
⁴²¹^T^N⁹⁵^,700,021^671,753.92^-^221,143^-^221,227
⁴²²^MA¹⁰^6,355,568^669,935.69^-^229,996^-^230,072
⁴²³^N^Y²⁹^19,005,180^666,951.07^-^605,789^-^605,818
⁴²⁴^C^O⁷⁴^,312,618^665,451.40^-^128,056^-^127,372
⁴²⁵^PA¹⁹^12,300,670^665,144.06^-^359,731^-^359,885
⁴²⁶^T^X³²^20,903,964^663,701.52^-^567,227^-^567,519
⁴²⁷^MO⁹⁵^,607,732^660,877.55^-^128,854^-^127,450
⁴²⁸^C^A⁵²^33,930,803^658,881.60^-^679,229^-^679,651
⁴²⁹^MN⁸⁴^,925,683^658,222.08^-^93,767^-^93,814
⁴³⁰^G^A¹³^8,207,036^657,088.76^-^142,345^-^142,386
⁴³¹^IA⁵²^,931,981^655,610.88^-^44,359^-^44,337
⁴³²^F^L²⁵^16,029,332^654,394.74^-^213,172^-^212,933
⁴³³^O^H¹⁸^11,374,543^650,239.34^-^79,546^-^79,688
⁴³⁴^C^A⁵³^33,930,803^646,330.32^-^33,507^-^33,940
⁴³⁵^N^C¹³^8,065,000^645,716.78^-³⁰⁹^-³^,084
^L^a^s^t^s^e^at^as^sⁱ^{gned by l}^a^w
⁴³⁶^U^T⁴²^,236,743^645,692.09⁸⁷⁸⁵⁵
⁴³⁷^N^Y³⁰^19,005,180^644,335.95^40,730^47,245
⁴³⁸^T^X³³^20,903,964^643,275.02^79,349^86,268
⁴³⁹^MI¹⁶^9,955,865^642,648.32^47,537^50,891
⁴⁴⁰^IN¹⁰^6,090,803^642,027.01^35,005^37,057
⁴⁴¹^MT²^905,316^640,155.08^7,866^8,168
⁴⁴²^IL²⁰^12,439,316^638,123.45^148,022^152,465
⁴⁴³^MS⁵²^,852,925^637,933.42^34,809^35,763
⁴⁴⁴^C^A⁵⁴^33,930,803^634,248.32^613,537^624,984
⁴⁴⁵^WI⁹⁵^,371,239^633,006.59^107,851^109,696
⁴⁴⁶^O^K⁶³^,458,817^631,490.70^77,920^79,090
⁴⁴⁷^PA²⁰^12,300,670^631,011.06^286,668^290,837
⁴⁴⁸^F^L²⁶^16,029,332^628,722.13^433,281^439,176
⁴⁴⁹^O^R⁶³^,428,576^625,969.47^108,161^109,365
⁴⁵⁰^MD⁹⁵^,307,886^625,540.36^171,204^173,020^a
^E^ach^st^at^e’s^cl^ai^m^t^o^rep^r^esen^t^a^tⁱ^oⁿⁱⁿ^t^h^e^Ho^u^s^eⁱ^s^b^a^sed^oⁿ^{a “p}^ri^o^rⁱ^t^y^val^u^{e” d}^e^t^e^rmⁱⁿ^ed^b^y^t^h^{e f}^o^l^l^o^wⁱⁿ^g^½^th
^f^o^rm^u^l^{a: P}^V^{= P}^/^[ⁿ⁽ⁿ^{- 1}^)]^{; w}^h^e^r^e^P^V⁼^the^s^t^a^t^e^’^s^pr^ior^ity^v^a^lue^,^P⁼^the^s^t^a^t^e^’^s^popula^{tion, a}^{nd n =}^the^s^t^a^t^e^’^sⁿ^th
^seatⁱⁿ^t^h^{e Ho}^u^s^e.^F^o^{r exam}^p^l^e,^t^h^{e p}^rⁱ^o^ri^t^y^val^u^{e o}^f^No^rt^h^Caro^lⁱⁿ^{a’s 1}³^se^a^t^is:^½
^PV^NC¹³⁼⁸^{,065,000 / [}¹³⁽^{13 -}^{1 )}^]^½
⁼⁸^{,065,000/ [}^{156 ]}
⁼^8,065,000^{/12.489959967968}
⁼^{645,716.780219014}
^T^h^e^a^c^t^ua^{l s}^e^a^t^a^s^sⁱ^gⁿ^m^e^nts^a^r^e^m^a^de^by^ra^nk^ing^a^{ll of}^the^s^t^a^t^e^s^{’ priority}^v^a^lue^s^f^r^om^hig^h^e^s^{t to low}^e^s^t^{until 435 s}^e^a^t^s
^{are al}^l^o^cat^ed^.^b^th
^T^h^e^s^e^fⁱ^g^u^r^e^s^r^e^pr^e^s^eⁿ^t^the^popula^{tion a}^s^t^a^t^e^w^{ould ne}^e^d^e^ithe^r^{to los}^e^{in or}^de^r^{to dr}^{op be}^low^the⁴³⁵^seat
^c^u^tof^f^,^or^{to g}^a^{in to r}ⁱ^s^e^a^bov^e^the^c^u^tof^f^.^I^f^{, f}^o^r^e^x^a^m^ple^,^N^o^r^t^{h C}^a^r^o^lina^ha^{d 309 f}^e^w^e^r^pe^r^s^ons^{in its}^popula^{tion tota}^l,
^the^s^t^a^t^e^’^s^pr^ior^ity^v^a^lue^w^ould^de^c^r^e^a^s^e^{to 645,691.04, w}^h^ic^{h w}^{ould r}^e^s^u^{lt in a}^ne^w^s^e^que^nc^e^num^be^r^of⁴³⁶^b^ecau^se

^Utah^{’s 4}^th^s^e^a^t^w^{ould oc}^c^upy^the⁴³⁵^th^{position in the}^priority^{list. Sim}^ila^rl^y^,^if^U^t^a^h^ha^{d 87 m}^o^r^e^pe^r^s^ons^{in its}
^popula^{tion tota}^{l, it w}^{ould dis}^p^l^{ace No}^rt^h^Caro^lⁱⁿ^a,^b^ecau^{se Ut}^ah^{’s p}^rⁱ^o^ri^t^y^val^u^{e w}^o^u^l^d^b^eco^m^e^645.716.91.^c
^T^h^e^s^e^fⁱ^g^u^r^e^s^s^how^how^m^a^ny^m^o^r^e^or^f^e^w^e^r^pe^r^s^ons^w^{ould ha}^v^e^be^eⁿ^ne^c^e^s^s^a^r^y^f^o^r^s^t^a^t^e^s^to^g^aⁱⁿ^or^los^e^a
^s^e^a^t^ba^s^e^{d on the}^of^fⁱ^cⁱ^a^l^a^ppor^tionm^eⁿ^{t f}ⁱ^g^u^r^e^s^s^e^{nt to the}^P^r^e^s^ideⁿ^{t a}^{nd r}^e^le^a^s^e^d^{on D}^ecem^b^e^{r 2}⁸^,^2000.
^Sour^c^e^:^C^o^m^puta^tions^of^pr^ior^ity^v^a^lue^s^a^{nd pe}^r^s^ons^ne^e^d^e^d^f^o^r^s^t^a^t^e^s^{to g}^a^{in or}^los^e^a^s^e^a^t^by^C^R^S^{. Se}^e^C^R^S
^R^e^p^o^r^t^R^L³^0711,^T^h^e^H^ous^e^Appor^tionm^eⁿ^{t For}^m^{ula in The}^o^r^y^{and Pr}^ac^tic^e^,^b^y^Davi^d^C.^Hu^ckab^ee,^f^o^{r an}
^exp^l^an^atⁱ^oⁿ^o^f^f^o^rm^u^l^{a f}^o^{r al}^l^o^catⁱⁿ^{g Ho}^u^s^{e seat}^s.
Could the CQR Changes Shift a House Seat?
As noted previously, the preliminary CQR corrections released on September 30,
2003, would not alter the apportionment of House seats when compared to the official
results, and the Census Bureau does not plan to alter the official apportionment figures.
If further corrections would change the apportionment, there are 19^th-century precedents
for the Congress to change apportionments after the initial results have been released.
In 1852, Congress increased California’s delegation by one Representative because
it appear[ed] “that the returns of the population of California are incomplete.”⁸ Section
2 of the act provided that if “at any future decennial enumeration of the inhabitants of the
United States, the census of any district or subdivision in the United States shall have
been improperly taken, or if the returns of any district or subdivision shall be accidentally
lost or destroyed, the Secretary of the Interior shall have power to order a new⁹
enumeration of such district or subdivision.”
Additional Representatives were added to the totals initially provided by the
reapportionment acts following the 1860 and 1870 censuses. Eight states received one¹⁰
additional seat by a supplemental apportionment act approved in March 1862. Another
supplemental apportionment act enacted in May 1872, assigned one additional¹¹
Representative each to nine states.
When Congress accepted challenges to census results, the 19^th-century precedent was
to increase the size of the House, not to take seats from states. There are no examples^th^th
similar to the 19-century precedents noted above in the 20 century, but the last time
new states were admitted between censuses, the House size was temporarily increased to¹²

437 until the next reapportionment, when it became 435 again.

If further refinement of the 2000 Census data reveals that the apportionment would
be changed if the corrected numbers were to be used instead of those used in December

2000, Congress could alter the apportionment by legislation.

⁸ 10 Stat. 25.
⁹ Ibid.
¹⁰ 12 Stat. 353.
¹¹ 17 Stat. 192.
¹² By authority of section 9 of P.L. 85-508 (72 Stat. 345) and section 8 of P.L. 86-3 (73 Stat. 8),
which admitted Alaska and Hawaii to statehood, the House size was temporarily increased to 437
until the reapportionment following the 1960 census, when it returned to 435.