OAG 92-58

April 9, 1992

Thomas M. Smith

Boehl, Stopher, Graves and Deindoerfer

417 South Lake Drive

P.O. Box 606

Prestonsburg, Kentucky 41653

Dear Mr. Smith:

You have asked how a county reapportionment commission should proceed when it discovers a discrepancy between the population of a precinct as designated in the federal census and the actual population of the precinct as shown by other evidence. Specifically your question is:

We need to know, as a committee appointed by the Fiscal Court to reapportion the districts, whether we can take into account the number of registered voters in the various voting precincts within the district in terms of redistricting, or whether this should be done strictly according to the census count as presented to the county clerk.

The county reapportionment statute, KRS 67.045, says at subsection 2 that the boundaries of magisterial districts shall be drawn so that “the population of each district shall be as nearly equal as is reasonably possible.” Although subsection 3 of the statute provides for reapportionment to take place during “the second year following the decennial census of the United States,” there is no express requirement that the census figures be used in determining the population of the districts.

Courts have held that the federal census may be the best evidence of local population, Woodring v. Walter, 118 A. 510 (Pa. 1922); that the federal census provides legal evidence of population, Fulham v. Howe, 14 A. 652 (Vt. 1888), and State ex rel. Smith v. Neal, 65 P. 188 (Wash. 1901); and that courts may take judicial notice of the federal census, Brown v. Reeves, 92 So. 825 (Miss. 1922). We believe that the federal census should be accorded a strong presumption of correctness in determining the population of county districts. The reapportionment commissioners should deviate from the census figures only upon clear and convincing evidence of an inaccuracy in those figures.

We emphasize that the evidence must be very strong in order for the commission to deviate from the official census. For example, voter registration records do not by themselves provide strong evidence because the residency standard for voting is based on intent rather than actual presence. Under certain circumstances a voter may register to vote in Floyd County even though he resides elsewhere. The fact that the census did not include that voter in Floyd County's population does not mean that the census is inaccurate; it simply means that the census used a different standard for determining residence.

Moreover, we believe that evidence before the commission must be county-wide in order to justify using a different population figure. If the commission determines that the population of a particular precinct is higher than the amount reported in the census, the commission cannot immediately reject the census figure, since the same underreporting may have occurred in other precincts. In other words, before the commission may reject the census figures because they are inaccurate in one district, it must determine that the figures are accurate in the other districts.

Based on these conclusions we believe that the census figures should be rejected only in extremely rare cases. We do not believe that the facts recited in your letter, which are based solely on voter registration information, constitute sufficient justification for deviating from the census.

You also ask “whether we can establish new boundaries for the voting precincts within the various magisterial districts as well.” Under KRS 117.055 precinct lines are to be drawn by the county board of elections. The reapportionment commission has no authority to establish a precinct boundary.


Chris Gorman

Attorney General

Ross T. Carter

Assistant Attorney General