Removal of the mandatory response requirements would make it impossible, for example, for Census to vouch for the accuracy or completeness of data if respondents could determine whether or not, and how they would answer specific questions. To exempt housing questions from the mandatory response requirement would ignore completely the dependence of the Congress, the Administration, state and local governments, and the building industry upon the availability of reliable facts on housing and its occupants. Quantification of the Nation's housing needs and planning effective programs for the meeting of these needs depend heavily upon reliable data on the size, condition, and extent of utilization of the existing housing stock. It requires information on the characteristics of the Nation's households such as their income, family size, and place of work. The inability of Census to collect the necessary information in an orderly and systematic fashion would eventually lead to a proliferation of special surveys by a wide variety of current users of Census data. Rather than reducing the burden upon the public, the end result could be an increased burden. Sincerely yours, ROBERT C. WEAVER. U.S. DEPARTMENT OF LABOR, Hon. CHARLES L. SCHULTZE, Director, Bureau of the Budget DEAR MR. SCHULTZE: This is in response to your request for my views on H.R. 10952 and the report thereon of the Department of Commerce. The bill would "amend title 13, United States Code, to limit the categories of the questions required to be answered under penalty of law in the decennial census of population, unemployment, and housing, and for other purposes.” Although I am sympathetic with the objective of preserving a citizen's privacy, I oppose limiting the census questions which are to be answered on a mandatory basis. The public interest to be sacrificed is substantial, whereas we are aware of no evidence supporting a countervailing need for the additional protections afforded by the bill in view of the anonymity preserved through the use of census data. I fully concur in the views of the Department of Commerce in its report on H.R. 10952 dated August 1, 1967, to the House Committee on Post Office and Civil Service. I would reinforce the conclusions of the Commerce Department concerning the undesirable effects which the enactment of H.R. 10952 would have on the conduct of the census, as well as on the reliability and usefulness of its content, by pointing out that the Department of Labor also would be particularly affected by the bill's provisions. The decennial census is the basic source of detailed information for the monthly survey on employment and unemployment which the Bureau of the Census conducts for this Department's Bureau of Labor Statistics. This information in turn provides monthly economic indicators, such as the unemployment rate, which are so important to the overall economic policy of the Federal Government. We might add that State and local governments also require certain data from the decennial census for the effective formulation of their economic policies; and for them the decennial census is the only source of much of the employment and economic data they need. More concretely, the Department of Labor needs complete and unbiased labor force data, for example, for more effectively planning and administering the various programs aimed at locating and alleviating pockets of underemployment and employment, especially among younger persons. We need to know more about the economic situation of persons who have migrated in recent years from the rural areas to the city in order to improve assistance for these people in their efforts to respond to the challenges this transition presents for them. To jeopardize the gathering of necessary census data and to depreciate the validity of its collected data by removing the mandatory basis necessary for the effective conduct of the decennial census would, in my opinion, create a serious imbalance contrary to the public interest. Sincerely, W. WILLARD WIRTZ, Hon. THADDEUS J. DULSKI, U.S. DEPARTMENT OF JUSTICE, Chairman, Committee on Post Office and Civil Service, House of Representatives, Washington, D.C. DEAR MR. CHAIRMAN: This is in response to your request for the views of the Department of Justice concerning H.R. 10952, a bill "To amend title 13, United States Code, to limit the categories of questions required to be answered under penalty of law in the decennial censuses of population, unemployment, and housing, and for other purposes." The bill would amend title 13, United States Code, by inserting immediately after section 141 a new section (141A) limiting the categories of information required under penalty of law in certain censuses. Refusal or neglect to furnish information not within the categories listed in this section would not be an offense under section 221 of title 13, U.S.C. In addition, the Secretary of Commerce may request additional information in connection with the census of a voluntary basis. Section 141 of title 13, United States Code, now provides that the Secretary of Commerce shall in the year 1960 and every ten years thereafter, take a census of population, unemployment, and housing (including utilities and equipment ) as of the first day of April, which shall be known as the census date. This measure would operate in an area for which the Department of Justice does not have primary responsibility. Accordingly, whether this legislation should be enacted involves policy considerations as to which the Department of Justice makes no recommendation. The Bureau of the Budget has advised that there is no objection to the submission of this report from the standpoint of the Administration's program. Sincerely, WARREN CHRISTOPHER, (The following are the statements and letters submitted for the record by Congressman Betts, all of which are excerpts from the Congressional Record.) [From the Congressional Record, Sept. 28, 1967] THE CENSUS: COMPULSORY VERSUS VOLUNTARY APPROACHES Mr. BETTS. Mr. Speaker, the 1970 decennial census of population and housing is now in its planning stages. The chairman of the Subcommittee on Census and Statistics of the Post Office and Civil Service Committee, Congressman Green of Pennsylvania, has indicated this subcommittee will meet later this year to examine the proposed questions to be asked the American people on that census. In anticipation of this review of the census questionnaires, justification for including many subjects and the mandatory nature of this inquiry, I have investigated many facets of data gathering operations of the Bureau of the Census. As I see it, there are four principal areas of concern which confront the Congress as the 1970 census approaches. First is the concept and use of mandatory features of this census; second deals with nonessential questions included in recent censuses and proposed for continued use; third relates to the rights of privacy infringed upon by this extensive public interrogation by the census; and fourth is the matter of competition with private market research firms or nonprofit institutions in the conduct of many census projects. For the purposes of this discussion, let me initially consider the matter of compulsion which is sometimes said to be the vital and prime source of providing complete and accurate decennial census statistics. The latter three areas of concern which I mentioned will be presented in succeeding reports. Section 221, title 13, United States Code, provides penalties-$100 fine and up to 60 days in jail-for noncompliance with various censuses-including the decennial Census of Population and Housing-conducted by the Bureau of the Census. An examination of the theory behind this mandatory provision, the number of violators prosecuted in recent years, and the deterrent affect it is contended to have on compliance, would be useful at this point. Statements and correspondence I have received from Dr. A. Ross Eckler, Director of the Bureau of the Census indicate that in his opinion the penalty provision and official nature of the census forms lend significantly to maximum compliance by the general public. The mandatory provision prompts greater cooperation with enumerators and would similarly result in better response when the mail-out/mail-back method of conducting future censuses is employed, Dr. Eckler contends. The mandatory rule also is said to discourage organized local opposition to the census. In order to receive an over all view of this provision, I asked the Attorney General and Director of the Census their policy toward enforcement of this penalty section: "What has been the policy or attitude of the Department of Justice (Bureau of the Census) toward the enforcibility of this statute and what steps have been taken to provide procedures for the Bureau of the Census and its enumerators to apprehend violators? How many incidents or cases of violations of Section 221 were reported to the Department of Justice in conjunction with the 1960 census and how many individuals were actually prosecuted under this provision?" Fred M. Vinson, Jr., Assistant Attorney General advised me on September 8, 1967, that: "Whenever the Department of Commerce feels that the facts surrounding a refusal to furnish desired census information justify prosecution, the file in each case will be forwarded by the department to the appropriate United States Attorney. In all instances of refusal to answer Census questionnaires affecting companies, businesses and other organizations, the United States Attorney should make certain that efforts have been made to persuade the delinquent to comply with the Census Bureau's report. Prosecution should be instituted under 13 U.S.C. 224 if the delinquent persists in refusal to supply the required census data." The Justice Department has no record of how many prosecutions were requested in conjunction with the 1960 census, but at least two convictions were reported, Mr. Vinson indicated. It appears from the response I received from Acting Director, Bureau of the Census, Robert F. Drury, that no figures are available on the number of persons refusing to give information to an enumerator, or the number of cases involved in informal counseling with local U.S. attorneys. I can only conclude that the need to pursue the enforcement provision is minimal, reflecting well on the attitude of the American public on filling out census questionnaires. This is not to say the American people like these requests but nevertheless they patriotically have complied with them. The Bureau of the Census has focused its fears frequently on the "vulnerability of the decennial census to organize local groups" who would thwart its completeness and/or accuracy. Dr. Eckler on August 2 wrote me : "A major concern with your proposal (H.R. 10952) for eliminating the mandatory reporting requirement for certain of the questions asked in the decennial census lies in possibility that organized local or national campaigns urging citizens not to answer particular census questions, which certainly are in prospect if the law is so changed, undoubtedly would make part of the census results unusable." Shortly thereafter, I asked Dr. Eckler to document the basis upon which he made that assertion: "You referred to the vulnerability of the decennial census to local efforts to discourage public cooperation and that the penalty provision for compliance was vital to maintain a high level of participation. Would you give me whatever experience you have had with efforts to thwart full participation with the Bureau in decennial censuses? How extensive have these revolts been and what effect have they had on a local or regional collection of statistics?" (Emphasis added.) Dr. Eckler's reply of August 9 gave not one instance where any groups of citizens had organized on a local or national level or been in collusion to sabotage a decennial census or a portion thereof. In his letter Dr. Eckler referred to scattered resistance by businesses to form requirements and one instance of city government officials opposing a certain questionnaire, but completely failed to show that his charge of organized local efforts working against a census has any validity. Mr. Speaker, I believe this contention, that people will not cooperate if a portion of the decennial census is voluntary is a smokescreen for complete unwill ingness to allow the principle of volunteerism in the gathering of data by the Federal Government. I asked Dr. Eckler to test this principle, to conduct a pilot project to determine the variance in response between a survey where the respondents believe or are told their participation is required under penalty of law from a survey where compliance is clearly stated as being voluntary. I suggested that perhaps such a test case could be conducted in one of your periodic mail surveys which would not require a great deal of additional effort or expense. The answer I received was less than enthusiastic. Dr. Eckler wrote: "To carry out a pilot project of the type you suggest in your letter of July 19 poses some operational problems which are not readily dealt with." Even if testing the principle of volunteerism with a special pilot project is ruled out by Dr. Eckler, the Bureau of the Census already admits that wholly satisfactory results are possible with this approach. Let us look at results from sample surveys and quote the Bureau of the Census: "In addition to the decennial censuses, the Census Bureau regularly conducts many household surveys, covering a variety of subjects. Among the most important is the Current Population Survey, which has been conducted monthly for over 20 years and serves as the source of the official government statistics on total employment and unemployment. A sample of approximately 52,000 households throughout the Nation is visited each month in connection with the survey, and the results are published by the U.S. Department of Labor. Households are selected for the survey by address only, using scientific sampling methods. Each household is visited once a month for four consecutive months one year, and again for the same time period a year later. This technique permits us to obtain the needed information while minimizing the inconvenience to any one household. "Participation in the Current Population Survey is voluntary. However, we have had cooperation over the years from the vast majority of the people contacted. Fewer than 2 percent, on the average, refuse to participate in the survey. In some other surveys the refusal rates have been higher, depending in part on the nature of the survey." Now these results may not bear on a 100-percent census, but I believe they would prove feasible for a 20-percent, 15-percent, or a 5-percent sampling undertaken to secure data in a decennial census. In almost every communication I have received from the Census Bureau, officials praise "the demonstrated cooperation by the American public." The American people, if I am any judge of their character, are more than willing to cooperate with the Federal Government to provide the basic, essential information to meet constitutional requirements on population and other facts about themselves. There is little disagreement that questions seeking this information should, if necessary, carry penalties for noncompliance. Our citizens do object, however, to harassment, invasion of privacy or questioning which has no public purpose. That is why I wish to limit the number of mandatory questions and require a separate, voluntary form for any extraneous inquiries the Census Bureau wishes to pose. I believe this plan would work. In reviewing the Census Bureau's reasons and justifications for mandatory provisions covering any and all questions they care to ask, I find their rationale faulty both in theory and practice. To provide documentation to this position, in addition to the logic the facts themselves reveal let me turn to two other major statistic gathering sources in the United States. I refer to the 50 State governments and the large number of private market research firms in this country. STATE CENSUSES VOLUNTARY Many State agencies conduct censuses and surveys to obtain vital information for the operations of State government. In order to learn the extent of such State statistic gathering activities, the U.S. Bureau of the Census itself in 1965 submitted a questionnaire to numerous agencies in each State to determine the scope and type of work they have undertaken. The principal emphasis of this census questionnaire was on population and housing information of the same type that appears on the decennial Federal census. Here is a summary of the reports the Bureau of the Census received: "In all but one State, North Dakota, some State agency reported making population estimates for counties or other local areas. In a number of instances, census counts rather than estimates are available. Thus, the State of Kansas takes a State census every year as of March 1. Massachusetts takes one in years ending in 5; the results of the last one, taken as of January 1, 1965, have recently become available. The Washington State Census Board counts the population in selected places and supplements these counts with estimates of the population of other cities and towns. In all other instances, the data reported here represent population estimates derived by various methods. "As in earlier surveys, the State departments of health lead other types of Statewide agencies in the preparation of local population estimates. Out of a total of 66 different State agencies making such estimates, 27 were departments of health. This is approximately the same number of State departments of health reported as preparing estimates in our earlier surveys. State universities are the second most important source of such estimates; 21 such agencies reported making population estimates. Ten of these were Bureaus of Business Research at State universities and the remainder were represented by Departments of Sociology and newly established Population Study Centers. Other types of agencies preparing estimates were: economic development commissions (6), employment security commissions (4), State planning commissions (3), and other agencies. These agencies include the State Census Boards in Oregon and Washington. In the State of California, population estimation is the responsibility of the Population Research Unit in the Department of Finance. In Utah, an interagency committee has the responsibility for such estimates. "Table A below summarizes the sources of population estimates by type of agency preparing such estimates. The results from the earlier surveys are also shown for comparative purposes. In general, the changes reported over time are true representations of shifts in responsibility of preparing such estimates. It is quite possible, however, that the increase in the total number of agencies reporting work in this area since 1960 reflects the more extensive coverage of the 1965 survey. "TABLE A.-STATE AGENCIES MAKING POPULATION ESTIMATES FOR LOCAL AREAS, PERIODIC 1 Includes California State Department of Finance, Kansas State Board of Agriculture, Utah Population Committee, Washington State Census Board, and the Office of the Secretary of the Commonwealth of Massachusetts (census every 10 years). This is an impressive report of what our States are doing themselves in the field of data gathering. However, it was not revealed in this Census Bureau report whether such State agencies have been given mandatory powers to secure such data from citizens and businesses. So, I wrote to the attorneys general of the 50 States to ask that question. The replies are significant. I received 45 responses and only in Massachusetts and Missouri did the attorney general indicate statutes existed requiring public compliance with agencies seeking general population and housing information. My two questions to the attorneys general and their responses are presented in the following table: |