the testimony of Dr. George H. Brown, Director of the Bureau of the Census, at hearings on June 10th before the Subcommittee on Census and Statistics of the House Post Office and Civil Service Committee.

We vigorously urge and recommend that the Administration reconsider its action and give strong affirmative support to a mid-decade census.

The proposal for a mid-decade census has been the subject of extensive study and comprehensive hearings before appropriate Congressional Committees since October, 1961. Over the course of the past ten years, 15 days of hearings have been held, with a total hearings record of over 380 statements in over 1,000 pages.

Almost all levels of the Federal government, along with State and local governments, and many private organizations and individuals have given their unqualified endorsement of a mid-decade census. Its supporters included Members of Congress, Governors of States, Mayors of cities, Cabinet Members, Federal officials, associations and chambers of commerce, private research and advertising firms, university professors, business executives, newspapers and other publications, and private planning organizations.

The record indicates a wide range of essential needs and uses for data that would be collected in such a census. More recent hearings have indicated that there are even stronger justifications than ever before in having census data available more often than once in ten years.

At hearings before both House and Senate Committees, held in 1967 and 1968, the then Bureau of the Budget went on record as recommending a mid-decade census. At the same hearings, the Department of Commerce supported pending mid-decade census legislation which was identical to a draft of legislation that the Acting Secretary of Commerce submitted to the Congress on March 17, 1967. On August 10, 1967 the House passed a mid-decade census bill, H.R. 7659, by a 2-to-1 vote. The vote was: yeas 255, nays 127, not voting 50. The bill had strong bipartisan support, with 18 bills similar to H.R. 7659 introduced by members on both sides of the aisle.

In addition to hearings on the specific subject of a mid-decade census, in 1966 the Senate Subcommittee on Executive Reorganization held year-long hearings on the Federal role in urban affairs. After these hearings, the Chairman of that Subcommittee concluded that a mid-decade census of cities was sorely needed and introduced a bill that would provide for a census of urban areas every five years. He said "A continuing problem in our inquiry into urban affairs was the inability of Government and other witnesses to provide specific demographic and social statistical information. It became apparent from the outset of our hearings that little was known about our urban areas and the conditions that exist. The best information we were able to gather was based on personal observations. Statistical information was fragmentary and dated. Most of the statistics came from the 1960 census, which was taken six years prior to the start of our hearings. But it is not enough to have up-to-date general statistics on housing, unemployment, and populations of cities. We must have these figures on a neighborhood-by-neighborhood-even block-by-block-basis.”

It should be noted that the Census of Population and Housing is now the only one of the regular censuses in the United States that still is taken on a decennial basis. The censuses of agriculture, construction, manufactures, mineral industries, transportation, business, and governments have been on a quinquennial basis for some time in recognition of the fact that the comprehensive data they provided are needed more often than once in ten years.

A number of other countries have found it necessary to take their census more often than once in ten years and have gone over to a five-year census. These include Australia, Canada, Japan, New Zealand, United Kingdom, Ireland, Turkey, and Denmark.

A primary concern of our organization is the development of Federal statistical programs of optimum usefulness at minimum expense. Other than a decennial census, a mid-decade census would be of greater optimum usefulness than any other statistical program of the Federal government. FSUC firmly believes that statistical programs which serve multiple and widely-felt needs should have priority over those which serve limited purposes. A mid-decade census unquestionably fits this criteria. We know of no substitute or alternative that could provide the kinds of data or detail that would be provided in a quinquennial census. Quality ond accuracy would also be lacking in any other approach.

We understand that budgetary situation is the principal reason for the Administration's failure to endorse a mid-decade census. Four options have been proposed for taking a mid-decade census ranging from a mere population count

with six identifying characteristics to a full replication of the 1970 Census of Population and Housing. The range in cost would be from a low of $150-$160 million to a high of $230-$240 million. In two instances the costs would be spread out over a four-year period and in the other two over a five-year period. The maximum cost in any fiscal year for these four alternatives would range from $90-$110 million to $140-$160 million.

As with any statistical program, the issue in effect is "would the additional cost justify the benefits that are produced?" We believe it would. Former Congressman Thomas B. Curtis (R-Mo.), a senior member of the Joint Economic Committee made the following comments in November, 1967 regarding the benefits and costs of census data. These remarks were related to decennial census data, but we believe they are equally applicable to a mid-decade census. Mr. Curtis said, "The Census of Population and Housing should be viewed in its total context. It is central to our entire statistical system-public and private. The Federal statistical system is engaged in transforming inputs of labor and capital into information which has value. Thus, from an efficiency standpoint, a particular type of information should be collected and processed in a manner that is least expensive, most accurate, and most useful for statistical purposes. Presumably, the information which is most valuable to the country will be produced and the value of the uses of the information will equal or exceed the alternative costs of the resources used." We concur wholeheartedly with this statement.

It was pointed out in hearings of the 89th Congress that while the savings inherent in the mid-decade census plan are difficult to state in concrete terms, it is estimated that with the billions of dollars which are being expended based on statistics as old as six years or more, a savings at least equivalent to the cost of a mid-decade census will accrue through efficiencies introduced in the use of up-to-date data. Particularly significant is the fact that more than $10 billion annually is now allocated by the Federal government to the States or their subdivisions according to formulae using population or housing data as a major component. Additional large amounts are distributed by States to their counties and municipalities largely on the basis of population.

In recent testimony, the Director of the Bureau of the Census said: "Since 1960, the Congress has enacted some 23 laws which require population information for their proper administration, and the appropriate distribution of funds. Some of these funds are to be distributed on the basis of total population, some on the basis of the population in specified groups (as children of school age), and some on the basis of more complicated formulas such as the Aid-to-Education program, which requires information about the number of children of school age living in families whose income was below a specified level. Equitable distribution of funds under many of these programs requires current information on the population of small areas. Programs for economic development, as well as housing, call for data for very small areas.

"Current information will become even more essential as new programs are adopted. Billions of dollars would be distributed in the proposed special revenuesharing programs for manpower training, urban and rural development, education, and transportation. In all these programs, population data represent a significant factor in the allocation formulas not only in grants to States but in the distribution to counties and metropolitan areas within States. At present the sample surveys which provide the national data can provide only a limited amount of information for regions, and for major metropolitan areas."

The most clear and concise statement of the specific governmental uses for census data and the needs for current data was presented in testimony by David B. Walker, Assistant Director of the Advisory Commission on Intergovernmental Relations. He said:

"Census data provide the factual basis for a multitude of policy decisions by the Congress and Executive Branch, by governors and state legislators, and by local chief executives and legislative bodies. This information is required for: (a) the development of long-range plans for public facilities and services at all levels of government; (b) the allocation of block grants and categorical aid by the Federal Government; (c) the distribution of State grants-in-aid; and (d) the measurement of changing demands on the Federal, State, and local governments as a result of shifting population patterns. Moreover, if revenue sharing is enacted, it will be essential to have reliable recent census data to compute the State area entitlements and substate 'pass through' allocations.

"Because governmental problems are now and will continue to be shaped by the growing urbanization of our population, more current statistical measures of the social, economic, and characteristics of the American people are indispensable for designing and implementing public programs to meet these needs. We emphasize the significance of currency, for the pace of population movements is altering substantially the characteristics of many urban as well as rural areas over relatively short periods of time. Yet, the Census of Population and Housing remains the only one of the censuses in our Nation that is taken on a decennial basis."

Members of the Advisory Commission on Intergovernmental Relations include three Members of the U.S. Senate; three Members of the House of Representatives; three officers of the Executive Branch of the Federal Government including th Counsellor to the President, the Secretary of Housing and Urban Development and the Director of the Office of Management and Budget; four governors; four mayors; three members of State legislative bodies, three elected county officials and three private citizens.

In addition to the widespread governmental needs mentioned, American business has an actual and vital need for up-to-date information to guide it in its many policy, operating and investment decisions. U.S. business enterprises need current information for analyzing the demand for and supply of products and services in a trillion-dollar economy. They need to know the facts about labor markets. They need up-to-date information in order to make projections of economic growth. They especially need current census data for determining sites for plants, warehouses and service facilities. According to your 1971 Economic Report, business fixed investment amounted to $99.3 billion in 1969 and is projected to reach $192 billion in 1975 and $198 billion in 1976. This is a vast expenditure of money which will have a considerable impact on the economy. It is essential that American business have adequate and current data to make intelligent decisions for the proper allocation of its funds and resources for these purposes. A misplacement of investment or wrong investment would be not only costly to the companies involved but to the economy as a whole.

With regard to the needs for current data by business in planning site locations, we direct your attention to your special messages to Congress on the environment and on clean energy. Both messages pointed up the need for longerrange planning by the producers of electric power to project their future needs and identify environmental concerns well in advance of construction deadlines. The Administration bill for "Power Plant Siting" calls for the establishment of guidelines which, among other things, includes establishing criteria for evaluating relative environmental effects of alternative sites, and criteria for evaluating projected electric needs. Surely, attempts to meet these needs will require certain types of census data available more frequently than once in every ten

years.

Your interest in the Federal statistical system and its improvement was demonstrated by your appointment last year of a special Commission on Federal Statistics. Your concern regarding the broad range of problems associated with population growth and their implications for America's future is evidenced by the establishment of a Commission on Population Growth and the American Future. We commend you for the establishment of these commissions and look forward with great interest to their reports. We hope your interests in population and statistics will not stop there, and that the Administration will recognize the needs of the Nation for having more current information regarding the growth, movement and characteristics of our population.

Although the United States is recognized as having the best statistical system in the world, there is always room for improvement. We believe that the Federal statistical system cannot and should not remain static in this dynamic U.S. economy. With the growth in government and its myriad programs designed to deal with the problems of health, education, welfare, poverty, manpower, etc., it is imperative that our statistical system be responsive to the growing and changing data needs for more current information.

We look to the Administration for taking the leadership in making our statistical system more responsive to the information needs of the Nation and again strongly urge support for a mid-decade census.

Respectfully yours,

JOHN H. AIKEN,
Executive Director,

Federal Statistics Users' Conference.

DUALABS,

SUBCOMMITTEE ON CENSUS AND STATISTICS,

Cannon House Office Building,

Washington, D.C.

(Attention: Mr. George B. Gould).

Arlington, Va., July 13, 1971.

Enclosed is the latest issue of our newsletter, Data Access News, which reports on the recent mid-decade census hearings and has an editorial strongly underlining the urgent need for a census in 1975.

A copy was sent to Mr. George Shultz, Director, Office of Management and Budget, with a covering letter. We sincerely hope the Subcommittee's efforts to forward this legislation and modify the Administration's opposition will be successful.

Enclosure.

(Mrs.) CONSTANCE F. CITRO,
START Program Director,
Editor of Data Access News.

DATA ACCESS NEWS

(June 1971, Volume 3; Issue No. 3)

PROSPECTS FOR MID-DECADE CENSUS DIM-ADMINISTRATION VOICES OPPOSITION AT HEARINGS

The House Subcommittee on Census and Statistics of the Post Office and Civil Service Committee recently concluded a series of hearings on various proposals for a mid-decade census of population and housing. The four major alternatives considered by the Subcommittee were:

1. A census of complete-count population items only providing geographic detail down to the ED and blockgroup level.

2. A census of complete-count population and housing items just as in 1970 with geographic detail down to the level of city blocks.

3. A 25-percent sample survey including the 1970 sample items only, providing geographic detail for states, SMSA's, large counties and cities. One problem with this alternative, as pointed out in Census Bureau testimony, is that not including any complete-count items at all might make it harder to secure the cooperation of the public since the Bureau could not stress that everyone must be counted.

4. A fullscale replication of the 1970 census with all complete-count and sample data items and full geographic detail.

At the first hearing on May 18, the Subcommittee heard from Dr. Ross Eckler, past Director of the Census Bureau, Mr. Walter Duffett of the Dominion Bureau of Statistics, who commented on Canada's experience with taking a fullscale census every five years, and Mr. John Aiken, Executive Director of the Federal Statistics Users' Conference. Mr. Aiken expressed complete support for a middecade census comprehensive enough to provide small-area detail, particularly on basic housing characteristics, although perhaps a fullscale repeat of 1970 would not be necessary.

At the second hearing on June 2, witnesses representing all levels of government as well as trade and professional associations uniformly expressed an urgent need for a mid-decade census in 1975 and every ten years thereafter. Generally, they favored a complete enumeration providing small-area data, although not necessarily including all the content of the 1970 questionnaire.

Rep. Charles H. Wilson (D.-Calif.), Chairman of the Subcommittee, noted that the testimony on June 2 amply reinforced the previous experience of the Subcommittee from extensive hearings held in prior years. Rep. Wilson was particularly pleased at the support for a mid-decade census voiced by the Advisory Commission on Intergovernmental Relations. This body includes Congressmen, Senators, Cabinet members, Governors, Mayors, and County officials. George Shultz, Director of the Office of Management and Budget, is a member and Wilson hoped that the support of the Commission would win over the Executive Branch which in the past decade played a decisive role in opposing a mid-decade census.

At the final hearing, June 10, Dr. George Brown, Director of the Bureau of the Census, documented the need for up-to-date population figures for small areas and provided cost estimates for each of the four major alternatives being con

sidered by the Subcommittee. The costs ranged from $150 million minimum for alternative 1, the most restricted proposal in terms of subject content, to $240 million maximum for a fullscale replication of the 1970 census.

When asked to convey the Administration's position on a mid-decade census, however, Dr. Brown had to state that the Administration remained opposed to a census in 1975, regarding the project as too expensive. Instead, the Administration will ask for a supplemental budget appropriation of $500,000 for the Census Bureau to conduct a 12-month feasibility study of using administrative records for making small-area population estimates.

Rep. Wilson expressed his dismay at the Administration stance, particularly since he had been hopeful that the present series of hearings would only need to decide among the various possible alternatives, previous hearings having more than established an overwhelming need for a census-type operation to obtain more timely small-area data. Wilson asked Dr. Brown to find out the strength of the Administration's opposition and specifically whether the President would veto a mid-decade bill if it passed Congress. Wilson felt that despite the great need, if a Presidential veto was likely, there was not even much point to the Subcommittee's reporting out a bill for a vote. However, Wilson hoped it would not come to that and expressed his intention to bring pressure to bear on the Administration from census data users. (See following story for Wilson's statement.)

REP. WILSON BLASTS ADMINISTRATION POSITION ON MID-DECADE CENSUSDAN CONCURS WHAT CAN BE DONE

Following immediately upon testimony from Census Bureau Director George Brown stating the Administration's opposition to a mid-decade census, Rep. Charles H. Wilson's office issued a press release denouncing this stand. The text of the release, dated June 10, is reproduced below.

"CENSUS CLOCK DIED TODAY"

Congressman Charles H. Wilson (D.-Calif.) blasted the Administration today in a hearing of the Subcommittee on Census and Statistics. Wilson, who is Chairman of the Subcommittee, was told by Dr. George H. Brown, Director of the Bureau of the Census, that the Administration "does not recommend a mid-decade census."

The California lawmaker angrily retorted, "I am deeply disappointed and very perplexed by the Administration's refusal to endorse the mid-decade census. Quite frankly, I cannot understand how the Administration can oppose the proposal of a mid-decade census when all levels of the Federal Government, along with State and local governments and including recognized national organizations representing statistical users of all types, have told us of their unqualified support for a mid-decade census."

Chairman Wilson went on to say that his Census Subcommittee has heard from hundreds of witnesses over a period of two and a half years without hearing one single witness oppose the mid-decade census. Wilson continued, "Recognizing that a mid-decade census could cost around $140 million [sic], it must be emphasized that our nation distributes over $10 billion in Federal funds using census data as a guide. There is no doubt in my mind that relying on old data from the decennial census later in the decade costs our Government considerably more than $140 million."

Wilson concluded by saying, "I plan to contact every witness who has appeared before us-governors, mayors, county supervisors, representatives of census user organizations, and members of the business community-asking them to contact the President advising him of their desperate need for a mid-decade census. Let me make it perfectly clear that if our nation goes without a mid-decade census, it is the fault of the Administration-not Congress. The President will have to accept the responsibility of robbing our nation, especially our minority citizens and our other citizens in our newly developed suburban areas, of a tool which would develop data of great need while saving our citizens millions in tax dollars. The Administration today has broken-if not destroyed-the census clock.

(The Census Clock is on Exhibit at the lobby of the Department of Commerce. It has a meter which records the changing population in clocklike fashion daily.) Data Access News concurs in the opinion that a mid-decade census will be essential by 1975 to update the small-area population and housing figures from the 1970 census. The money would be well spent in terms of the benefits from 63-434-72—16