[House Hearing, 106 Congress] [From the U.S. Government Publishing Office] THE AMERICAN COMMUNITY SURVEY--A REPLACEMENT FOR THE CENSUS LONG FORM? ======================================================================= HEARING before the SUBCOMMITTEE ON THE CENSUS of the COMMITTEE ON GOVERNMENT REFORM HOUSE OF REPRESENTATIVES ONE HUNDRED SIXTH CONGRESS SECOND SESSION __________ JULY 20, 2000 __________ Serial No. 106-246 __________ Printed for the use of the Committee on Government Reform Available via the World Wide Web: http://www.gpo.gov/congress/house http://www.house.gov/reform __________ U.S. GOVERNMENT PRINTING OFFICE 73-836 WASHINGTON : 2001 _______________________________________________________________________ For sale by the Superintendent of Documents, U.S. Government Printing Office Internet: bookstore.gpo.gov Phone: (202) 512-1800 Fax: (202) 512-2250 Mail: Stop SSOP, Washington, DC 20402-0001 COMMITTEE ON GOVERNMENT REFORM DAN BURTON, Indiana, Chairman BENJAMIN A. GILMAN, New York HENRY A. WAXMAN, California CONSTANCE A. MORELLA, Maryland TOM LANTOS, California CHRISTOPHER SHAYS, Connecticut ROBERT E. WISE, Jr., West Virginia ILEANA ROS-LEHTINEN, Florida MAJOR R. OWENS, New York JOHN M. McHUGH, New York EDOLPHUS TOWNS, New York STEPHEN HORN, California PAUL E. KANJORSKI, Pennsylvania JOHN L. MICA, Florida PATSY T. MINK, Hawaii THOMAS M. DAVIS, Virginia CAROLYN B. MALONEY, New York DAVID M. McINTOSH, Indiana ELEANOR HOLMES NORTON, Washington, MARK E. SOUDER, Indiana DC JOE SCARBOROUGH, Florida CHAKA FATTAH, Pennsylvania STEVEN C. LaTOURETTE, Ohio ELIJAH E. CUMMINGS, Maryland MARSHALL ``MARK'' SANFORD, South DENNIS J. KUCINICH, Ohio Carolina ROD R. BLAGOJEVICH, Illinois BOB BARR, Georgia DANNY K. DAVIS, Illinois DAN MILLER, Florida JOHN F. TIERNEY, Massachusetts ASA HUTCHINSON, Arkansas JIM TURNER, Texas LEE TERRY, Nebraska THOMAS H. ALLEN, Maine JUDY BIGGERT, Illinois HAROLD E. FORD, Jr., Tennessee GREG WALDEN, Oregon JANICE D. SCHAKOWSKY, Illinois DOUG OSE, California ------ PAUL RYAN, Wisconsin BERNARD SANDERS, Vermont HELEN CHENOWETH-HAGE, Idaho (Independent) DAVID VITTER, Louisiana Kevin Binger, Staff Director Daniel R. Moll, Deputy Staff Director James C. Wilson, Chief Counsel Robert A. Briggs, Clerk Phil Schiliro, Minority Staff Director Subcommittee on the Census DAN MILLER, Florida, Chairman THOMAS M. DAVIS, Virginia CAROLYN B. MALONEY, New York PAUL RYAN, Wisconsin DANNY K. DAVIS, Illinois MARK E. SOUDER, Indiana HAROLD E. FORD, Jr., Tennessee ------ ------ Ex Officio DAN BURTON, Indiana HENRY A. WAXMAN, California Jane Cobb, Staff Director Vaughn Kirk, Professional Staff Member Erin Yeatman, Professional Staff Member Andrew Kavaliunas, Clerk David McMillen, Minority Professional Staff Member C O N T E N T S ---------- Page Hearing held on July 20, 2000.................................... 1 Statement of: Collins, Hon. Mac, a Representative in Congress from the State of Georgia........................................... 15 Crowe, David, staff vice president of housing policy, National Association of Home Builders; Edward Hudgins, director of regulatory studies, the Cato Institute; Chuck Fluharty, director, Rural Policy Research Institute; Richard Kulka, vice president of statistics, health and social policy, Research Triangle Policy, accompanied by Judith T. Lessler, director, Statistics Research Division; and Barbara Welty, president, National Center for Small Communities, Board of Directors, National Association of Towns and Townships........................................ 51 Emerson, Hon. Jo Ann, a Representative in Congress from the State of Missouri.......................................... 9 Spotila, John, Administrator, Office of Information and Regulatory Affairs, Office of Management and Budget, accompanied by Katherine Wallman, Chief Statistician for the United States; and Kenneth Prewitt, Director, Bureau of the Census................................................. 23 Letters, statements, etc., submitted for the record by: Collins, Hon. Mac, a Representative in Congress from the State of Georgia, prepared statement of.................... 17 Crowe, David, staff vice president of housing policy, National Association of Home Builders, prepared statement of......................................................... 55 Emerson, Hon. Jo Ann, a Representative in Congress from the State of Missouri, prepared statement of................... 12 Fluharty, Chuck, director, Rural Policy Research Institute, prepared statement of...................................... 83 Hudgins, Edward, director of regulatory studies, the Cato Institute, prepared statement of........................... 71 Kulka, Richard, vice president of statistics, health and social policy, Research Triangle Policy, prepared statement of......................................................... 91 Maloney, Hon. Carolyn B., a Representative in Congress from the State of New York, prepared statement of............... 7 Miller, Hon. Dan, a Representative in Congress from the State of Florida, prepared statement of.......................... 3 Prewitt, Kenneth, Director, Bureau of the Census, prepared statement of............................................... 33 Serrano, Hon. Jose E., a Representative in Congress from the State of New York, prepared statement of................... 103 Spotila, John, Administrator, Office of Information and Regulatory Affairs, Office of Management and Budget, prepared statement of...................................... 25 Welty, Barbara, president, National Center for Small Communities, Board of Directors, National Association of Towns and Townships, prepared statement of................. 96 THE AMERICAN COMMUNITY SURVEY--A REPLACEMENT FOR THE CENSUS LONG FORM? ---------- THURSDAY, JULY 20, 2000 House of Representatives, Subcommittee on the Census, Committee on Government Reform, Washington, DC. The subcommittee met, pursuant to notice, at 9:30 a.m., in room 2358, Rayburn House Office Building, Hon. Dan Miller (chairman of the subcommittee) presiding. Present: Representatives Miller, Maloney, and Davis of Illinois. Staff present: Jane Cobb, staff director; Chip Walker, deputy staff director; Vaughn Kirk and Erin Yeatman, professional staff members; Michael Miguel, senior data analyst; Andrew Kavaliunas, clerk; David McMillen and Mark Stephenson, minority professional staff members; and Earley Green, minority assistant clerk. Mr. Miller. A quorum being present, the subcommittee will come to order. Today we are here to begin the process of eliminating the problematic census long form. The issue of the census long form hit home in early March when census forms began to arrive in the mail. It was at that time we realized there was a newly surfaced discontent with the nature of the long form questions. From newspapers to television, and from talk radio to congressional offices, everyone was talking about privacy and the perceived intrusive nature of the long form questionnaire. Even though the long form was the shortest it had ever been and only contained one new question since 1990, this didn't seem to matter to some people. They were legitimately concerned about their privacy. This change in attitudes was not simply the Census Bureau's fault. Congress had given its tacit approval of the questions earlier in 1998, but even then, no one had sensed that the privacy concerns would be as intense as they were this spring. Even many of the special interest groups that loudly complained about the content of the questionnaire this spring, were silent on its content during the public comment period 2 years ago. So what changed? What was so different about the 2000 census as compared to 1990? Simply put, we changed. The American people changed. The American people, over the last, decade have become more concerned about their privacy, more concerned about the intrusive nature of government, and more concerned about the intrusive nature of private businesses. A story I've told before highlights these concerns: My wife assisted one of our elderly neighbors complete her census form. This neighbor was adamant that there was certain information like her phone number and her income that she was simply not going to give to the government. Her reasoning was that she couldn't trust the government. She mentioned how certain State governments had sold driver's license information to private businesses, and she felt strongly that her trust in government had been betrayed. She made no real distinction between local, State or Federal Government. I believe many people feel this way, and who could blame them? However, these privacy issues cannot simply be laid at the feet of government either. Businesses from the traditional to the new dot-coms exchange volumes of information on us every day. Recently an Internet toy store that is going out of business was caught trying to sell its customer data base, personal information about parents and children. This sale of personal information was never approved by the consumers. Financial institutions and medical facilities share records about people every day without their permission or knowledge. Census Bureau Director Kenneth Prewitt has said before this subcommittee, and I think he's right, that we are at an impasse, a catch-22, if you will. We are a society that thrives on information. Successful business models build on, ``knowing your customer.'' The more a business knows about its customers, the more efficiently and profitably it can provide goods or services. Yet, at the same time, we, as Americans, love our privacy. We fight for that privacy every day in State legislatures, the U.S. Congress and the courts. So today we examine a piece to that puzzle, the American Community Survey. While I think most of us here today support eliminating the long form, is the American Community Survey the answer? I'm not sure. It would be a disservice to the American people if we were to reflexively approve the American Community Survey in the wake of the long form controversy, without giving it careful consideration to determine if it addresses today's privacy concerns. This and other key concerns must be addressed before any long form commitment from Congress can be made. Is the American Community Survey cost-efficient? Two, should the American Community Survey be a mandatory or voluntary survey? Three, are rural areas getting quality and timely data? Four, will it be implemented in an accurate, efficient and consistent manner? And finally, five, does the American Community Survey address the privacy concerns of the American people? Not until these questions and their components are answered satisfactorily can Congress give its full blessing to the American Community Survey. [The prepared statement of Hon. Dan Miller follows:] [GRAPHIC] [TIFF OMITTED] T3836.001 [GRAPHIC] [TIFF OMITTED] T3836.002 Mr. Miller. I now turn to Mrs. Maloney for her opening statement. Mrs. Maloney. Thank you, Mr. Chairman. Today's hearing on the American Community Survey is both important and timely; important because it represents a new, hopefully improved, way to collect information about the American people. Such information is vital to making informed policy decisions. This survey is timely because we all remember the controversy over the long form which flared a few months ago. First, I want to take a moment to compliment the Census Bureau and Director Prewitt for how well the census is doing so far. I know that there have been press reports in Florida and Illinois about the Census Bureau having to take corrective action where procedures apparently were not adhered to. I am sure that problems like these were not unexpected given the fact that we have had to hire over a half a million people for temporary work on the census. The problems need to be addressed, as I'm sure the Bureau is doing, but they also need to be kept in perspective. If only 1 percent of all of the people hired hadn't followed directions correctly, we would be hearing many more complaints than just the handful we have. The decennial census does two things. It counts the population, and it obtains demographic, housing, social, and economic information by asking one in six American households to fill out a long form. This information is necessary for the proper administration of Federal programs and the distribution of approximately $180 billion Federal dollars per year. It is also vital not only for the Federal Government, but for local governments, health researchers, transportation planning, businesses across the country and a dozen other fields. The census is done once every 10 years, and the information collected by the long form goes out of date after 2 to 3 years. State and local governments, development organizations and other planners are therefore often reluctant to rely on census data at the end of a decade for decisions that are expensive and affect the quality of life of thousands of people. The American Community Survey is intended to provide data communities need every year instead of only once every 10. It will be an ongoing survey that the Census Bureau plans will replace the long form in the 2010 census. When fully implemented, the ACS will provide estimates of demographic, housing, social and economic characteristics every year for all States, as well as for all cities, counties, metropolitan areas, and population groups of 65,000 people or more. For smaller areas, it will take 2 to 5 years to accumulate sufficient data, but after that, information will also be available every year. The questions on the American Community Survey questionnaire are the same as those on the long form; questions which have been around for decades and which every Member of Congress received 2 years before the decennial census. The questions asked by the census represent a balance between the needs of our Nation's communities and the need to keep the time and effort required to complete the form to a minimum. These questions are required by a multitude of Federal statutes, and I look forward to hearing from the authorizing committees, perhaps at a future hearing, about why their committees have required this data to be collected by the Census Bureau. Federal and State funds for schools, employment services, housing assistance, road construction, day care facilities, hospitals, emergency services, programs for seniors, and much more will be distributed based on these data. I think those who criticized the long form either don't know or maybe don't care how essential this information is to solving the problems of the people of our country, and they may have similar criticisms of the American Community Survey. Let's look some at just one of these questions, take the question on plumbing that the talk radio shows seemed to focus on. It may shock some, but there are places in this country where Americans don't have plumbing, in the Colonias in Texas, on Indian reservations, and in isolated rural communities across America. We can't help these places if we don't know where they are. Or let's look at question 17 concerning a person's physical, mental or emotional condition in the last 6 months. Don't we need to know how big a problem this is, how many disabled Americans there are in this country? Small communities need to know where the disabled live in order to provide transportation and other services called for under the Americans With Disabilities Act. In the information age we need reliable information in order to make good decisions for this Nation. Without good data you cannot administer the laws of this country fairly. I for one will continue to do all I can to make sure that the Census Bureau has the capabilities to provide the Congress, and this Nation, with the ability to provide all of us with the high-quality data needed by the public and its elected representatives to make informed public policy decisions. Thank you, Mr. Chairman. I thank my colleagues that have come to report to us. [The prepared statement of Hon. Carolyn B. Maloney follows:] [GRAPHIC] [TIFF OMITTED] T3836.003 [GRAPHIC] [TIFF OMITTED] T3836.004 Mr. Miller. Mr. Davis, do you have an opening statement? Mr. Davis of Illinois. Thank you very much, Mr. Chairman, and I will make a brief comment. Let me, first of all, commend you and Ranking Member Maloney for the manner in which you have both gone about trying to make sure that this issue remains in front of the American public and to make sure that we do, in fact, not only get the best and most accurate count that we can for the census that is under way, but also that we have understanding as we move into the future. I also want to thank Representatives Collins and Emerson for their willingness to come and share with us this morning and to voice concerns about the issue. I also commend Dr. Prewitt for sharing his findings with us in relationship to what we have been able to do up to this point with the 2000 census. I look forward to hearing the proposed missions to be accomplished by the American Community Survey. Will it be as efficient as the long form questionnaire in maintaining the most accurate count possible and in securing the information that is needed? Especially as we talk about all of the different aspects of American life, I find it somewhat incomprehensible that at the same time we would talk about denying the information or not generating the information that is needed in order to make rational, logical and informed decisions. Of course that is the beauty of democracy. We all have a right to do whatever it is that we want to do, even if it is wrong. Does it ask the questions necessary to determine where the most financial assistance should be targeted? In addition, I look forward to our expert witnesses as they express concerns relating to the use of the American Community Survey. I am excited about the work that the Census Bureau has done. I have really enjoyed serving on this committee and have enjoyed the interaction with the leadership as well as the Bureau. I look forward to a good count for this year, but more importantly, I look forward as we move ahead to really having the kind of information and the kind of data that the American people need. I also want to thank my intern Detris Brown, who prepared this statement. This was her first one, and she has demonstrated a serious grasp of the issue. Detris, thank you very much. I yield back the balance of my time, Mr. Chairman. Mr. Miller. Thank you, Mr. Davis. We are pleased to have a couple of our colleagues today. The long form controversy created concern by many of our colleagues. I think Mrs. Emerson is concerned about some of the rural concerns. Mr. Collins, is it all right if Mrs. Emerson goes first? We are both at a markup across the hall. Mrs. Emerson, would you like to go first? STATEMENT OF HON. JO ANN EMERSON, A REPRESENTATIVE IN CONGRESS FROM THE STATE OF MISSOURI Mrs. Emerson. Thank you, Mr. Chairman, Mrs. Maloney and Mr. Davis. I speak as the cochairperson of the rural caucus. As you know, this is really a period of rapid change for rural communities, for rural governments. Economic and demographic shifts and changing relationships between Federal, State and local governments are really having major impacts upon rural communities. As Congress continues to devolve decisionmaking and resources to the local level, it has become critical that the local leaders in the public and private sector have accurate information upon which to make critical policy choices which will impact the future well-being of their communities. Greater responsibility than ever is being transferred to our local units of government, not only by this Congress and the administration, but also by our State governments. As this important task of moving government closer to the people unfolds, there remains an inherent potential that rural places and people will be disadvantaged. Rural decisionmakers have significantly much less access to the effective decision support tools necessary to make informed public choices. There are many reasons for this rural disadvantage, but because accurate and timely data is seldom available for rural communities and smaller rural jurisdictions, and because these entities have limited budgets, and in some cases severely limited budgets, and are often led by part-time decisionmakers, timely, empirically based assessments of policy alternatives are seldom available. I would say based on the communities in my district, which is composed of 26 counties and very remote areas in some instances, that really probably is an understatement. Unfortunately, our rural citizens are increasingly disadvantaged in this regard. Urban and suburban jurisdictions with full-time research staffs are at a significant advantage in competing for the Federal and State resources available to support their communities. Rural communities must wait for the decennial census for the locally based data upon which to base their decision. Often this data is already out of date when the census is published, and then this information remains the only available data source for most rural communities for the next 10 years. Obviously in this era of decentralized community- based decisionmaking, these communities are in dire need of more accurate and timely information upon which to base future decisionmaking. Let me give a recent snapshot of rural America which really does reinforce this reality. During the 1990's, a significant rural population rebound occurred, totally reversing the rural out-migrations of the 1980's. Three-fourths of our country's 2,350 rural counties grew in population between 1990 and 1997; seven-eighths of these communities derived some or all of their population increase from in-migration of metropolitan residents. Of the rural population increase of 800,000 between 1995 and 1997, 400,000 came from metropolitan areas, and 100,000 came from immigrations. Due to these shifts, many growth counties are experiencing unique new diversity in ethnic, racial and cultural composition with their attendant challenges and conflicts. From 1990 to 1998, metropolitan America experienced domestic out-migration of over 10 percent. By contrast, over 50 percent of nonmetropolitan America was domestic in migration. Only nine States in the Nation had a net nonmetropolitan out- migration. Just taking a brief snapshot of my own State reinforces this reality. In this decade over the last 10 years, 145,000 more people moved into Missouri than moved away. Nearly all of these folks moved to places other than Kansas City and St. Louis. In fact, over 80 percent of the population growth in our 92 nonmetro counties resulted from in-migrations. Using 1990 census data entirely misses these new rural realities. Given these changes, it is evident that any policy, program or resource allocation decision based upon a 1990 rural America would have completely missed the mark. Because the face of rural America is changing so quickly, and these new rural realities are often so misunderstood at all levels of governance, the importance of accurate rural data is critical. A number of issues continue to be raised regarding the fairness of the American Community Survey for rural constituencies. The Congressional Rural Caucus is particularly concerned with these issues, recognizing there are unique challenges which must be addressed to effectively implement the American Community Survey in rural areas. However, I am also aware that an ongoing discussion with rural social scientists and rural community organizations has resulted in specific attention to these issues. I would urge the committee to remain vigilant to assure that these issues are adequately addressed, and I urge the census and ACS staff to continue to be sensitive to these challenges. I would also ask that particular attention be paid to the testimony of Chuck Fluharty, who is the director of Rural Policy Research Institute for the subcommittee. He will be testifying, I think, on panel three. Simply put, having accurate and timely data is critical to assuring our rural communities do not continue to be significantly disadvantaged in our Federal statistical processes, however inadvertent that harm may be. The Congressional Rural Caucus would be pleased to work with the subcommittee and the full committee and Congress to ensure that this is achieved, and I thank you very much. Mr. Miller. Thank you, Mrs. Emerson. [The prepared statement of Hon. Jo Ann Emerson follows:] [GRAPHIC] [TIFF OMITTED] T3836.005 [GRAPHIC] [TIFF OMITTED] T3836.006 Mr. Miller. We will go ahead and ask questions. Mrs. Emerson. Thank you. Mr. Miller. I am glad that you are here today because I think so much of our focus has been on urban concerns and issues with the decennial. One question that I have, the Agriculture Department does a certain amount of surveying of its own. I don't know very much about it, whether or not there is a way to make use or contracting with the Agriculture Department because they do a farm survey. Mrs. Emerson. They do a farm survey, but keep in mind even though I would have 26 counties, the majority of which are agricultural in nature, by just looking at farms, because it depends on farm size, etc., you really don't capture the number of people and a lot of the other challenges that we face in rural America. The fact of the matter is when you get your decennial census information 10, 12, 11 years late as it somehow applies to rural America, we do get missed in any kind of allocation of resources. For example, Medicare reimbursements to rural hospitals are based on such inaccurate data that we are losing hospitals right and left. When you look at education funding, we get the short shrift every time. There has to be some mechanism where we can measure what is going on in rural America in the in-between times. I know that there is concern about privacy issues, but using a 3-year rolling average, which, in fact, the ACS does, would, in fact, help us simply because of the unique nature and the challenges we face in rural America. If you really look at rural America, it is very similar to the challenges urban America faces, and a quarter of the population lives in rural America. Oftentimes we simply get missed in the overall assessment and don't have the resources to lobby, if you will, for those things that we need as effectively as others do in more affluent areas. Mr. Miller. Thank you. Mrs. Maloney. Thank you very much, Jo Ann, for your testimony. In reviewing the testimony of Representative Collins, one of the proposals is making only the short form mandatory, and I want to know what your response is to that suggestion and whether or not you see any negative effect on rural communities. Mrs. Emerson. As many calls and letters as we got, people complaining about the long form, and it was onerous in some respects, because we have difficulties in obtaining some of that information in rural America, I suppose I would say some combination would be in order. And I realize that there is a financial portion of that, but in order that we be able to have the resources we need to not only keep rural America viable, but to let it flourish as it once did, I think the more information that we have, as Mr. Davis said, the better off we are. Mr. Miller. Mr. Davis, do you have any questions? Mr. Davis of Illinois. Well, actually she answered my question just at that moment. What I was going to ask, Representative Emerson, as an expert on rural America, on needs, quality of life, indicators, as one who really understands and who is there, without the information how difficult would it be to really understand what was going on in these areas of the country? Mrs. Emerson. I think given the spread-out nature of rural America, and my district is small compared to those in the Western States, if you have to drive 50 miles to a grocery store, that is not too far. It is difficult to accumulate the information. I am not an expert in statistics, and perhaps Mr. Fluharty can help answer the more technical part of that, but I think it is extremely difficult to obtain the data from rural America, and I think if we do not attempt to find some means to interview, assemble information in the in-between years, that we in rural America will not be able to compete on a level playing field with the cities and suburbs. Mr. Davis of Illinois. Thank you. Mrs. Emerson. Thank you, and if you will excuse me, I will run to my markup. Mr. Miller. Mr. Collins. STATEMENT OF HON. MAC COLLINS, A REPRESENTATIVE IN CONGRESS FROM THE STATE OF GEORGIA Mr. Collins. Thank you, Mr. Chairman. Mr. Chairman and members of the committee, I appreciate the fact that you allow me to come and speak to you this morning about a bill, H.R. 4188, entitled the Common Sense Census Enforcement Act of 2000. The bill was actually inspired by constituents, many of whom contacted me to complain that this year's census was too intrusive. Their complaints centered on the long form, which they said took too long to complete and asked questions which were too personal. They wanted to know why the Census Bureau needed to know about their plumbing or about the size of their paychecks. Many of them were also worried about the fines for those who either lie or refuse to answer the questions on the form. Mr. Chairman, we know these are legitimate questions. Mrs. Maloney, they are legitimate questions, and I understand the reason for asking those questions. The Census Bureau looks at the census as a once-in-a-decade chance to gather a great deal of information that may be very useful. Nonetheless, we have to ask ourselves if this additional information is absolutely necessary to fulfill the constitutional purpose of the census, which is to enumerate the population for the purpose of redistricting. Further, the complexity of the census long form and the threat of the financial penalties is actually counterproductive. The long form questions tend to intimidate Americans and lead to a lower participation in the census. This leads to an undercounting of the poor, members of minority communities, children, and those living in rural areas and inner cities. In other words, it intimidates all segments of the American population. And it is important that we have as accurate a count as we can have. I know, and we all know, the Census Bureau has a very, very difficult job, but as every member of this committee is acutely aware, the census is constitutionally mandated for the purpose of apportioning Federal legislative districts, and the population information gathered is also used in drawing State legislative district lines. The Constitution requires the Federal Government to conduct the census, and Federal law also requires that the residents answer the census completely and truthfully. Failure to answer any questions can result in fines up to $100. Furthermore, if one intentionally provides inaccurate information in response to the census, the law provides for fines up to $500. These penalties are understandable with regard to the questions directly related to apportionment, in light of its central importance to our constitutional system. I do, however, question the appropriateness of imposing such penalties for refusal to answer questions unrelated to apportionment. Congress should eliminate the penalties for failure to answer census questions unrelated to apportionment. To accomplish this I have introduced H.R. 4188, which would eliminate the fine for failure to answer census 2000 questions unrelated to apportionment. By taking this action, Congress can limit the intrusive nature of the census while still providing the government with the basic information necessary to administer our Republic. H.R. 4188 does not prevent the Census Bureau from collecting information. It does not stop the Census Bureau from collecting information through other surveys. It only prevents the levying of penalties on those Americans who choose not to participate. I must say we had a lot of assistance from the regional census office in Atlanta when we had these questions asked us, am I going to be penalized, and am I going to be fined if I refuse to answer or inaccurately answer. According to the law, the answer is yes. But I will say that the regional office and the D.C. office both were very cooperative in saying, ``we don't intend to levy any fine on anyone,'' but that is not what the law says. Someone else could come back later and say, ``well, you didn't answer, and the law is this,'' and that is the reason that we are so encouraged and want to see some changes in the penalty portion of the nonapportionment portions of the census. In closing, I share the belief of many Georgians who find it inappropriate for the Federal Government to coerce citizens to provide personal information by packaging non-apportionment- related questions with the constitutionally required and legally enforceable apportionment questions. In the future the information should be collected separately. There has been one proposal for dividing the Census Bureau into two divisions--one which conducts a postcard census for reapportionment and another which handles surveys and polls. This proposal has gained support inside and outside the Congress. While this proposal should be examined, it should be made clear that no penalty will be applied to those who refuse to answer questions unrelated to apportionment. I urge my colleagues to join me in supporting H.R. 4188. Thank you, Mr. Chairman. Mr. Miller. Thank you, Mr. Collins. [The prepared statement of Hon. Mac Collins follows:] [GRAPHIC] [TIFF OMITTED] T3836.007 [GRAPHIC] [TIFF OMITTED] T3836.008 Mr. Miller. Your area is not as rural as Ms. Emerson, but it is relatively suburban rural? Mr. Collins. The district at the northern end is just outside of Atlanta, which is a bedroom community, and runs through a rural area to Columbus, which is a rather large city in relation to other cities in Georgia. So we have a mix. We are very proud of that, and we want everyone counted. I think it is important that everyone should be counted so when it comes to redistricting, you have a number that is as accurate as possible so you can have fair reapportionment so people are represented by people within their communities. Mr. Miller. The long form is a sample going to one out of six on average, but in rural America it can be one out of two. In Mr. Davis' or Mrs. Maloney's district, it may be one out of seven or eight. We have talked about this before. They have made major improvements on the long form since 1990. They have focus-grouped the questions and reduced the questions; and as you know, the plan or the discussion for 2010 is to go to--we will talk more about a postcard, but a postcard type of census concept. One of the questions that you raise on the fine, the fine hasn't been used since 1960. And if we are going to have a fine, and I would think that Henry Hyde or John Conyers would agree, if you have a fine and you don't enforce it, what good is it? It is a legitimate question. It is like jaywalking; if you don't enforce it, why have the fine? Certainly on the proposed plan, which is just six questions or so for the short form, which is a Constitution requirement, but how do we ensure the data is accurate, and we have to weigh that. As we go through this process preparing for the 2010 census, and that is what this hearing is about, you bring up some valid points. Thank you. Mrs. Maloney. Mrs. Maloney. Thank you for coming and showing the concern that you have and the concern expressed by your constituents. One thing that is so very special about the census is that it is a great civic ceremony that is mandatory. Literally every American is called upon to participate in the census, and that is one of the responsibilities that we have to be residents in this country. And if you went forward with your proposal, you would basically make the census a voluntary activity, particularly for the long form. I really can't think of any other activity we have in the country that absolutely everyone--it is mandatory, it is required that we do this. Education, you can educate in your home. There is flexibility. But the census is an important--it is important because of the information that we get. It is important for planning for the future, for the country in general and our communities. One of the things that came out of the research from the Census Bureau is that by designating a survey as mandatory on the envelope, that increases the mail-back response rate dramatically. When people see mandatory, they fill that form out. It not only benefits them individually, but it benefits their community. To the extent that all of the residents in your district are counted, then it is more fair in the distribution formulas for the $180 billion. One of the things that we would have to do if it was not mandatory and the mail-back response rate fell, we would have to spend more money going out with the other efforts of telephone calls, knocking on doors, so in effect it would raise up the cost of the census. You are Ways and Means, but we have an appropriator sitting next to us who has to vote on those appropriations for all of our expenditures, including the census. I guess all of us are concerned about keeping costs down for activities, so there is a cost factor to it. I wonder how you measure that. By just stamping ``mandatory,'' the mail-back response rate goes up, and that saves money because then you don't have to pay for the field resources to go out. One of the things that we are proud of in the census, for the first time in 30 years we reversed the decline of the mail- back response rate. It literally went up, which helps with the accuracy and literally saves hundreds of millions of dollars. This change could literally cost--we will hear from the experts later, but it could cost hundreds of millions of dollars, I would think, in implementing the accurate count of the census. Mr. Collins. Well, I think you are asking a question that there are really no known answers to. As far as being on the Ways and Means, without Ways and Means and the activities of our committee, there would be no appropriations. Mrs. Maloney. That is true. We know the hierarchy around here. It is a lot higher than the census committee. Mr. Collins. At home, my wife and I have a residence, and then we have this little lake cottage, and then we have the apartment here in Washington, and we also very frequently occupy two seats on Delta Airlines. I received a census form, more than one, at each one of those locations with the exception of the seats on Delta Airlines, and I kept looking for those. But even though we, my wife and I, filled out--and we fortunately got--the short form, we would have filled out the long form. I know that there are a lot of Americans who had no problem doing it. Even though we filled out the short form and submitted it, there was an inaccuracy on the address that came to us. My wife changed that in the proper place, but still it didn't register, and we received phone calls at each location. We had noticed on the doors of each location that people had been from the census to see us to get the final form filled out. And then, too, about 2 months prior to the last day of the deadline, I received a call from staff saying that the Washington office had called my office, and I had failed to answer the census. They wanted to make sure that as a Member of Congress I did fill this thing out, you know. I did, too, because I didn't want to be subject to a fine. I have to look after my finances, too, you know. But we had filled it out, but it had been in some way through the system not registered. But how many people are there who just simply say, well, I have this thing here, this long form. I don't want to fool with this thing. If I don't fill it out, and I fill out the important portion of it, but if I send it back without any other information, then there is a question that you failed to answer on the form. What actually in law is a failure to answer the question. There is a fine for inaccurate or failure. The information is important. As you suggested with the plumbing in many areas of the country, there are still areas that have problems when it comes to drinking water or even disposing of sewage. But I wonder if we are getting the response from the census that we really want for apportionment, and if we are actually getting the response to the questions that we do need in order to be able to assist people. I am not in any way trying to deny any of those questions from being answered. If we are going to tell people through the Census Bureau or through the regional office or the D.C. office, hey, look, we are not going to fine you if you didn't do this, but please fill it out and send it in, why not make the law conform with what you are telling people? We are leaving in place the fines that deal with apportionment so we have fair and equal representation across this country. Mrs. Maloney. Thank you. Mr. Miller. Mr. Davis, any questions? Mr. Davis of Illinois. Yes. Thank you very much, Mr. Chairman. Representative Collins, I listened very intently to your testimony, and I have a great deal of appreciation for what you said in terms of the concepts of freedom, the individuality of citizenship, the rights of people not to be violated unduly and unnecessarily, and as I thought about that, I also thought about the concepts of citizenship responsibility. It seems to me that part of the responsibility of citizenship in a free and democratic society is for each member of that society to contribute as significantly as he or she can to the decisionmaking, the total well-being of not only themselves individually, but the group as a whole. And so I thought that there is a delicate balance here in terms of how do you satisfy both roles, your social role obligation as well as your responsibility. My question is are you aware of any instances where any person has ever been convicted or penalized in any way for having refused to give the information that was being asked for? Mr. Collins. You have just made the case of why we need to take the fines off this particular provision of law. No. The Census Bureau says we are not going to do this. But, you know, they had a little--I am not an attorney, what's the word-- caveat to it. It says that, but that is what they are telling us. Someone else may have a different idea because it is the law. And is it really freedom to answer the questions if you can possibly be monetarily fined for not answering? Mr. Davis of Illinois. I think you make somewhat of a point, but I have spent a lot of time lately with my father, who is 88 years old. I have a lot of octogenarians in my family right now. My father is 88, and my mother has a brother who is 96, and my father has a brother who is 94, and he has another brother who is 86. I grew up on a farm. We had a very gentle horse, and the horse just wouldn't go anyplace. He would just hang around the house and the yard. My father kept him in a little pasture. I said, since the horse isn't going anywhere, why do you put him in the pasture? He said, I know that he is not going anyplace. I just don't want him to be tempted to wander away if there are no restrictions, if there are no boundaries, and I think this requirement is kind of like that. I don't think that it is designed in any way to be punitive or even to be harmful, but it is just to guard against individuals wandering off and not contributing their fair share of information. There is not much more than information that is being generated, but I really appreciate the concept of the legislation that you are projecting. I just think citizens aren't giving enough right now in this country, and I think to the greatness of this Nation all of us owe a great deal, and certainly one of the things that a citizen can, in fact, do if nothing else is share that kind of information with the rest of us. Mr. Collins. I appreciate the fact that you spent a lot of your time with your father. Both of my parents are deceased, and my dad lived to be 86. When he was about 84, I carried him to the doctor 1 day. As we left the doctor's office, I kind of stepped back to speak to the doctor. When I came out and we got in the car, my dad asked me, he said, what did that doctor say? I said, that doctor said I was going to live to be 100, and that you were going to be one of my pallbearers. He said, any guarantee on that? My dad, like yours, was one of the smartest people I have ever known even though he only had a third-grade education, and he also had a lot of old sayings, and one related to that old horse that you wanted to keep in the pasture and not attempting to leave without restraints, and that is that you could lead that horse to water, but you couldn't make him drink. Let's don't intimidate people by trying to get them to voluntarily give us information and going and asking them for information. The Census Bureau visits numbers and numbers of people all across this country, and they can probably get a better, accurate--more accurate accounting of information upon those visits than the possibility of never receiving any of the information back because of the intimidation that is there with these possibilities of fines, in addition to the fact that I think one of the reasons we have had such a good response to the census is that in Georgia, the Governor of Georgia was on television with 30-second spots time and time again, and a lot of other people, encouraging people to answer the census form. It is important. We need this information so that we can accurately count people so that we can as fairly as possible apportion, and so that we have information that helps cities and counties with funding that comes from the Federal Government. I think that is probably going on all across the country, and that had a lot to do with having an increase in count. We told people that we need the information to do a better job for the representation of the people. Thank you. Mr. Miller. Mr. Collins, thank you for being with us today. Mr. Collins. Thank you, and I hope that you will consider H.R. 4188. Mr. Miller. Dr. Prewitt, Mr. Spotila, and Katherine Wallman, would you come forward, and I'll swear you in. [Witnesses sworn.] Mr. Miller. Let the record show that all three answered in the affirmative. This is kind of the beginning of a process that will continue for the next year or so as we go through this process preparing for the 2010 census and also preparing for the needs of our society. We look forward to your testimony and an opportunity to ask questions. Mr. Spotila, would you go first. STATEMENT OF JOHN SPOTILA, ADMINISTRATOR, OFFICE OF INFORMATION AND REGULATORY AFFAIRS, OFFICE OF MANAGEMENT AND BUDGET, ACCOMPANIED BY KATHERINE WALLMAN, CHIEF STATISTICIAN FOR THE UNITED STATES; AND KENNETH PREWITT, DIRECTOR, BUREAU OF THE CENSUS Mr. Spotila. Good morning, Mr. Chairman and members of the committee. Thank you for inviting me here to discuss how the American Community Survey can help the American people by improving the quality and timeliness of the data that our Federal statistical system provides. I am accompanied by Katherine Wallman, who serves as chief statistician for the U.S. Government. The ACS can provide us with key statistical information on a much more current basis. The Census Bureau has piloted the ACS since 1996. It hopes to implement the ACS in every county in the United States, starting in 2003. By 2010, the ACS may replace the census long form and greatly simplify the census process so that the decennial census can focus solely on counting our population. We see this as a very promising and positive initiative. We know that better information can help agencies make better decisions about how well the government is working, whether new services are needed and whether existing programs are still necessary. Better information can also improve decisions made by businesses, local organizations and individual citizens. The American people need timely, accurate information on a wide range of topics affecting their daily lives and business activities. Each year thousands of Americans cooperate with Federal requests for data because they understand the value and importance of their participation in these surveys. While information plays a critical role in good government, we recognize that the collection of that information imposes a cost on the public. The Paperwork Reduction Act of 1995 [PRA], emphasizes that agencies must strike a balance. They should collect the right information to meet their responsibilities to the public, but should not require information that is unnecessary. In carrying out our responsibilities under the PRA, OMB reviews and approves agency requests for information. We seek to ensure that the information is necessary, that duplication of effort is minimized, and that the collection methods used are as simple and fast for respondents as possible. Most of the information needs of the Federal Government flow from statutes passed by Congress. Decennial census data that historically have been collected on the long form are among key sources of this critical information. Many information requirements help implement legislatively based programs, including data used in formulas to allocate nearly $200 billion annually in Federal funds. A large percentage of funding formulas distribute moneys to States and localities. If fully implemented, the ACS would provide, beginning in July 2004, far more current data for use in these formulas. Because conditions in some communities can change rapidly, having current data is critical to identifying the most deserving communities. Although we believe that the ACS can generate better, more current data in a cost-effective manner, we recognize that we must still implement it wisely and well if it is to fulfill its promise. The Census Bureau has the lead on this endeavor. Its leaders and staff will work closely with Congress, OMB, other Federal agencies and data users to ensure that the data collected by the ACS are needed and that the survey design methods are both efficient and effective. As Dr. Prewitt is noting today, OMB recently launched an interagency committee to broaden the dialog on this subject. Over the next 3 years, this committee will examine a variety of issues relating to the ACS, including a comprehensive review of the questionnaire content. The ACS test instrument currently in use will be the starting point for this review. As with the census 2000 long form, every question on the ACS test instrument is required by Federal law to manage or evaluate government programs. The committee will examine these statutory requirements, determine whether the ACS is the best vehicle for meeting them in the years to come, and consider new data needs that may be best met by the ACS. It will also work with the Census Bureau to develop approaches for considering longer-term ACS content issues. These approaches will be responsive to congressional concerns and will address the needs of Federal agencies that rely on the statistical system. OMB ultimately will review and approve the proposed 2003 ACS instrument through the standard PRA clearance process. In doing so, we will carefully consider the recommendations of the interagency committee and ensure that the ACS collects the right information with minimum burden on survey respondents. We welcome the committee's participation and interest in all of these matters. We look forward to working with you closely and to sharing the results of our collective efforts. Thank you. Mr. Miller. Thank you. [The prepared statement of Mr. Spotila follows:] [GRAPHIC] [TIFF OMITTED] T3836.009 [GRAPHIC] [TIFF OMITTED] T3836.010 [GRAPHIC] [TIFF OMITTED] T3836.011 [GRAPHIC] [TIFF OMITTED] T3836.012 Mr. Miller. Director Prewitt. Mr. Prewitt. Mr. Chairman, Mrs. Maloney and Mr. Davis. In my few minutes I will offer three observations and then identify four issue areas. My three observations emphasize the importance of the American Community Survey in its own terms, note the substantial and significant positive consequences of the ACS for decennial census, and, third, report that the ACS is feasible; that it is meeting its operational tests. First, its importance. The American Community Survey is the single most important innovation in Federal household statistics since the middle of the last century when sampling methods were first introduced. This innovation is timely. The country will be challenged to sustain its economic well-being and enhance its capacity to democratically govern itself under the new conditions brought about by, among other things, dynamic demographic changes resulting from immigration and geographic mobility, the emergence of the knowledge economy, and the changing balance of responsibilities between the Federal and local government and between the public and private sector. To navigate in these new social, economic and political circumstances with the decennial census long form data is like driving in a country we have never visited where the directional signs are in a language we do not read and with an outdated map, guessing as best we can where we are and where we are going. The ACS is an up-to-date map allowing us to navigate in a language we understand; that is, facts. Mr. Chairman, if we do not launch the ACS this decade, we will in the next or the one thereafter. The country will get increasingly impatient with a data collection strategy initiated in the middle of the 20th century to deal with the conditions of the 21st century. My second observation: The ACS will revolutionize the way we take the decennial census and for the better. With good reason, the Congress has been concerned that the long form is a drag on the decennial census, that it introduces a complication in carrying out the basic constitutional purpose of the census. The best solution is to radically simplify the census by eliminating the long form. Obviously we cannot eliminate the long form without an alternative method of collecting detailed population and housing characteristics. Congress must have the data it has mandated to run the programs it has written into law. Some 400 pieces of legislation now directly or indirectly require long form data. Indeed we are aware of at least 35 new bills requiring long form data that have been introduced in the 106th Congress alone. With the long form task shifted to the ACS, a much simpler decennial census comes into view. The Census Bureau has started planning how to reengineer many aspects of the decennial census to achieve substantial cost savings in every aspect of the decennial--address listing, form design, printing, labeling, questionnaire delivery, enumerator training, nonresponse followup, data capture, data editing, and data tabulation. The ACS, for example, can lead to a better master address file because we can continuously update the address list in an ongoing partnership with local officials. It will improve 2010 operations by allowing us to use the highly trained and seasoned ACS field staff as a cadre of key supervisory staff for the 2010 census. A simplified census 2010 can make extensive use of the Internet, further reducing costs and perhaps taking more advantage of administrative records to improve coverage. The ACS can further improve coverage in the decennial head count by providing current rather than decade- old data to target areas where special field procedures are called for. In short, the ACS offers multiple opportunities for substantial cost savings and improved coverage in 2010. The Census Bureau will need to conduct operational research if it is to do this reengineering in a timely fashion, and the clock on census 2010 is already ticking. As will be made clear by other witnesses today, the ACS can also produce dramatic improvement in the entire infrastructure of the Federal statistical system. It is now costly and time- consuming to create samples for new surveys or to adjust samples of existing surveys to provide data to meet the needs of policymakers in unforeseen areas. My third observation, the ACS is feasible. With strong support from the Congress, the Census Bureau has been field- testing the ACS in selected sites around the country, and we are conducting an additional and critical test of the ACS in the decennial environment. All signals are showing green. The clearest, strongest indicator that the ACS is feasible is its successful household response rate, which at better than 97 percent is higher than that of any other demographic survey conducted by the Census Bureau. Also, we have maintained a demanding schedule. Next week, for example, we will release those 1999 site data scheduled for this month, which is exactly on the schedule we have set for ourselves and have pledged for the actual ACS when it is fielded. I turn now to four issues that will have to be addressed as the ACS moves forward. This is not an exhaustive listing, but underscores some key areas for further attention. I start with what is perhaps the most important: constructing an optimal working relationship with the Congress. The ACS is part of the decennial, and as such must be planned and executed in close collaboration with the Congress. How best to accomplish this requires more time than we can devote today. To anticipate, just one of the many issues is how to balance concerns about minimizing respondent burden with requests for additional information. Just as some in Congress want minimal data collected, others have expressed the need for additional questions. The only new question added to census 2000, that concerning grandparents as caregivers, was required as part of the welfare reform legislation. The Census Bureau determined that the data should be collected on a sample basis, although some Members supported asking it of all households. And, there was a sense- of-the-Senate resolution just a few months ago expressing strong concern that the marital status question had been moved from the short to the long form in census 2000. So we must always look for ways to balance these competing pressures, and we look forward to working with the Congress to do that. There are, of course, other congressional oversight issues. My point here is that this is an issue to be addressed. Second, the Census Bureau will need to establish strong, ongoing relationships with a large number of key stakeholders. The interagency committee launched by the OMB is a very timely and welcome initiative. We will want to maintain the community partnership program, which has proven to be so very effective in the early ACS test sites, as it has been for the census 2000. More specifically, there can be no successful ACS unless local and regional governments are on board. We are confident that they will be. And we will want a working relationship with the private sector so that any number of reciprocal benefits can be realized, especially with the survey and data dissemination industries. Finally, the Census Bureau will redesign its advisory committee structure that has served it so well in census 2000 in order to draw upon wide-ranging expertise in the conduct of the ACS. Third, timing. For a number of operational and planning reasons, it is critical that we maintain the schedule for ACS recommended by the Census Bureau. Initiating the ACS in 2003 allows us to be confident about key design issues for Census 2010 in advance of when those decisions have to be made. This in turn will allow for a high level of congressional comfort about removing the long form from the decennial operation. Fourth and finally, the privacy issue. Long form questions are not less, nor more, intrusive because they are asked in the ACS rather than the decennial environment, but the environments are wholly different. It matters whether 20 million housing units are asked long form questions in one intense timeframe or whether those questions are asked in a series of monthly surveys. One big difference. Instead of having a large army of temporary enumerators, the ACS field interviewers will be highly trained permanent staff who will be better prepared to deal with the public's questions about the questions. As I said at an earlier hearing, the Census Bureau's experience in conducting hundreds of surveys, some much more demanding than the long form questionnaire, gives us confidence that once the public understands that their answers are protected by law and that every question asked serves an important purpose, they do perform their civic duty to respond. In fact, our experience thus far with the ACS has been that our trained interviewers have achieved good cooperation with few complaints from the public. I believe it is hard to sustain the argument that government data collection is an invasion of privacy when there are such strong protections of the data, when they are used only for statistical purposes, not for regulation or law enforcement, and when each questionnaire item is linked to a program that the people's representatives have enacted. I take note that some Members of Congress believe that long-form-type questions should not be asked, period. That is an issue for Congress to resolve. It is not for the Census Bureau to decide what kind of society we should be or even whether we should have timely and relevant data to make that vision possible. It is our role, however, to inform the Congress about the most efficient, effective, and modern ways to collect data, once the administration and the Congress have determined which data are necessary. That is what we are doing today by presenting our plans for the ACS. Mr. Chairman, in conclusion, I reemphasize my initial three observations: The American Community Survey is the single most important innovation in Federal household statistics in more than a half century and positions the country well for the century we are just entering. With the ACS in place, the decennial census can concentrate on its core constitutional task, population counts for apportionment and redistricting, and do so less expensively and more efficiently. The ACS works. Mr. Miller. Thank you. [The prepared statement of Mr. Prewitt follows:] [GRAPHIC] [TIFF OMITTED] T3836.013 [GRAPHIC] [TIFF OMITTED] T3836.014 [GRAPHIC] [TIFF OMITTED] T3836.015 [GRAPHIC] [TIFF OMITTED] T3836.016 Mr. Miller. This is a significant issue we are undertaking, and I think this next Congress will be addressing it in even more detail, and I am glad we have an opportunity toward the end of the 106th Congress to begin the congressional oversight part of that responsibility. Let me start with--I have a number of different questions, and a lot of it is getting a better understanding and justification for some things. When we talk about sample size and response rates, my understanding is that it is 3 billion in a year. First of all, is--that is a stratified sample. For large population areas, will we get usable information? Mr. Prewitt. For any unit of the country or population group of 65,000 or greater, we will have highly reliable estimates of its characteristics after 1 year. Mr. Miller. So this rolling average we would combine year after year? Mr. Prewitt. So by combining data for 2 to 3 years, we will be down to 20,000. By combining all 5 years, we will be down below 15,000. Very small communities. Mr. Miller. What is the impact of having a rolling type of average? Mr. Prewitt. It improves the statistics in some respects because you smooth out some irregularities in data collection. Obviously, most questions will work very well by averaging across 3 to 5 years. Some questions will have to be examined. That is what the interagency committee will be doing. In fact, do you want us to go back and forth? Mr. Miller. Please. And describe the interagency working group. Ms. Wallman. We established an interagency committee officially about 2 weeks ago to help us in the process of examining the content for the American Community Survey. We had a similar process in census 2000, bringing together the 20 agencies that use census data to implement legislation, as Dr. Prewitt described in his statement to you. Using that same model, we have brought together not only a set of statistical agencies--the Bureau of Labor Statistics, the National Centers for Health Statistics and Education Statistics, and so on--but also several agencies in agriculture, Health and Human Services and other Departments that administer Federal programs using the kinds of information that the American Community Survey will provide. One of the things that we need to do is repeat the process essentially--and perhaps enrich that process--that we carried out in looking at the census 2000 content for the long form. We need to review each of those data items against its statutory bases to look in particular at things such as, given the ACS environment, would we need to collect that particular data element every single time; or given the annual nature of the American Community Survey, could there be some adjustments in that. That is a process that we began in our meeting on July 13 and look forward to carrying out further. We will have a lot of dialog not only with that committee, butt with other users outside, such as the Congress, as we proceed. Mr. Miller. Will Congress have any role in this? Mr. Spotila. The intention is to work closely with the Congress. Although it is an interagency committee, I think the intention is to communicate regularly, to be responsive to your interests and to other Members who would like to participate, and to try to work for a meaningful, cooperative approach. Mr. Miller. The sample size issue, how do we come up with that number? We are talking about a 30 million number over a 10-year period. Does it need to be that large? Could it be reduced in future years? I want to ask some questions about cost when we have time. Mr. Prewitt. The sample is designed so that after 5 years we will have the same degree of statistical resolution as we now have for the long form data, and then by averaging the 5 years, we have that every year. So we actually don't think of it as a 30 million sample over 10 years, but a 15 million sample over 5 years because it is 5 years' worth of data which will allow us to bring it down to the current geographic refinement that we now have for long form data. So that is how the sample is designed. Obviously if you cut the sample in half, then you would get to the degree of resolution you want after 10 years, but you are now averaging across 10 years, and so the data simply get less stable. We believe if we are going to do this, it makes more sense to make sure that we hit that point of statistical resolution after a 5-year cycle. Mr. Miller. This data is usable at the census tract level? Mr. Prewitt. That's correct. That is roughly the degree of resolution. It is roughly 15,000 people, any community of 15,000, or fewer, even down to the census tract level, which is what the long form takes us to now. Mr. Miller. Is that needed? Is there any reason---- Mr. Prewitt. Well---- Mr. Miller. I am asking you to justify the sample size. Mr. Prewitt. The first witness you had put it on the table, the issue of getting good rural data. The issue is not rural data, it is geographic data of low population density. You can have very spread-out suburbs, not just rural areas. You can have inner cities, because of their characteristics; they are more commercial than residential where not very many people live. The real issue is what is the population density of an area. If we want geographically refined data, then it takes this kind of sample size. Congress could decide to deal with half the sample size and only bring it down to a population of 50,000 or 45,000, but I would ask you to talk to your Rural Caucus representative. Mr. Miller. I want to ask some questions about all of the questions. Do we need them in that detail at the tract level section? Obviously you do surveys that don't go to the tract level. Do we really need this large of a sample size? Ms. Wallman. If we are to provide the level of resolution that we are talking about here, we need this large a sample size. If we want to talk about alternatives that would give us less detailed information for these lower levels of population density, then there are alternatives that could be explored. Mr. Spotila. There are tradeoffs. If we want quality information in a timely manner so we can inform decisionmakers, then our best professionals have come up with their best judgment of what that sample size should be. If we work back from that because of cost reasons or concerns about how many people are being asked how many questions, then we have tradeoffs. These tradeoffs will be reflected in the information that we gather and, therefore, what can be provided for decisionmaking. Mr. Miller. This gets into the question of cost and other factors and privacy. I think that needs to be discussed some more later. Let me just ask--should we leave to vote? Mrs. Maloney. I have one question. Representative Collins feels very passionate about his legislation, which would basically make the long form voluntary. I want to ask you what you think of his proposal and how that would affect the long form now, and what do you propose for the American Community Survey? Do you propose to make that mandatory or voluntary, and really to just followup on the testimony of Mac Collins on essentially prohibiting any penalizing of people and not making it mandatory? Mr. Prewitt. First, the American Community Survey has been conducted under the umbrella of the decennial framework. Therefore, it has been conducted as a mandatory exercise. It has been field-tested in that way. The points that I would make in response to Mr. Collins are as follows: First, as Mr. Davis said, there is obviously a question of responsibility and obligation in society as well as rights and benefits. We ask so little of our citizens that it doesn't seem to me to ask for 45 minutes every 10 years to help create the kind of data that we need in this society is a big, onerous task. I know that it has been described that it takes a long time, but it takes 45 minutes, maybe an hour, every 10 years. Even under the ACS panel, no household could possibly get this survey more than every 5 years, and most will never get it. It strikes me as odd that we can't ask our citizens that much. On the other hand, it is possible for us to conduct this in a nonmandatory fashion. I don't think that you can separate mandatoriness from some sort of penalty whether it is imposed or not. And indeed, to go back to Mr. Collins' testimony, the reason that the Census Bureau said that it would not try to impose fines, is that we are not an enforcement agency. That would have to be a decision made by the Justice Department. Even if we wanted to, we could not have. That is why we were insistent that the Census Bureau was not going to impose fines; we can't. It is not our job. But I think the thing that you mentioned, Congresswoman Maloney, in your question to Congressman Collins, is it will be more costly. The reason one wants to pause before taking the mandatory framing off the table, is that by saying it is mandatory, what we really are saying is that the Federal Government takes this seriously. That is why jury duty is mandatory. That is why military conscription is mandatory. We take very few things seriously in our democracy. I would urge us to pause before we said to ourselves that we don't take getting the fundamental data that we need for this country seriously. I don't think that it is intimidating, and it is not only because of cost or quality issues, although data will be affected if it is not mandatory. I think Mr. Collins is just wrong in saying that by making it mandatory we reduce the level of cooperation with the census. We know better because we have done a lot of studies on this. You get a higher response rate and cooperation; but also in the field, people take it seriously because the government has said, ``this is something that we believe strongly in.'' I think there are lots of things at play. I do want to say, however, if the U.S. Congress decides that this ACS should be voluntary, that we would be able to get reasonably good data. It would not be as good as it is under the mandatory rulings, and it would be more costly. We would have to have more knocking on the doors and followup work in order to get the quality of data that we think that we owe the country, but it could be done. Mr. Miller. Let's take a 10 minute recess so we can go vote, and then we will continue with the answers here. [Recess.] Mr. Miller. Let's return to questioning. Mrs. Maloney will be back shortly. I will continue. There will be another vote in about 45 minutes. We will just proceed. We have to finish everything by 1 p.m., and we have a very important third panel that we want to save time for. On the sample questions, and I am sure that there is more information available and you will be clarifying the justification for it, sample size, I don't know if you want to add anything else. Mr. Prewitt. Just one other sentence, Mr. Chairman. I was a little more cautious than I should have been. When we say census tract level, we are talking about populations that can be in the neighborhood of 4,000 to 5,000. We primarily talk about 20,000 to be very cautious, but we will be talking about small jurisdictions having data then on an annual basis. So it is a fundamental transformation to produce small area data, and that is what sample size is about. If you want to pull it up to higher levels of jurisdiction, we can cut the sample. It is an issue of tradeoff, as John said. One other sentence on the mandatory issue: it is important for us to understand that we do a lot of very important surveys for the country--the Census Bureau does and HHS does and the Department of Justice does, which are voluntary. The difference in the ACS and all of these other important surveys is that the American Community Survey is the platform against which all of the other surveys create their statistical controls. So if it is weakened by low response rates or item non-response, all of the other surveys will also be weakened. We would want to be very prudent and cautious before we lifted the mandatory part off of ACS. It would be costly not just the ACS in terms of dollars and accuracy, but it would have implications for all of the other surveys that we do. Mr. Miller. Since we don't enforce it, what difference does it make? As Mr. Collins said, and as you have said publicly, you have not enforced it since 1960, and it was for one person. What impact does it really have? Can we focus group in a test sample to see if it makes a difference? I understand the psychological thing. One thing about the sample size, my understanding is that-- because you don't have the fixed deadline like April 1, you can do a second mailing and telephone, and then you will knock on doors. But some of the mail response rates even with the second mailing were not that high. We don't have the advertising campaign, and maybe that is an indication. But the mail response rates are 50, 60 percent. I think they projected with a second mailing under the decennial we would be close to 70 percent. Mr. Prewitt. In the high 60's. Mr. Miller. Do you have a comment, and what can we do to improve them? Mr. Prewitt. On the mail-back response rate, and we are comparing ACS to long form mail response rates, what we are hoping is that as this ACS gets embedded in the counties and the local leadership understands the importance of these data on a regular basis, they themselves will become a part of our promotional campaign. We will have a standing partnership. We cannot use mass advertising, obviously, for something like this, and we will hope that the local community will take leadership in promoting the importance of it. But I don't think that we have a magic way. We know that mail-back response rates are simply down in the industry across all kinds of surveys. We are pleased we do as well as we do with the ACS questionnaire, even in the 60's. But I think you are right. We will get better at the targeted second mailing, how we describe it and urge it, because we can go so quickly to the CAPI instrument, the computer assisted personal interview, and the computer assisted telephone interview, and we still end up with 97 percent. It will be more cost-effective if we get people to mail it back in. We have to create a presumption that this is part of what the society needs in the 21st century. Mr. Miller. Mrs. Maloney, would you like to continue? Mr. Spotila. I would like to add one comment. Just to clarify, we have not taken an official administration position on the Collins bill since it is so new. We are clearly going to take these views into consideration, but I did want to clarify that we were not stating an administration position at this point. Mrs. Maloney. In your testimony, Dr. Prewitt, you said that the American Community Survey, you got a 96 percent participation? That is astonishing. I would like to ask how you got it; and No. 2, would it affect the participation if we did away with the long form, because, as you know probably better than anyone else in this room, the tremendous effort that we put into heightening awareness and a sense of responsibility to put out the long form. I thought the advertising campaigns were absolutely great. Many constituents and people have commented on it. The census in the schools was my personal favorite. There were programs that we tried to put forward to raise the awareness of the long form. We had everything coming together with the census. If you don't have that there, would that have an affect on the response of the American Community Survey, and then how in the world did you get a 97 percent response? I think that is astonishing. Mr. Prewitt. I think a number of things went into that. It is a slightly different design. It is mail-out/mail-back, but then it is a targeted mail-back to nonrespondents, and so that has a little bit of a bump, and then it is a targeted telephone interview and then a targeted personal interview. The big difference is we are dealing with a very well trained professional staff, and they know how to find people. They know how to explain things more quickly. It is fundamentally different to conduct a survey with your professional permanent trained enumerators than this army of volunteers. It is different. That is why we continue to get at the Census Bureau quite high response rates, and it is partly because of that. We are particularly pleased with the ACS response rates in this trial period because we are also plugging people to respond to the census. Mrs. Maloney. If you didn't have that environment, do you think that it would drop dramatically? Mr. Prewitt. It is an interesting question, whether the census environment helped us improve the response rate on the ACS, but we are simply putting more of a burden on people who got both sets of questionnaires. We made sure that no ACS respondent also got the decennial long form. That was some tricky engineering to make certain that did not happen. We didn't want anybody to get both the long form and the ACS. What we are hoping for, Congresswoman Maloney, is that by embedding the ACS in the community with a community ownership of it by local leaders, mayors and commissioners, is that what we will get is a felt sense at the community level that this is critical data and we want high response rates. Mrs. Maloney. We talked earlier about the questions--about the questions on the long form, and I would like to ask all of you if you would comment on what the approval process is for the questions that will be on the American Community Survey, and will it be different from the questions on the long form, and at what point would it be appropriate to have input from Members of Congress? Mr. Spotila. Broadly in terms of the ACS questionnaire, the starting point, is a set of questions on the decennial long form. The interagency committee will be looking at the questionnaire to examine whether it is appropriate to continue in that way or to modify it. That process is one which we hope will be inclusive. The committee will not only discuss it among themselves, but will communicate with the Congress and people in the private sector to try to get other viewpoints on this. Ultimately, recommendations from the committee will lead to a submission that OMB will review under the Paperwork Act. There would be a final approval through that process where we would assess the recommendations and what has gone into them, including comments that were received throughout the process. That is very similar, I think, to the decennial census process. One of the issues that is still under discussion is the precise way in which the committee will interact with the Congress. The ACS questions, we certainly feel, need to be discussed with the Congress. We are still open as to the best way to do that. That is one of the issues that the committee will be talking about, and we are interested in your thoughts on that. Mr. Miller. Mr. Davis. Mr. Davis of Illinois. Thank you very much, Mr. Chairman. If this question has been asked and it is redundant, let me apologize. Dr. Prewitt, let me just see if I understand. If we were to replace the long form with the American Community Survey, let's just say that, would we use it as extensively as the long form is currently being used in terms of the number of surveys that would be done? Are we talking about more responses, fewer responses? Are there any projections in terms of that? Mr. Prewitt. Yes, sir. The way that it is designed, in a 5- year period we will be talking to approximately the same number of people we now ask the long form questions of, actually a few fewer, but give or take about 15 million households over a 5- year period. Then the data, we believe, will be much more extensively used because, as the witnesses said earlier, it is available almost immediately, and it is eventually available every year down to very small jurisdictions, maybe as few as 5,000 people. But the burden on the American public in a 5-year period would be similar to what the current long form burden is. Because we would like to make the data more timely by continuing to roll the sample through, in effect you are increasing the burden on the American people over a 10-year period. Now, they won't feel it that way because it won't be that intense environment. If you add up the minutes that people use to answer the ACS questionnaire, it will certainly be higher than the number of minutes invested in the long form in the decennial census environment. But it does really improve the possibility of reengineering the decennial process so we can make it the civic ceremony it is and look for 100 percent coverage with new technologies, because doing the long form in a decennial environment is a very complicated part of our operations. So it is better on both sides. It is better for long form data, and it is better for the basic population count that goes into the apportionment and redistricting numbers. Mr. Davis of Illinois. If the data is more timely, does that necessarily make it more reliable? Mr. Prewitt. In some respects it does, because all of the data can now be summarized across a 4 or 5-year period, even if you don't--let me just say it can be summarized. You are reducing some of the variability that you get from responses. You know, when you are doing the long form, suddenly it is 3 p.m., on a Tuesday. We can only take that. That has to sit there for 10 years. If you are summarizing across different respondents across a 5-year period, you minimize some of those erroneous fluctuations and biases that can occur, so we think the data will be more reliable. The other big difference is that you are doing this with a permanent, professional, trained staff, which does give you more reliable data than doing it with a large army of part-time temporary employees. Mr. Davis of Illinois. I assume that the updates--for example, I often will make use of data in terms of saying when the census was taken in 1990, well, this is 2000. What does that really mean in terms of the actuality of what has taken place in some communities or some instances? And so we really won't be talking about information that is 10 years old because we will have these constant updates in terms of the data constantly coming in; is that correct? Mr. Prewitt. Exactly. Mr. Davis of Illinois. I have no further questions, Mr. Chairman. Ms. Wallman. May I add one point to complement something that Dr. Prewitt said? We have focused on 15 million sample households for the ACS, 30 million over 10 years, in contrast to 20 million for the census 2000 long form. But the complementary point should be underscored that we are still in the process of evaluating the specific content for the American Community Survey in the long run, and we may or may not need to ask every single question every time when we are in this annual American Community Survey environment. So the burden on individual households of responding to this could be, in fact, less than we currently perceive in the current long form mode. Mr. Davis of Illinois. I appreciate your comment because I think what I am getting, at least for me, is some assurance that we might be talking about something that is going to be more beneficial as opposed to simply responding to some criticism that we might have gotten in relationship to the long form, which I think is very different. Mr. Spotila. That is an extremely important point. This is not a defensive reaction to criticism of the decennial long form. It is a result of a lot of careful planning and thought on how we can produce information that we need on a timely basis and do it in a way that makes sense. The Census Bureau should be complimented, along with those in the Federal statistical community that have worked so hard to do this. Mr. Miller. The timeliness of the data, and this gets back to the sample question, do we need to take 3 million or 3 million over 2 years, ideally you have your data on a larger sample which gives quicker data, but it gets down to a question of cost. I want to talk about that. We have to address the cost over a 10-year period. This goes into effect 2006. Mr. Prewitt. 2003 we would like to field the first. Mr. Miller. So we are looking at a 10-year period because it will reduce the cost of the decennial. We mentioned the postcard census, and you mentioned the Internet. How do you envision how the decennial will be done in 2010, assuming that we have the American Community Survey and we are down to those few basic questions? Mr. Prewitt. Well, we do obviously think that we would use the Internet much more than we used it in 2000. We are looking at using the Internet in the ACS environment, too. We hope that the Internet would be one of the major ways in which people responded to the American Community Survey questionnaire. We have that in the field right now. We will be reporting on it soon. Looking forward to 2010, it is so complicated because you don't know what the technology is going to be like then; and we talk about the Internet like it is going to be the Internet in 2010. We know that technology is moving so very, very fast. We have a group who tries to track this and pay attention to it. I think we would be able to use more administrative records. Now we are trying to get to the hard-to-count population groups, and we may well be able to use--the way we do with the military, we go to administrative records. Now we are trying to get the population count and very basic information like gender and race and ethnicity, if we still need those. We certainly need it now to administer the Civil Rights Act of the 1960's. If that were to change, that would change what we need on the short form. If all you need is a population count and enough evidence about the population count to make certain that you have reliable data, we can go to a very simple form to get a population count in the mail system, if we are still using the same mail system in 2010. And then maybe, Mr. Miller--you have been concerned about this, the second mailing--we may be in a position with only the short-form data and with our much better address list and a much better technology to track things to target a second mailing. I would think that we could do that in 2010, which we did not think that we could do in 2000. But if you can do a targeted second and third mailing, you can do more efficient followup work. You can redesign the entire thing with the same level of coverage with less expense. I don't know whether you still need the big kinds of promotional and advertising effort. It is hard for me to know how important that kind of apparatus, which was very important to 2000, would still be. Mr. Miller. I want to go back to the cost question, and I know we have this 10-year number. What is the 10-year cost of ACS combined with the decennial, versus the decennial with the long form? Mr. Prewitt. I will tell you our goal, and we are working very hard on it. We are obviously poring through the ACS, and with regard to the second mailing and telephone followup, we are poring through that data. Our goal is to come back to the U.S. Congress and say that if you take the current decennial budget, which we know is about $6.5 billion, and you put that into constant dollars in 2010, we would be able to come to you with a decennial design and an ACS design which would be within that budget. We would be giving much more timely data and better long form data because we are now doing it every year rather than every 10 years, and then we would get the decennial head count, and we would love to be able to present to you a design that would do that within the same framework. Mr. Miller. So the cost would be equivalent, you're hoping? Mr. Prewitt. That is what we are hoping. Mr. Miller. It costs $250,000 a month, and what kind of infrastructure expansion would be needed as far as field offices, regional offices? Mr. Prewitt. It is very important--that is a very important question. We will do the ACS out of our current structure. There is no additional infrastructure we need. We obviously need enumerators to do it. Second, we do all of the data capture in Jeffersonville. The only kind of infrastructure that we need for the ACS that we don't need for the normal surveys is the partnership structure. We would like to keep some of our local partnership people to maintain the address file. It is very key to maintaining the address file, the ACS. Mr. Miller. The address file we are using for this decennial, the plan and intent is to keep it current? Mr. Prewitt. Absolutely, every year current. We are working with the local county people so there is no extra 2010 expense if we are still using a mail-back. Mr. Miller. You have that built into your costs? Mr. Prewitt. Yes. That is what part of the test sites are telling us. Mr. Miller. What kind of costs are you projecting for the 2003? ACS is $40 million? Mr. Prewitt. I think it is $20 million; $25 million is what we asked. The house mark is $20 million. Mr. Miller. What are you projecting for 2003 when it is fully operational? Mr. Prewitt. I would love to give you that number. We simply have to get the field data analyzed, and that is what we are doing right now. We think by the time we present the 2002 budget, which will be in the standard appropriation cycle, that we will have a very good indication of what the 2003 budget will be, but right now it would be--the error term is too high. Mr. Miller. Mrs. Maloney. Mrs. Maloney. Go ahead, Mr. Chairman, you are on a roll. Mr. Miller. This is almost like a work session. Will you replace any surveys with this? Is that some of the intent? Are there any surveys that could be incorporated into this that would no longer be needed? Ms. Wallman. I think at this stage we are prepared to say two things on that topic. First, what we anticipate at the moment is not so much replacement of ongoing surveys, but more efficient ways to carry out some of our major household surveys. The second thing that I would add is that we are now beginning to look at opportunities where we could use the infrastructure of the American Community Survey to avoid having to develop additional new surveys. If the Congress has a requirement for a new piece of information, for example, for the Children's Health Insurance Program, we could use the vehicle of the American Community Survey to collect that one additional piece of information in the framework of the American Community Survey that has the other basic demographic information we need, rather than fielding a whole new effort. We see those kinds of opportunities, and that is exactly the kind of thing that our interagency group will be looking at and will be looking to work on with the Congress and other organizations that need to field new kinds of information collections. Mr. Miller. What questions need to be included is another issue, and I know we had all of the justifications 2 years ago, and I don't think that there was much input from Congress, and the question is how much is really needed at the tract level? You say that you are in the process or are planning on doing a review of that issue, and I think it is very legitimate. The temptation is to ask more questions. There are 400 citations now requiring it. The temptation--so you have a tough job. Mrs. Maloney. And that is from us. Mr. Miller. I know. Mrs. Maloney. It is not their fault. Mr. Miller. I know. As we plan for this, we need to say do we really need this at this level of detail. Mrs. Maloney. Reclaiming my time, that would be a good focus of a committee hearing that we could have, all of the various--or just a meeting on the citations and just our own review of whether or not we need them. Mr. Miller. You might expand how you did it for this census. I know that you scrubbed it and you did focus groups, and I don't think that people appreciate the difference until you look at the 1990 census form, which was done before our involvement, and the Bureau did a good job. But the plan is to scrub the questionnaire and decide how? Mrs. Maloney. Reclaiming my time, remember that we all got a huge booklet 2 years in advance of the completion before they printed the forms to go out for our own input and our own questions or suggestions or whatever, and as we all know, it was exactly like the census that Bush and Reagan oversaw with the exception of adding one new question because of welfare reform and deleting four. Mr. Miller. Would you comment on what you are going to review, and it may take legislation. Mr. Spotila. These are important issues. We are looking forward to the interagency committee working closely with the Census Bureau to take a look at these issues. We know that there are 130 laws which require various kinds of information. The potential is there for more, but as to the decision about how much detail we include in the questions, which questions should be on the American Community Survey versus some other survey or asked in some other way, it is fair to have a discussion about whether there are alternatives to do the same thing. That is not to minimize the fact that if by law the information is required, then we need to obtain that information. But there is also going to be an ancillary review that will be necessary. If the American Community Survey proves successful, and if this is an approach that we want to follow, we collectively, including the Congress, then it will also be necessary to review statutes that make references to the decennial census. We are going to need to determine whether any other kind of legislative adjustments will be needed to make sure that we cross-reference properly. This is a major shift. We think that it is a positive one. But there will be a very strong need for this process to be a collegial one with the Congress. In working sessions with the Congress not only like this one, but even outside the context of a hearing, we can have this kind of a discussion and identify what the choices are and get some meaningful input as to perhaps how it can best be done. We are going to rely heavily on the Census Bureau as the lead on this to develop a lot of this information. They have been very good at doing it in the past, but we are conscious that this will be a successful process only if it is inclusive and very cooperative with the Congress. Mr. Miller. In Congress if, for example, the welfare reform bill had the question about grandparents, do they make that decision unilaterally? Your staff is nodding their heads behind you. Maybe questions get added without us realizing. That may be legitimate, but is that needed at the tract level? Mrs. Maloney. Reclaiming my time, I tell you that the welfare reform question is a tremendously important one, particularly in New York where many children are having babies and their mothers are not there, and the grandparents are becoming the mothers. This is widespread, and to understand the ramifications of that is very important. Personally, I fail to understand the criticism of the long form. I got the long form. It didn't take me 45 minutes. I think it probably took me 25, and I considered it a great honor to fill out the form, and they were reasonable questions about real information that planners would need for the city, for the State and for the Federal Government. But--so I welcome as much oversight, and I think oversight is important and hearings, and maybe we should send three or four books to the Members of Congress as opposed to one book 2 years in advance. Maybe we should send a book a year and have more discussion about it. But I have heard one complaint about this glorious American Community Survey which I would like to ask a question about. A member of the private sector contacted my office and he alleged that the American Community Survey will give the Census Bureau an unfair competitive advantage over private sector research firms in competing for government contracts. What is OMB's position on this issue? Do you believe that such a competitive advantage will exist; and if so, what can be done to remove that advantage so that the private sector can compete fairly for government contracts? Mr. Spotila. This is a point that we are going to be looking at and have the committee look at. We know, if the American Community Survey is introduced and implemented on a broad scale, that we will get a tremendous amount of very useful information that the private sector will benefit from. Even people in the research business, if you will, will benefit from it. But there can be issues. Since the Census Bureau needs to keep information that it collects confidential, if it is going to put some of this information to use in a way that benefits the American people, it may find itself doing new things with this new information. Some of the new things are being done in the private sector now by companies that charge for them. Some perhaps operate less efficiently. It is a fair question, and I think that the committee should examine this. We should have some discussions with affected parties. We certainly don't think that it is appropriate that the government compete with the private sector in areas where the private sector can and should have the lead, but there are going to be other areas where the government can serve the American people very well by getting them better information and getting it to them less expensively. There could be a public purpose in doing that. Change always brings ramifications that one has to work through and develop a better understanding of. We all have to adjust to that change. I think it is a fair question and one that the committee should look at and will look at. Mrs. Maloney. Would you like to comment, Dr. Prewitt? Mr. Prewitt. I, of course, came from that industry. That is, I was director of the National Opinion Research Center for 5 years. I was in active competition with the Census Bureau at that time for government contracts. I really strongly believe that the private survey industry is a critical source of innovation of survey methodologies, of new ideas and so forth, and that the government survey efforts will be harmed if we don't have a robust private sector survey industry. I think the issue that has been raised really--there are three or four different dimensions of it. They have to do with screens for certain rare population groups or small-size population groups and address files, the degree of detail that we can provide the private industry. You will shortly hear from Richard Kulka from RTI who will address this issue with you. But I want to put the principle in play that the Census Bureau itself really does work closely with our colleagues, statistical and survey colleagues in the private sector. I think those problems can be worked out. People from survey houses understand that. We simply have to do that. We have started those conversations, and I am convinced that there is a way to do it, even honoring our title 13 obligations which we have to honor. Mrs. Maloney. Does the Census Bureau compete with the private sector for government contracts now to gain data? What areas do you compete with now in the private sector for government contracts? Mr. Prewitt. Yes. If the National Science Foundation wants to collect data on earned doctorates, they may come to the Census Bureau to ask how we would do it, or they may go to the private sector and ask, how would you do it? In that sense we are in a competitive situation. Mrs. Maloney. Is there a process where you actually bid against the private sector, or is it just a choice between--by government? Mr. Prewitt. We can't bid, so we don't bid. Mrs. Maloney. So they can make an in-house decision that you would be a better vehicle? Ms. Wallman. I would be happy to add to that slightly. In general we are talking about major surveys that involve the household as the unit we are querying. We have several surveys across the government, such as the Health Interview Survey that is conducted by the National Center for Health Statistics, education statistics surveys, and so on, and the choice that these agencies face is, do they go to the Census Bureau and take advantage of the household survey frames that the Census Bureau holds as a result of its ongoing work, or do they go to the National Opinion Research Center, Research Triangle Institute, and so on, and have a competition among those outside organizations. We use both mechanisms extensively. There are in some cases household surveys conducted by the Census Bureau in partnership with the other statistical agencies. The Current Population Survey that produces the monthly unemployment numbers, the biggest of them, is a partnership between the Bureau of Labor Statistics and the Bureau of the Census. In many other cases, particularly when we are doing longitudinal work, the agencies choose to partner with National Opinion Research Center, Research Triangle Institute, Westat, and others of that ilk. Mrs. Maloney. Thank you. Mr. Prewitt. Certainly the private sector has enormous competitive advantage vis-a-vis the Census Bureau. They pay better salaries. They can sometimes work more quickly and efficiently because we have a lot of constraints in terms of how a Federal agency functions, of course. The fact is that the Census Bureau has been around for about 100 years doing survey research for the Federal Government. The private sector industry doing government contract survey work began in the 1950's and 1960's, and it has grown like that. This is an industry that has been enormously successful, and very importantly so for society, doing government contracts over the last 30 or 40 years. So it is not as if they are not there. They are growing and robust. They are statistically sophisticated, they do things we cannot do, and they pay much better salaries. Mrs. Maloney. Ms. Wallman, you usually come to see me about the Statistical Efficiency Act. This has been a project that you have promoted and worked on for many years, and I am pleased that the House finally passed it last year. What effect will the final passage of that have, if any, on the American Community Survey? Would you comment on that? You were so dedicated to passing this, so I just had to ask you about it. I applaud you and your hard work in that area. Ms. Wallman. Thank you, Congresswoman Maloney. I always appreciate the opportunity to say more about the Statistical Efficiency Act which we are delighted had bipartisan unanimous support from the House of Representatives, and we are still working to get passed in the Senate. I want to underscore a couple of things in that vein: First, the Statistical Efficiency Act first and foremost gives statutory protection for the confidentiality of statistical information that is collected by several agencies, a privilege that the Census Bureau currently enjoys but some of its sister agencies do not. Second, we do see this as part of the package of improvements that we can bring to bear to more efficiently gather information and ultimately to more efficiently provide information for public use. The American Community Survey is a fundamental piece of the statistical collection infrastructure that ultimately, with our Statistical Efficiency Act in place, could bring us some of those benefits that we see. I want to take the opportunity of this question to underscore that in addition to the statutory confidentiality protection in the Statistical Efficiency Act, in any case where we are proposing to use the information either to go back to respondents for the kind of sampling that we have been talking about here from the American Community Survey or to use the American Community Survey in concert with statistical programs of other agencies, we will always use the same practice we do now to inform respondents of these potential uses of the information before we collect the information from them. I think that question has been raised in some of the prehearing dialog. That is a policy which has been in place that we would keep in place, and we would not intend to go out, collect the American Community Survey, and then use the information in additional ways without prior notification to respondents. Mrs. Maloney. Thank you. My time is up. Mr. Miller. Let me make one comment. Following up on that question about the private sector, I think some of the concern is that American Community Survey may give a huge competitive advantage to the Census Bureau because it gives a new ability on an ongoing basis to have a very large sample to work with. That is one of the concerns of the private sector. Mr. Davis of Illinois. Mr. Davis of Illinois. I have no further questions, Mr. Chairman. Mr. Miller. As I said before, this is a great opportunity for the Census Bureau and the country to get better, more accurate data and more timely data. We do have a lot of questions that need to be answered. Thank you all very much for being with us here today. I will now call up the third panel. [Witnesses sworn.] Mr. Miller. The record should note that all witnesses answered in the affirmative. I appreciate all of you being here. We may have a vote on the floor in a few minutes. Mr. Crowe, if you would like to proceed with your opening statement. STATEMENTS OF DAVID CROWE, STAFF VICE PRESIDENT OF HOUSING POLICY, NATIONAL ASSOCIATION OF HOME BUILDERS; EDWARD HUDGINS, DIRECTOR OF REGULATORY STUDIES, THE CATO INSTITUTE; CHUCK FLUHARTY, DIRECTOR, RURAL POLICY RESEARCH INSTITUTE; RICHARD KULKA, VICE PRESIDENT OF STATISTICS, HEALTH AND SOCIAL POLICY, RESEARCH TRIANGLE POLICY, ACCOMPANIED BY JUDITH T. LESSLER, DIRECTOR, STATISTICS RESEARCH DIVISION; AND BARBARA WELTY, PRESIDENT, NATIONAL CENTER FOR SMALL COMMUNITIES, BOARD OF DIRECTORS, NATIONAL ASSOCIATION OF TOWNS AND TOWNSHIPS Mr. Crowe. Thank you, Mr. Chairman. I thank you for the opportunity to participate in the hearing on the Census Bureau's American Community Survey and the collection of important data in the post-census 2000 era. I will summarize my statement and ask my full written statement be included. Mr. Miller. We will include everybody's full written statement for the record. Mr. Crowe. My name is David Crowe. I am the staff vice president for housing policy at the National Association of Home Builders. I also represent a coalition of housing organizations, called the Housing Statistics Users Group, on the 2000 Census Advisory Committee to the Secretary of Commerce, and 2 years ago I was pleased to testify before this subcommittee on behalf of the Coalition to Preserve Census Data, a group of broad-based industry and professional associations and individual companies representing diverse economic sectors, including retail, print communications, housing, transportation and marketing. We are grateful to the committee for supporting the coalition's request for preserving the collection of data in the 2000 census. We applaud your continued monitoring and review of this complex issue related to the collection of data. We encourage you to examine these issues objectively to ensure that the Nation's information needs are met in the most cost- effective and efficient, reliable manner. Business, industry and professional data users are studying the proposed ACS carefully as well. I am pleased once again to speak on behalf of the many stakeholders today. I have three points I would like to convey. First, the collection and dissemination of basic data about people and their conditions are core to the statistical infrastructure of our country and a legitimate and necessary function of a democratic government. Second, demographic and socioeconomic data are essential tools for informed decisionmaking, prudent investment and an efficient allocation of resources in both the public and private sectors. The detail of America's homes and communities is the meat on the bones of the basic count of people. Third, alternatives to the collection of demographic information should be considered. The ACS appears to be a sound alternative to the traditional census long form, and Congress must be willing to provide sufficient funding and support reasonable content before deciding to eliminate the long form. On the question of Federal support, I ask two questions: Should the Federal Government collect and publish statistics on the population and the characteristics of our communities? Is establishing a basic statistical picture of America a legitimate function of the government? I say yes. Data collection not only is a legitimate function of the government, it is in many respects a necessary activity of a democratic society. Long-form data are employed by policymakers, businesses and public advocates because it is reliable, consistent, and comprehensive. No privately run organization could duplicate or replicate the conditions and the infrastructure or certainly gain the trust of respondents. Critics suggest that data collection is used to justify unneeded expenditures and promote government largesse. I contend the opposite is true. An informed citizenry exercises the most effective check on wasteful and inefficient government spending and guards against improper or inefficient use of resources and power. Some observers believe that the wide range of information traditionally collected on the long form constitutes corporate welfare. I think this is also false for four reasons. One, while census data provide basic information about the characteristics of our population, businesses must conduct more focused research and data collection tailored to their specific needs. Census data represents a universally consistent and respected basis of fact that no private company could replicate. It is a statistical benchmark rather than a tool of competition. Second, there are Federal laws, as has been mentioned, requiring the collection of these data. Third, the census is the only source of information for small areas such as tracts that allow government planners and the private sector to prepare their communities for the future. And finally, businesses do pay taxes which help support this endeavor. Concerns about individuals' privacy are legitimate. Assigning the task of data collection to the Census Bureau provides better protection than alternatives. Most people understand that the collection and dissemination of data from the census is far safer, infinitely less compromised and effectively isolated from revelation because the Bureau is responsible for the security rather than a private for-profit firm. The Census Bureau does ask some personal questions to produce impersonal data. The fact that many people may not fully comprehend this link is more a question of better education than a revolt against the government. The development of the ACS presents a valuable opportunity to examine this issue and build a foundation of mutual trust. I would like to take just a minute to explain how this gathering of information promotes economic growth and improves the quality of life. Business and industry need basic demographic and socioeconomic information and housing data for the same reasons that the government does, to make informed decisions, to make accurate projections, to make prudent investments. These decisions and judgments allow businesses to create jobs, to provide appropriate products and services to our communities, to invest resources in underserved urban neighborhoods and rural areas, and to assess work force readiness. Let me use an example from my industry, the housing industry. The National Association of Home Builders represents about 200,000 firms which build 80 percent of the 1.5 million homes in this country. Residential construction is 5 cents of every $1 spent in this country. The census long form is the only source of geographically detailed, nationally comparable data on our Nation's housing stock and the people who live in those homes. Local planners, home builders and financial institutions rely on it to decide whether to invest in new housing, where it should be and what needs it must fulfill. Without basic housing conditions gleaned from the census, these companies either wouldn't invest in some areas, or they would make inaccurate decisions that could hurt both business and consumer. They do collect data on their own, but it serves to amplify the basic data rather than complement it. It is also instructive to consider the transportation sector, another large component of the economy. Transportation services contribute $378 billion in the economy, up 21 percent from 1992, and the long form serves as the only source of geographic information for that sector as well as providing information on work trips, transportation preferences and household and working characteristics. The American Community Survey offers a promising opportunity for information collected in the census long form. By measuring many characteristics of our population and housing stock on a continuous basis, it would provide far more timely and more accurate information. It would also be useful for emerging data needs. Replacing the census long form with the American Community Survey also carries risks. We know that every 10 years the census puts forth a massive infrastructure to count the population and allocates a substantial amount of money to that task. While the ACS may require more direct funding over a 10- year period than the census long form, the additional costs might be offset, as was said in the earlier testimony, by more prudent investment and more precise targeting. I encourage you to consider the value of that investment as you weigh the cost of replacing the long form with the continuous measurement. There are a number of issues to consider before Congress and external stakeholders embrace such a paradigm shift in the strategy for collecting geographically detailed information. These issues include a process for determining what questions are asked and how they are asked, and an assessment of sample size to ensure adequate coverage and cost. Mr. Miller. Can you summarize? Mr. Crowe. In summary, congressional oversight is an important objective. I would simply summarize my three points: Basic data collection is a responsibility of our government. Basic demographic and social data provides the infrastructure of information about our country, and the ACS is a promising alternative. Thank you. Mr. Miller. Thank you. [The prepared statement of Mr. Crowe follows:] [GRAPHIC] [TIFF OMITTED] T3836.017 [GRAPHIC] [TIFF OMITTED] T3836.018 [GRAPHIC] [TIFF OMITTED] T3836.019 [GRAPHIC] [TIFF OMITTED] T3836.020 [GRAPHIC] [TIFF OMITTED] T3836.021 [GRAPHIC] [TIFF OMITTED] T3836.022 [GRAPHIC] [TIFF OMITTED] T3836.023 [GRAPHIC] [TIFF OMITTED] T3836.024 [GRAPHIC] [TIFF OMITTED] T3836.025 [GRAPHIC] [TIFF OMITTED] T3836.026 [GRAPHIC] [TIFF OMITTED] T3836.027 [GRAPHIC] [TIFF OMITTED] T3836.028 [GRAPHIC] [TIFF OMITTED] T3836.029 Mr. Miller. We are voting across the hall, and so I am sitting on the side so I can leave quickly. We do have lights for 5 minutes, and please watch the light. Mrs. Maloney. And I have to run to another committee meeting, and I will be right back. Mr. Miller. Mr. Hudgins. Mr. Hudgins. Thank you. I will summarize my longer remarks. I am Dr. Edward Hudgins of the Cato Institute, and I commend the committee for holding these hearings and thank you for the opportunity to speak today about the census. While there are legal and methodological questions concerning the proposed American Community Survey, I raise the more fundamental issue: Should the Federal Government be asking the questions currently contained in the census and mandating citizens to answer? I have done many call-in shows and interviews on the census, received numerous e-mails and telephone calls of concern. I report to you today a sentiment I believe is shared by millions of Americans. The lack of proper decorum and its expression is not aimed at the individual Members of Congress in attendance, but rather at the system as a whole. An accurate summary of that sentiment would be ``most of the census questions are none of your damned business. We hire you to protect our lives, liberties and property, not to butt into our affairs. Stop your meddling.'' Let me return to the proper decorum and explain this position by answering four questions. First what does the census suggest about America's civic order? The Constitution authorizes a census to enumerate persons in order to apportion electoral votes. Yet the 53 questions about income, how we get to work, how many toilets we have, have nothing to do with that purpose. The civics lesson is that Washington political elites need the information so they can redistribute wealth and limit liberty according to their visions of a good society. Without census data, political elites would find it difficult to convince the public about the needs for their policies. We are told that filling out census forms helps our communities and ourselves obtain aid for roads, schools, child care and recreation. In the past, the Federal Government took far less from families in taxes and did not so dominate public policy that it reduced State and local government tax bases and functions. Now we are urged to answer census questions so we can ransom back our own money. The decline of American federalism provides the impetus for the intrusive census questions. I also note, by the way, that the census seems to be obsessed with race. We are asked three questions with numerous subdivisions of answers. It is instructive by the way that we are asked what race we consider ourselves to be. I guess this puts off until the future the need for Nuremberg-type race laws or mandatory DNA tests. A second question is why are individuals so upset about the 2000 census since it doesn't contain more questions than the 1990 one? The first reason is that the information and communications revolution and the Internet have made individuals much more sensitive about their privacy. I would be happy to talk about the private sector response to this. Mr. Miller. Excuse me. I have to go across the hall for 2 minutes. If we can take a short recess. I apologize. [Recess.] Mr. Miller. I apologize. When you are on Appropriations, you are not supposed to serve on another committee of Congress, and I am the only Member allowed to serve because of the census issue. They are doing the full markup of D.C. appropriations, and they are right across the hall. I apologize. Please continue, Dr. Hudgins. Mr. Hudgins. The second question is why are individuals so upset about the 2000 census since there are not more questions than on the 1990 one? The first reason is that the information and communications revolution and the Internet have made individuals more sensitive about their privacy, and I would be happy to talk about the private sector response in that area. The second is that individuals have seen unprecedented assaults by government on their privacy in recent years. There are reports of census takers asking for and being given access to records of apartment tenants from rental offices. The FDIC proposed a regulation which would require bank tellers to ask customers about any suspiciously large deposits or withdrawals. The Postal Service regulations last year would have made available to anyone off the street the home addresses and phone numbers of customers of private mailbox companies. The administration's medical privacy regulations would eliminate the need for the government to obtain individuals' permission to use or distribute their medical records. One administration proposal for a unique health identifier would require a DNA sample from each American. The Kidcare Program can allow bureaucrats armed with psychobabble to spy on parents and interfere with child rearing. Medicare now encourages health care workers to spy on the elderly in their own homes, and bills before Congress would allow Federal agents to enter homes, make copies of personal papers or computer hard drives and not notify citizens that their homes had been searched. I think Americans see a pattern and lump the census into a pattern of invasion of privacy. Third, what problems do these intrusions cause? I think if you see the census as part of this pattern, those dangers are quite clear. I would add that I think the census is a free marketing survey for corporations, and I do consider that corporate pork. Finally, what should be done? The Federal Government should retain only those census questions necessary to exercise its constitutional mandate to enumerate the population. Citizens should not be required to answer under penalty of law the questions in the long form nor in the American Community Survey; that is, the answers should be voluntary. Also, we should question what questions are asked in the survey. Remember, because Congress tends not to honor constitutional limits on its jurisdiction, there are no logical, only political limits on what powers it can exercise. Thus, there are no logical, only politically imposed limits to what questions the census might ask pursuant to policy goals. Will we see future questions on whether we smoke in the home or about what our diets and our exercise is? In summary, the American citizens should not justify or have to justify to government why they should keep their personal affairs private. The government should stick to the Constitution and respect the privacy of citizens and make the census and answers to the American Community Survey voluntary. Thank you. [The prepared statement of Mr. Hudgins follows:] [GRAPHIC] [TIFF OMITTED] T3836.030 [GRAPHIC] [TIFF OMITTED] T3836.031 [GRAPHIC] [TIFF OMITTED] T3836.032 [GRAPHIC] [TIFF OMITTED] T3836.033 [GRAPHIC] [TIFF OMITTED] T3836.034 [GRAPHIC] [TIFF OMITTED] T3836.035 [GRAPHIC] [TIFF OMITTED] T3836.036 [GRAPHIC] [TIFF OMITTED] T3836.037 [GRAPHIC] [TIFF OMITTED] T3836.038 [GRAPHIC] [TIFF OMITTED] T3836.039 Mr. Miller. Now we have Mr. Fluharty. Mr. Fluharty. Thank you, Mr. Chairman. I am going to dispense with my oral testimony and make four points. First of all, I want to commend you for looking at the rural implications. The rural reality most of us believe is out there is not the rural reality our citizens and public servants live in. Congresswoman Emerson shared some statistics, and my full testimony has others. I will simply name one. In the last 8 years, 50 percent of the population growth in rural America has been in-migration of metropolitan citizens. Inversely, metropolitan America has lost 10 percent of its population to rural areas. None of this is reflected in the data of the 1990 census. Second, growth is a phenomenon in rural America. Third, Federal transfer payments are absolutely key to the economic viability in most rural communities, 20 to 25 percent of per capita income. And last, rural governments and rural development organizations are key to sustaining private sector, regionally targeted enterprise in rural America. It is really important to understand that public servants at the local level, like yourself, should have a fair and equitable playing field in which to make good decisions. That currently is not the case. I will use two very short examples. The Congressional Rural Caucus just asked us to assess a GAO study on the identification of economically distressed communities. That study was done correctly, and used census tract data. The Rural Policy Research Institute took that data, used the commuting codes that the Economic Research Service developed, and came out with very different findings. Should we have had the American Community Survey in place, it would not have been necessary for two Federal agencies to end up with different data findings. Second, we work every day with county governments, small towns and development organizations, trying to help them make good public sector investments. To these entities and jurisdiction, the existing data simply does not reflect their choice pattern. It is a disadvantage in the American democratic experience that exists because these jurisdictions cannot afford consultants to do the studies that are timely. They don't have budgets or staff. In my testimony I raised two cautionary concerns. One is vigilance, and the other is a realistic assessment of what goes on in the ground. I have been very impressed by what ACS has tried to do to get rural social scientists, statisticians and rural policy analysts engaged in looking at the issues that are critical for rural America here. I participated in two of those. I am impressed with their willingness to engage these challenges. They are very real. I will mention three. The moving year average is critical. There are problems with it. Citizen inconvenience is a challenge in terms of survey groups, and the privacy issue in a small sector sample is an issue. Second, there is a cultural challenge in rural America, with gathering this data. Third, I believe this committee needs to stay engaged on those issues. Do I believe ACS is engaged in attempting to deal with this? Yes, I very much do, Mr. Chairman. Finally, I would simply say the reality impact needs to be there. The rural partnerships with local governments and development organizations need to be in place to get the right data, and I would urge the committee to continue to assure that ACS works with appropriate, designated jurisdictional partners to move that forward. Last, all of us that work to make rural America better have as a common core concern the fact that we do not have good data upon which to help public servants make decisions. Whatever this committee decides to do, I would urge that, at least, this rural differential is adequately considered, and I appreciate the time to bring that to you. Mr. Miller. Thank you. [The prepared statement of Mr. Fluharty follows:] [GRAPHIC] [TIFF OMITTED] T3836.040 [GRAPHIC] [TIFF OMITTED] T3836.041 [GRAPHIC] [TIFF OMITTED] T3836.042 [GRAPHIC] [TIFF OMITTED] T3836.043 [GRAPHIC] [TIFF OMITTED] T3836.044 [GRAPHIC] [TIFF OMITTED] T3836.045 Mr. Miller. Next we have Mr. Kulka accompanied by Judith Lessler. Mr. Kulka. Thank you for the opportunity to raise a couple of points. I am going to truncate what I might have said in the interest of time. The key issue is we are very supportive of the goals, objectives and fundamentals of the ACS and believe it would make a valuable contribution to our work and those of others like us and those of the Federal statistical system in general. So all of the things that have gone before today, we would endorse most of them. However, we are concerned about a couple of issues: that the ACS could have negative implications for survey research in general and the ability of RTI and others like us to conduct survey research business. We are concerned because the Bureau's publicity to date clearly indicates, and it has come out in this hearing, that it plans to go beyond its traditional mandate under title 13 of releasing data for statistical purposes only and profiles of groups of individuals within broad geographic areas to use the information collected in the ACS about specific individuals to construct sampling frames for other surveys. This expanded use raises concerns for three reasons, all of which we think could be resolvable if we wanted to raise them. First there is a conflict between the ACS mandatory reporting requirement, the confidentiality statements that are given to the public and the intended uses of the data. The Bureau's publicity notes that the public is required by law to respond to the survey, and that data will only be used in aggregate form for statistical purposes. But the Bureau plans to use specific information from specific respondents to select people for inclusion in other nonmandatory surveys. Most respondents will not interpret, we believe, statements that their data are confidential and only released in aggregate form as permitting such use. That has been addressed in earlier panels, but it is worth emphasizing. The survey research community depends on the trust that the public places on our promises of confidentiality. Throughout its history, the Census Bureau has a distinguished record as a leader in making sure that this public trust is not violated. We view such plans to use data in one way and without careful consent of the other as a serious threat that needs to be resolved. Second, these expanded uses of the ASC will adversely affect the ability of the private sector to compete with the Census Bureau. The Census Bureau contracts with other government agencies for surveys. We also contract with other agencies for surveys. Clearly the use of the ACS to select samples that are specifically tailored to the needs of other surveys will give the Census Bureau a large competitive advantage over the private sector organizations which will not have access to this information because of title 13 restrictions. Third, this unfair competition could severely damage private survey research. While the Congress might decide that it would prefer that the Census Bureau conduct all Federal surveys, there is a danger in doing so. The expertise that the private sector has in conducting scientific surveys is valuable, as Dr. Prewitt referred to earlier. Agencies and businesses have options when they need to conduct a scientific survey. Private survey research organizations have led the way in developing innovative survey procedures, and restricting competition can decrease innovation. Creating conditions that unfairly disadvantage the private sector is counter to the current emphasis on strengthening the private sector so that the government can reduce its expenditures in its need to conduct the Nation's business. The net result would likely be a significant increase in the proportion of survey research conducted by the Federal Government at the expense of the private sector. We would like to reemphasize our support for the ACS if it is conducted in a balanced way such as under the traditional mandate of the Census Bureau to provide aggregated information for use by the government and nongovernment organizations and the public. However, if the uses of the ACS are expanded to include selecting samples for other surveys using privileged information, the pledges of confidentiality, as they are currently understood by citizens, will be violated, and the private sector survey research organizations will be harmed by unfair competition. There are a number of potential solutions to this problem. First, the Bureau could abandon its plan for expanded use of the individual data. Second, if the ACS were not conducted under mandatory reporting requirements, but with proper informed consent, the types of followup surveys envisioned by the Census Bureau with all of the advantages cited could be done. There are a number of other potential solutions that I think others in the Bureau and we have talked about. Ultimately we believe that the Congress must decide how to achieve the appropriate balance in this area. We stand ready to work with our colleagues at the Census Bureau, OMB and other Federal agencies to arrive at an optimal system if this goes forward as planned. Mrs. Maloney [presiding]. Thank you. You have raised a number of important points. [The prepared statement of Mr. Kulka follows:] [GRAPHIC] [TIFF OMITTED] T3836.046 [GRAPHIC] [TIFF OMITTED] T3836.047 [GRAPHIC] [TIFF OMITTED] T3836.048 Mrs. Maloney. Next is Barbara Welty, president of National Center for Small Communities Board of Directors. Ms. Welty. Thank you, Mrs. Maloney, and good afternoon to you and to the chairman when he comes back. Thank you for the opportunity to testify again before the Subcommittee on the Census. Today I am here to emphasize the importance of the American Community Survey for rural communities. However, before I go into detail, I was going to give a background, but will shorten that because you have given some of that. NATaT represents approximately 11,000 towns and townships, mostly small and very rural. Although individual rural communities may be small in population, collectively they make up a significant and valuable portion of our citizenry. There are 36,000 general purpose subcounty local governments throughout the country; 82 percent of them have a population of 5,000 people or less. About half of all local governments have fewer than 1,000 residents. Because we are really a Nation of small communities, it is important that we ensure that the American Community Survey accurately represents the characteristics of those communities. The need for the ACS is clear. Community-specific, up-to- date data is essential for well-informed, long-term community planning. Knowing that the average age of a community's residents is increasing gives community planners warning that the community may need to have assisted care living, transit access to pharmacies and health facilities and other programs to help an elderly population. Similarly, an influx of younger families means that a community needs to budget adequate capital for schools and recreation facilities. Local governments also use demographic information to create long- term economic development plans to ensure the continued viability of these communities. The data assists in determining the infrastructure needs of their communities, including the maintenance and building of roads, sewers, shopping centers and libraries. Finally, data is necessary to determine eligibility and apply for State and Federal assistance. Many smaller communities depend on State and/or Federal assistance and require accurate and up-to-date demographic information to prove their eligibility. In sum, local officials are charged with protecting the health and environment and public welfare of their community residents. It is very difficult to fulfill these responsibilities without an accurate understanding of who is living in the community and what their needs are. By providing yearly refreshed data, the ACS will enable local government decisionmakers to remove some of the guesswork and insert hard evidence into their decisionmaking process. While the ACS promises to deliver an annually updated 5- year average of data for small communities, there have been concerns raised by data users about the quality and accuracy of data generated from surveying small communities. It should be pointed out that the Bureau has taken a lead in publicly examining some of the concerns raised, much to its credit. Some of the quality concerns that have been raised include: The difficulty of getting complete and accurate address lists in rural areas because of the very nature of rural addresses, i.e., post office boxes and rural route and box numbers; the difficulty of conducting personal interviews as part of a nonresponse followup in rural areas. Rural areas are expected to have lower mail response rates than more urban areas, and consequently, according to the Bureau, will produce less reliable data than areas where there is a high mail response rate; sample size and sample rate. Some of those who have reviewed the Bureau's methodology have argued that smaller jurisdictions should be sampled at increasingly higher rates to add precision and decrease disparities in sampling errors between differently sized governmental units. Before the design for the ACS is fully approved and signed off, the Bureau should conduct research into the best ways to mitigate these concerns that have been raised, and as before, such work would preferably be done in a public setting to allow a full airing of the concerns raised and solutions proposed. One other point I would like to make has to do with the cost of the ACS program. Some of the ways to make small community data more accurate may very well involve having to put more money into the ACS process. I know, Mr. Chairman, that you have made good on your promises to make every effort to ensure that the Bureau had the necessary funds to conduct an actual enumeration during the 2000 census. I believe that members of this committee must be equally adamant that the Bureau gets the funds necessary to generate ACS data from small communities that is as accurate as is economically feasible. This is a fundamental question of equity for small communities, which, again, are the vast majority of communities in the Nation. Although there is no NATaT policy that explicitly endorses the American Community Survey, it is clear from the attached resolution that the NATaT board recognizes the need for the specific demographic data that the ACS would provide. Indeed, the resolution suggests that the ACS can provide better quality data than the long form currently does. Thank you again for this opportunity to appear before the subcommittee, and I will be happy to answer any questions. Thank you. Mrs. Maloney. I would like to thank you very much for your testimony. [The prepared statement of Ms. Welty follows:] [GRAPHIC] [TIFF OMITTED] T3836.049 [GRAPHIC] [TIFF OMITTED] T3836.050 [GRAPHIC] [TIFF OMITTED] T3836.051 [GRAPHIC] [TIFF OMITTED] T3836.052 [GRAPHIC] [TIFF OMITTED] T3836.053 [GRAPHIC] [TIFF OMITTED] T3836.054 Mrs. Maloney. We will adjourn. We have a vote going on. We will be back shortly for questions. I would like to put in the record a statement of Jose Serrano who was supposed to testify today. [The prepared statement of Hon. Jose E. Serrano follows:] [GRAPHIC] [TIFF OMITTED] T3836.055 [GRAPHIC] [TIFF OMITTED] T3836.056 [GRAPHIC] [TIFF OMITTED] T3836.057 Mrs. Maloney. We will be back. We have a break for 10 minutes. [Recess.] Mr. Miller [presiding]. We will reconvene, and again I apologize for the votes on the floor of the House. Mrs. Maloney will be back shortly. Let's have a few questions. Mr. Hudgins, as a Libertarian, once the government has a program going, you almost have to have some--I know philosophically you oppose the existence of it, but once it is there, what do you do? Mr. Hudgins. That is a good question. First, you look to the Constitution, and you ask whether the Constitution gives Congress authority to do something in a mandatory way. Mr. Miller. Let's say it was voluntary. Mr. Hudgins. If it is voluntary, that opens up a whole different issue. In terms of getting data for programs, some of the American Community Survey might be an appropriate way if Congress is going to do it. I would also maintain that this is something that States and localities should bear some of the burden on. I am a Federalist, and I think in some cases the responsibility might be knocked down a couple of levels. Mr. Miller. The Founding Fathers didn't trust the States to do the census. Mr. Hudgins. That is absolutely right. The Constitution, Article I, Section 2 is quite clear that the Federal Government has the authority to enumerate people for the purposes of assigning electoral votes. No argument on that. There is nothing in Article I, Section 2 that remotely says that the Federal Government can mandate that I explain how many toilets that I have and some of the other questions. I am suggesting that those be voluntary or taken out of the census entirely; or second, we should ask the question about what data does the Federal Government really need, and what are the best ways to collect it. Frankly, I am very concerned, as I mentioned, there are no logical or legal limits, it seems, to what Congress can do. Therefore, there are only political limits to what kind of questions might be asked. I mention not facetiously that at some point I can imagine the Federal Government wanting to do surveys on people's diets and exercise habits because lots of people have serious problems in those areas. I don't downplay the problems---- Mr. Miller. I suspect that there are surveys, whether by HHS---- Mr. Hudgins. I was just speaking on a couple of these. Private companies do these surveys. Mr. Miller. That is good information. There is some informed decisionmaking that needs to be made. Mr. Hudgins. That's correct. Mr. Miller. We have school lunch programs and Head Start programs. Mr. Hudgins. One of my earlier points, because of the decline in federalism in the United States, in a sense the system itself pushes Congress to seek this kind of information. I don't blame Mr. Prewitt or the Census Bureau in one sense. They are simply following orders, and I say that suggests a reexamination of the system itself, and is there any logical conclusion or logical end to what you can or cannot ask. Mr. Miller. Mr. Fluharty, the question is on rural America. The long form, they went to the long form in a sample basis instead of purely random sample, a stratified-type sample in 1940. Mr. Prewitt. It was 1960. Mr. Miller. What was the perception in the rural areas? In the rural areas it was one out of two. What was the perception? You both recognize the need--or anyone else who wants to comment--the need for the information. What is the perception because of the long form invasiveness and privacy concerns? Ms. Welty. Mr. Chairman, as you said already, the need is certainly out in rural America. We need it so badly. We need that data to be able to do just about anything that we are going to do. The budgets are not available locally to be able to do most of the projects that rural America needs to do now, and therefore we do depend upon assistance for the Federal programs, and so we need that data. As to how it was interpreted, and I am assuming that you are meaning in this immediate census, I think the feeling was pretty much the same. Why do you need that, and we at the local level tried our very, very best to explain. We need to know how many toilets there are and all of those things because--and then follow it up with that wherever we had an opportunity to put that forth. The problem being, of course, is that we were not able to relate to a lot of public, and I think that is a real big issue that we need to do, whether ACS or long form or whatever. In 2010, we need to educate why those questions are being asked. Mr. Fluharty. Two quick reactions. Talking about rural America is just simply impossible. There is a sociologist who once said, ``once you have seen one rural community, you have seen one rural community.'' Rural communities are very diverse. There is no doubt there is a cultural imperative at work here, in some communities. I will not--I don't think that we should minimize the concerns that exist with privacy in rural areas. I also do not think that we should minimize the need of local citizen leaders in our democracy, for better data. It is critical to our democracy, Mr. Chairman, and I would fully agree, this Congress must decide what, in your wisdom, you will do. When you do, I would simply urge that rural jurisdictions in a decentralized Federal governance structure are not disadvantaged by access to good data upon which to make public policy choices. That is also an urban issue. The 50 percent of our country that lives in the suburbs have ability to access Federal funds. Rural America and central cities have part-time public servants, such as these panelists, who work with no framework for better decisions. A cultural imperative exists. For democracy to survive we need to address that issue. I urge that we do that, in whatever form we end up fulfilling this congressional mandate. Mr. Miller. I don't know enough about what the Agriculture Department does, but they do every 5 years a survey of farms or something like that. I don't know whether anything can be piggybacked or contracted for. Mr. Fluharty. I need to speak to that because it is a critical issue. My family has been in agriculture for six generations. Agriculture right now is approximately 2 to 6 percent of the rural economy. The programs of USDA minimally provide the needs in transportation, venture capital, health care, business capital and expansion and economic development. In this current structure, USDA agricultural data is being collected. But rural America is much, much more than agriculture, and the integrative links of private sector and public sector funding is key to making sustainable economies in rural America. The one other thing that I would say, it that it is very important for the Census Bureau and the ACS to work with elected and appointed jurisdictional leaders that are already doing this in communities, as they build these partnerships. That is going to let local governance get expressed in how ACS does its work. Mr. Miller. Does each county need a specific Bureau liaison? Counties get very small. Mr. Fluharty and Ms. Welty, how it worked this past decennial, do you have a comment between the relationship between the Bureau and how that can be improved? Ms. Welty. Mr. Chairman, yes. I am probably a little bit biased because I am on the Census Advisory 2000 Committee, so I think we have had a very good working relationship. Mr. Miller. How about the local level? Ms. Welty. Within our organization and within our State, those people have been very satisfied with the way it has worked. There have been some glitches, but overall it has worked. And having the representation--or I should not say representation, but the communication that we have had between the regional office and the State office has been sufficient. We do not have specific county census offices in the State of Minnesota. There are a few, but very few. It all works well. Mr. Fluharty. I would second all of those comments. I think there are nuances in terms of public-private linkage with organizations in place, whether it is county, townships, development organizations, community or private sector players. There are some unique models that were talked about, and I want to reinforce this one more time. When ACS brought together the rural community, I was very impressed with their ability to listen to not only community practitioners, but also statisticians, urban and rural, and rural scientists about how that works. I think that experiment in public-private linkage has to go forward as ACS moves out, acknowledging that the rural jurisdictions are a unique challenge. I think in general there is a good relationship. Mr. Miller. Mr. Kulka, my comment before is with ACS it could give a new potential, almost monopolistic power to the Census Bureau. You are very supportive of the ACS, but then you also express your concern about the competitive ability for organizations such as yours, and one that Dr. Prewitt used to work with and others. What can be done to balance the ability of--and to some extent we are not sure what new potential power ACS gives to the Census Bureau. Mr. Kulka. I don't have the ultimate solution, but I think it has to do with thinking about what this new vehicle is going to look like. For example, Paula Schneider on our recess mentioned the issue of title 13, and the mandatory reporting in title 13 is not really the issue, that there are nonmandatory title 13 issues. A lot would be about what availability of data are there, and how is privacy guarded within the system. In the radical sense if title 13 were amended to allow under certain circumstances the very data I was talking about not be shared to other government agencies for specific statistical purposes, which may be going too far on the pendulum, then you would have balance, and maybe have a more healthy balance. I don't know what is possible within the legislative system. Mr. Miller. Let me ask you about whether it is mandatory or not. This was discussed earlier. If it was not mandatory, it would be different. What impact do you think that it would have? Mr. Kulka. I think that the assessment is correct, that nonmandatory will reduce mail-back response. I think they will achieve virtually the same results at a higher cost. We do a lot of information. A lot of the surveys that we do are very important for the government. The decision about what data items, which is what you were talking about, are mandatory and which ones aren't is really a tough decision because a lot of things that have been envisioned as potentially added to the ACS are not very different than things gathered in a nonmandatory way now. So the decision point you are at now is exactly there. What are the ones that we want at the tract level, as you pointed out, Mr. Chairman, but which ones need to be mandatory and which ones don't, and which ones can be gathered under different mechanisms. Mr. Miller. If the mandatory label goes away, you also believe it would be more costly. The fact is on the census form there was a $100 fine, and that upset some people. You are confirming that response rates would be less? Mr. Kulka. I think they would be somewhat less. And we do and you all mandate, and we do in many cases other surveys which are not mandatory, and we provide data, and response rates are somewhat lower, and it is more costly to gather the data. But putting the mandatory label on everything, I believe, would cause a problem. Mr. Miller. Mr. Crowe and others, the question of corporate welfare raised by Mr. Hudgins and others, we are collecting information for the housing industry, and the government is giving you free information and not having to pay Mr. Kulka's organization to get that same information, or for rural America. Why should they not collect their own information and pay for it? Mr. Crowe. Certainly. I think one thing is that businesses do pay taxes, and they expect that as one of the services is some basic core information about the condition and location of houses in this country. Also I would argue that a good deal of what the housing industry ends up doing with the information from the census is establishing some sort of a level playing field with the local governments. Most housing construction is regulated by local governments. Zoning and planning and building codes and all of the issues that govern what is really built and where it is built are locally controlled. The local governments and the building industry have some meeting point on data that were collected and formed by an independent agency, the U.S. Census Bureau. The Bureau has no interest, if you will, in perverting that information. It is honest and straightforward data about where is housing and what does it look like and, therefore, what is the next step. Where do the new houses and transportation corridors--all of the other community infrastructure which cannot be decided if you don't have some good basic picture of how it exists right now. From that standpoint I guess the industry feels that it is part of the infrastructure of our country to have that basic information. Third, it is mandatory information because of Federal laws. So we are really only encouraging the Congress to continue that responsibility to collect information, to evaluate Federal programs. Mr. Hudgins. I just want to add if you see the census in the context of the growing concern for privacy, what you are also seeing is--especially on the Internet--is an emerging market for information and for different degrees of privacy because Internet entrepreneurs need information for marketing purposes and other things, and what you are starting to see is people setting privacy policies. You see people on Internet sites in a sense saying, we will buy your information, and people want different levels of privacy. Some things people don't care. Other things people are very sensitive about. One of the things that we are going to have to discuss much more broadly in our society is how these markets for privacy are evolving and what should be the government role in it, because we have to protect privacy. On the other hand, there are valid needs for businesses and government and for others to have information. I don't want the Census Bureau stepping in and in a sense messing up the market with mandates. Ms. Welty. Mr. Chairman, my earlier answer to one of your questions was that in rural America and very small communities we don't have the data available any other way, and if we don't have it because of the nature of rural America, and particularly against small communities, the money is not there, so we would have to continue forward without data if it doesn't come from the census. That is the very bottom line. Mr. Miller. States could do it, too. Mr. Fluharty. One example in terms of Medicare policy, the private sector was very comfortable to differentiate hospital reimbursement for the very same procedure from an urban hospital to a rural hospital. That would not have changed were there not government data which said we have made a commitment to provide services across space, and I think that nuance in working out the relationship in public information and what it means to be in a culture that values place is a critical charge which this committee has to address. Without that information, your colleagues would not have the data to make a more informed, equitable decision for rural customers. Mr. Miller. Mrs. Maloney. Mrs. Maloney. David Crowe, you mentioned earlier the various uses of the information, both for public purposes and for private purposes, on housing. I can't resist asking you about the plumbing question since it did become such a topic of concern across this Nation and actually in the House of Representatives and in the Senate, too. So for the benefit of all of us, could you give us some examples of how the plumbing question is valuable to the National Association of Home Builders and for both the public and the private? Mr. Crowe. Gladly, Congresswoman Maloney. I am so happy you tossed that one to me. Mrs. Maloney. We should have put you on the talk shows. Mr. Crowe. First of all, the census does not ask how many bathrooms you have. I have heard that repeatedly from a number of people. That is not in the questionnaire. Whether or not it is complete plumbing, whether or not there are the three components to make what is considered to be a decent house, and I think that is where the reason for the question comes from, at least from the private standpoint. It is a good indicator of more general conditions of whether or not this is a decent house or not. It is a specific question that gives you a more broad view of whether we are talking about really deplorable housing or housing that is adequate enough that it just needs some repairs. Therefore, it gives us some sense of what housing needs are in a community, what buildings are really not useful anymore that should be replaced. And it is a larger answer than just the simplistic response to whether or not you have plumbing. The same is true of the kitchen facilities. Those two together are the only low quality housing indicator in the census. Mrs. Maloney. I would like to ask Mr. Fluharty a question. Although I now represent one of the great urban areas in our country, probably in the world, I was raised in a rural area. You mentioned in your statement rural bias. Would you talk a little bit about this perceived rural bias? What is it? Has this issue been addressed by the Census Bureau, and could you elaborate further? Mr. Fluharty. I believe I said unintended bias, and it does get to my earlier question about assuring that public servants and private sector actors have accurate data upon which to make informed choices. The reality is if you look at any of our rural counties in the United States right now, the data that you will make public or private sector choices on is phenomenally wrong. When we work with counties or townships to do either siting work for plans, or a county decides to invest in infrastructure, the data that is there for rural jurisdictions will lead to bad public choices. The challenge we have is suburban and larger urban jurisdictions can hire consultants, and I think that is an excellent service. We need to acknowledge that most rural governments and jurisdictions don't have that option. So while it is an unintended bias, it leads to unwise, in many cases, public sector choices in jurisdictions at the local level. Mrs. Maloney. Did you raise this with the Census Bureau? Mr. Fluharty. Indeed. And as I mentioned earlier, I am really impressed by their willingness to address these issues. As I said, we had a very fruitful 2 days looking at how ACS might address some of those issues. It is the reality if democracy is going to work at the local level in rural America. Mrs. Maloney. Thank you. Richard Kulka, you raised a lot of interesting points in your testimony, and you mentioned a concern that plans to use individual data collected under a mandatory reporting requirement to select specific individuals for followup studies could be a serious violation of confidentiality and undermine the public's trust in the Bureau. Would you explain how private survey firms correctly use information collected on the long form in a way which maintains the confidentiality of the data? Mr. Kulka. There are sort of two parts to that answer. One, private firms do not have access to individual information, so it doesn't become an issue. However, all private organizations in surveying have an informed consent procedure where they say what uses of the data are, and we are subject to institutional review boards who review very carefully what we tell subjects, potential respondents, what we do. So we are not subject to that. The danger which is correctable is that in this new environment that you are talking about, the ACS environment, if you are going to tell people this is mandatory reporting, you are going to be penalized by law and all of the things that have been discussed here, because this information has legislative or other mandatory characteristics of the populations, as a citizen you need to provide it, then you--you then use that information to follow them up for another survey on maybe a very important issue that comes up a year later. The question is how do you do the consent which allows people to understand that you might follow them up for purposes that are nonmandatory? That is what I was trying to address. Mrs. Maloney. Thank you very much. My time is up. But, Ms. Welty, some who have reviewed the ACS plans believe that smaller jurisdictions should be sampled at increasingly higher rates to decrease sampling errors. Is that a position that you agree with? Ms. Welty. We don't have a position at this point. We just know that we want the ACS to continue. We would, of course--of course we would like to see it done to the smallest group possible. We realize that financially that is impossible, but we are willing to compromise to a workable size of the sampling so that we can get down to somewhat of a smaller community. Mrs. Maloney. Thank you. My time is up. Mr. Miller. Let me thank all of you for being here today. This is a first of, I am sure, a number of hearings that we will have over the next couple of years on this issue. I think we have some great opportunities here. We appreciate your input. We have received written testimony from representatives of the Urban Coalition and the Association of Public Data Users that I would like to enter into the record. Without objection, so ordered. I ask unanimous consent that the record remain open for 2 weeks for Members to submit questions for the record, and that witnesses submit their answers as soon as practicable. Without objection, so ordered. Thank you very much. This hearing is adjourned. [Whereupon, at 1:16 p.m., the subcommittee was adjourned.]