Department of Commerce v. United States House of Representatives

PETITIONER: Department of Commerce
RESPONDENT: United States House of Representatives
LOCATION: Residence of Brenda Roe

DOCKET NO.: 98-404
DECIDED BY: Rehnquist Court (1986-2005)

CITATION: 525 US 316 (1999)
ARGUED: Nov 30, 1998
DECIDED: Jan 25, 1999

Michael A. Carvin - Argued the cause for the appellees in No. 98-564
Maureen E. Mahoney - on behalf of the Appellees in No. 98-404
Seth P. Waxman - Department of Justice, argued the cause for the appellants

Facts of the case

Under the Census Clause (Art. I, Sect. 2, Cl. 3), Congress is authorized to conduct a census of the American public every 10 years. Among other purposes, the census provides a basis for apportionment of congressional districts. Under the Census Act, Congress delegated this responsibility to the Secretary of Commerce (Secretary). When the Census Bureau (Bureau) announced plans to use two new forms of discretionary statistical sampling in the 2000 census, various United States residents, counties, and the House of Representatives challenged the constitutionality of the new sampling methods in two separate suits. On direct appeals from three-judge district courts enjoining the use of the new sampling methods, the Supreme Court consolidated the cases and granted certiorari.


Is the use of statistical sampling in the execution of the census inconsistent with provisions of the Census Act or in conflict with the Census Clause of the Constitution?

Media for Department of Commerce v. United States House of Representatives

Audio Transcription for Oral Argument - November 30, 1998 in Department of Commerce v. United States House of Representatives

William H. Rehnquist:

We'll hear argument now in No. 98-404, the Department of Commerce v. the United States House of Representatives, and William Jefferson Clinton v. Matthew Glavin.

General Waxman.

Seth P. Waxman:

Mr. Chief Justice, and may it please the Court:

Article I, Section 2, Clause 3 of the Constitution requires that representatives be apportioned based on the number of persons in each State.

And to effectuate that requirement, Congress is directed to provide for, quote, the actual enumeration, to occur every 10 years, in such manner as Congress shall by law direct.

In modern times, as the census has faced increasing challenges, Congress has delegated authority to the Secretary of Commerce to conduct a census in such form and content as he may determine.

Following exhaustive study and the unanimous recommendation of the Census Bureau, the National Academy of Sciences, and other professional groups, the Secretary has determined that for the 2000 census, employing statistical sampling, in addition to other means of enumeration, will best achieve the constitutional goal of determining the number of persons in each State.

The important--

William H. Rehnquist:

General Waxman, I understand there are... there are two kinds of statistical sampling involved here.

And I want to ask you about the first one... rather the second one... which I... I gather is a series of 750,000 housing units, selected randomly.

And they will then be used to adjust the figures from census tracts?

Seth P. Waxman:

--That's correct, Mr. Chief Justice.

If... just a small correction in the premise of the question.

There are three methods of sampling that the Census Bureau proposes to conduct.

One is not challenged by... at least directly... by any of the plaintiffs in this case.

But the third one, the integrated cover ment measurement portion of the survey, is actually quite similar in... in the means by which it will done... is going to be done... as in 1990, the post enumeration survey that the Court discussed in Wisconsin v. City of New York.

That is, after the first two phases of the census are completed and there is an initial enumeration roster with respect to every census block and tract, the--

William H. Rehnquist:

Well, that initial enumeration is first based on responses and then based on follow up interviews?

Seth P. Waxman:

--That's correct.

There are... there are three phases, Mr. Chief Justice, to the 2000 Census.

The first phase will be a... a series of mailings to every individual household, an 800 number for people who don't want to respond by mail to call in their answers, the distribution of questionnaires in malls and public libraries and other places.

That's the first phase.

The second phase involves what's called non response follow up.

And that includes both an effort to physically visit a certain number... up to 90 percent of the houses in each census tract, and the use of statistical sampling to impute the population characteristics with respect to those homes which did not send in... mail in their census information and were not located personally.

Antonin Scalia:

But when you say up to 90 percent, there... there is going to be an intentional effort not to do 100 percent?

Seth P. Waxman:

No, there... Justice Scalia, there is an intentional effort in this census to obtain census information from every household that is known to exist by mailings, by a Be Counted program, by telephone, by the Internet.

The Census Bureau has determined that in order to make the population totals for apportionment purposes more accurate, it will not have enumerators physically go to every single house that did not respond to those initial requests for information, but rather will go to enough houses so that 90 percent of every household in each census tract has been the subject of a, quote, physical count.

And it will then use statistical sampling on a completely random basis... and I think this is key, because none of the plaintiffs have alleged, nor could they, that they have been injured in any way by the effort... by the attempt to use sampling--

William H. Rehnquist:

Let... let's... let's stay on the question of describing this... this procedure.

Seth P. Waxman:

--Yes, sir.