6

We do not know what procedures we would use if we were required to exclude undocumented immigrants. The proposed methods for excluding undocumented immigrants that we have examined so far--including direct enumeration and estimation--would cause problems in carrying out the census and would be likely to introduce significant errors into the census. The problems would differ depending on what we would be directed specifically to do.

General Problems

In general, any change in our plans for the 1990 census could impose
serious timing problems that could affect the conduct of the census.
As I have testified several times before the Subcommittee on Census and
Population, we are on a tight time schedule with regard to getting the
questionnaires printed. We will submit the final questionnaires to the
OMB next month. Without questionnaires, we cannot conduct the census.
Another general concern is that changes in our plans now would mean that
we might have to conduct the 1990 census with untested and unproven
procedures. This could increase risk in the census.

Specific Problems

Now, I will discuss some specific problems with proposed methods for excluding undocumented immigrants.

One seemingly plausible approach would be to determine the legal status of every person. This causes both perceptual and operational problems that could jeopardize our ability to take the census. Asking about legal status could result in the misperception that the information we collect is used

7 for enforcement. This could seriously threaten cooperation even among legal residents. The answers obtained from such an inquiry would be questionable at best. Indeed, many respondents simply would not be willing or able to answer honestly or accurately. We would be asking respondents to make a legal determination, a process that normally follows a judicial procedure. We face similar problems if we direct undocumented immigrants simply to exclude themselves. Legal residents may find this a convenient way to excuse themselves from participating. Because the Census Bureau could not distinguish the legal and undocumented resident nonparticipants, there could be a large undercount among legal residents.

Another hypothetical procedure might be to estimate the number of undocumented immigrants from a separate information source and subtract them from the census totals. Concerns with possible alternatives for making these estimates include the lack of an appropriate information source; timing (the estimates could not be developed until the census is complete, thus delaying delivery of apportionment and redistricting counts); and the possibility that the resulting estimates would contain errors serious enough to affect the allocation of Congressional seats.

Additional Concerns

Now, I will discuss some additional concerns we have about excluding undocumented immigrants.

One of the ways we decide what to ask in the census is whether the subject is suitable for self-enumeration. This is important because we expect about 78 percent of the households in the country to answer their questionnaires

8

and return the forms by mail. The determination of legal status among types of aliens is not straightforward. In many cases, it requires administrative review by the Department of Justice. Many aliens do not know precisely what their legal status is. Neither the Census Bureau nor our enumerators in the field have the legal qualifications to assist respondents in answering such a question.

I will illustrate further some of the complexities:

Can only aliens who have been lawfully admitted as permanent residents be included in apportionment counts? This category of aliens may not be easily identifiable in either of the proposed methods. Other aliens, such as refugees and students, may be confused about their legal status and misreport themselves. Although INS records can give us numbers of persons admitted to the country as permanent residents, they do not show how many of those persons have subsequently left the country nor do they include current addresses for those who have stayed. This information is crucial to using INS records to estimate the number of aliens to be excluded.

What about aliens who apply for amnesty under the Immigration Reform and Control Act of 1986? If they have not completed the application process for permanent residence status by April 1, 1990, they would be excluded from the apportionment counts.

There are also groups here under an indefinite status, neither legal nor illegal. These include persons in the "extended voluntary departure" or "temporary safe haven" categories.

Mr. Chairman, that concludes my answers to the questions in your letter of invitation.

STATEMENT OF DAVID HUCKABEE AND THOMAS DURBIN

Thank you, Mr. Chairman and Members of the Committee.

My colleague, Mr. Durbin, and I are here at the Committee's request to discuss and respond to questions about the implications of excluding illegal aliens from the census counts used for reapportioning the House and to explain how apportionment works. As the Committee is aware, CRS takes no position on public policy issues. Accordingly, what we say here, and in our reports submitted for the record, reflect only the results of our research on technical, legal and practical aspects of this proposal.

Mr. Durbin is submitting for the record a CRS report that concludes that the weight of constitutional interpretation and analysis raises doubt concerning the exclusion of illegal aliens in the census counts used for apportionment. Mr. Durbin will be pleased to entertain questions about this. I am submitting for the record a report by me and my co-author, Dr. Jennifer Williams, that examines this issue from a practical perspective focusing on: What are some of the problems of enumerating the illegal alien population; and what effect might excluding illegal aliens have on the apportionment?

At the outset, let me briefly describe how the House is reapportioned by the Census Bureau. The intuitive and simple way to apportion seats in the House is to divide the U.S. population by 435 to get an "ideal district size." You then divide each State's population by the "ideal district size," and allocate to each State the number of seats indicated, rounding up or down as appropriate. There are two major problems with this approach when you consider that the House size is fixed at 435. First, it is possible that some States's populations might possibly be so small that they would be "entitled" to less than half a seat, yet the Constitution requires that each State must have at

Washington, D.C. 20540

Congressional Research Service
The Library of Congress

APPORTIONMENT OF THE HOUSE AFTER THE 1990 CENSUS:

CONSTITUTIONAL AND PRACTICAL IMPLICATIONS OF EXCLUDING ILLEGAL ALIENS

Testimony before the House Committee on Post Office and Civil Service

by

David C. Huckabee and Thomas M. Durbin

Congressional Research Service

June 24, 1988