[House Hearing, 106 Congress] [From the U.S. Government Publishing Office] OVERSIGHT OF THE 2000 CENSUS: NON-RESPONSE FOLLOW-UP AND OTHER KEY CONSIDERATION ======================================================================= HEARING before the SUBCOMMITTEE ON THE CENSUS of the COMMITTEE ON GOVERNMENT REFORM HOUSE OF REPRESENTATIVES ONE HUNDRED SIXTH CONGRESS SECOND SESSION __________ MAY 11, 2000 __________ Serial No. 106-203 __________ 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 71-177 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 David A. Kass, Deputy Counsel and Parliamentarian Lisa Smith Arafune, Chief 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 Lara Chamberlain, Professional Staff Member Amy Althoff, Professional Staff Member Andrew Kavaliunas, Clerk Michelle Ash, Minority Counsel C O N T E N T S ---------- Page Hearing held on May 11, 2000..................................... 1 Statement of: Mihm, J. Christopher, Associate Director, Federal Management and Workforce Issues, U.S. General Accounting Office, accompanied by Robert N. Goldenkoff, Assistant Director, and Randolph C. Hite, Associate Director, GAO.............. 18 Letters, statements, etc., submitted for the record by: Davis, Hon. Danny K., a Representative in Congress from the State of Illinois, prepared statement of................... 13 Maloney, Hon. Carolyn B., a Representative in Congress from the State of New York: Letter dated May 3, 2000................................. 46 Memo dated May 11, 2000.................................. 11 Prepared statement of.................................... 6 Mihm, J. Christopher, Associate Director, Federal Management and Workforce Issues, U.S. General Accounting Office, prepared statement of...................................... 22 Miller, Hon. Dan, a Representative in Congress from the State of Florida: Followup questions and responses......................... 57 Letter dated May 9, 2000................................. 61 Redacted memo............................................ 55 Prepared statement of.................................... 3 OVERSIGHT OF THE 2000 CENSUS: NON-RESPONSE FOLLOW-UP AND OTHER KEY CONSIDERATION ---------- THURSDAY, MAY 11, 2000 House of Representatives, Subcommittee on the Census, Committee on Government Reform, Washington, DC. The subcommittee met, pursuant to notice, at 11:20 a.m., in room 2247, Longworth House Office Building, Hon. Dan Miller (chairman of the subcommittee) presiding. Present: Representatives Miller, Ryan, Maloney, Davis, and Ford. Staff present: Jane Cobb, staff director; Chip Walker, deputy staff director; Lara Chamberlain, Michael Miguel, and Amy Althoff, professional staff members; Andrew Kavaliunas, clerk; Michelle Ash, minority counsel; David McMillen and Mark Stephenson, minority professional staff members; and Earley Green, minority assistant clerk. Mr. Miller. Good morning. A quorum being present of the subcommittee, the committee will come to order and we'll begin with opening statements. Today we welcome the nonpartisan General Accounting Office. The GAO is the investigative arm of Congress. As such, it provides an objective assessment of a wide range of issues of concern to Congress. GAO's mission, as stated on its Web site says the following, GAO's mission is to help the Congress oversee Federal programs and operations to assure accountability to the American people. GAO's evaluators, auditors, lawyers, economists, public policy analysts, information technology specialists, and other multidisciplinary professionals seek to enhance the economy, efficiency, effectiveness, and credibility of the Federal Government, both in fact and in the eyes of the American people. GAO accomplishes its mission through a variety of activities including financial audits, program reviews, investigations, legal support, and policy and program analyses. GAO is dedicated to good government through its commitment to its values of accountability, integrity, and reliability. The nonpartisan GAO has been an invaluable resource to this subcommittee and our enormously difficult task of overseeing the almost $7 billion 2000 census. At this point, I would normally talk about the status of a range of Census Bureau operations. However, very late yesterday, my subcommittee became aware of a very serious matter that cuts to the heart of this census and severely calls into question the Census Bureau's credibility. The subcommittee received information that a mid-level Census Bureau manager had instructed, in an e-mail memo to other Census Bureau managers under his supervision, to intentionally keep information from the General Accounting Office. The memo states in part the following, ``I will try to get the D-333D report to you all on a daily basis. However, this report must and can not be shared with any GAO representative. This is a report that must not be shared with anyone else except the management staff.'' Let me repeat the relevant part of that paragraph, ``This report must and can not be shared with any GAO representative.'' To say I am shocked would be an understatement. I am appalled. The document in question is an update on the progress of local census offices during the very difficult non-response followup phase of the census. The specifics in this document, while important, are really not at issue here. The issue is a clear attempt to prevent Congress, through the GAO, of having access to the information. Is this the first time, or just the first time the Census Bureau has gotten caught? I take this obstruction seriously. I take it personally. This Congress takes it personally. Director Prewitt on numerous occasions has said this will be the most transparent census ever. This Congress has been assured of the level of professionalism and transparency in which this census would be conducted. This Congress has been told to trust the Census Bureau. In fact, whenever there was a suggestion that the Census Bureau would attempt to hide something from the Congress, it was summarily dismissed. How could Congress impugn the integrity of the Census Bureau, the defenders shouted. I don't for 1 minute believe that this mid-level manager decided on his own that he would instruct his local census office managers to withhold information from the GAO. Someone in a more senior position had to give this mid-level Manager these instructions. Whether this reaches beyond the regional census office and back to the Census Bureau headquarters is a question that must be answered. No stone be left unturned. There must be full accountability. While the vast number of Bureau employees are very professional, there are those, some in very influential positions, who have their own agenda. These people have no respect for Congress. They view us elected officials as meddlesome and they don't respect legitimate oversight. These people and this attitude are dangerous. The most politically sensitive part of this census has yet to begin in earnest. What documents have been, or will be, hidden from the bright lights of scrutiny? Today there is reason to be worried; very, very worried. Thank you. There is a copy of the redacted version of the e-mail memo. [The prepared statement of Hon. Dan Miller follows:] [GRAPHIC] [TIFF OMITTED] T1177.001 [GRAPHIC] [TIFF OMITTED] T1177.002 Mr. Miller. Mrs. Maloney. Mrs. Maloney. I just would like to put my prepared statement in the record, I don't even know where my prepared statement is, and just respond to your accusations. You know, first of all, we can't even talk to anyone on the West Coast since it's not yet 9 a.m. there. And I suggest that we get Director Prewitt here so that we can find out what his response is before we start making accusations. I would like to point out that this is edited in your statement. You add two words. You add ``is'' and ``not.'' And you quote from this memo but in fact this chairman's statement contains alterations. And I am asking my staff, and I would like my staff to hand out redacted copies of this memo and show that what the chairman is stating is not the entire facts. And I'd like the memo, before it was edited, to be placed in the record. Mr. Miller. It's already in the record. Mrs. Maloney. And it's not clear what this means. It says, however, this report must and cannot be shared with any GAO representative. This is a report that must be shared with anyone else except the manager staff. So he's talking, I assume, about the census when you talk about the management staff. But I think the important thing is to get the area director here who allegedly wrote this. And we know that area directors are people who are hired only a year ago. And to have a few people reporting to them--and how in the world could this be some type of vast conspiracy to hide information which the chairman is attempting to put out? I want to remind everyone that the last time that the GAO was here, when Christopher Mihm was here, he said that he was getting more information than he got in 1990. So my suggestion is that we suspend this hearing and call the director here and have the response of the individuals so that we know what we're talking about. You put forward some criticisms that I think should be answered. And the appropriate information is not here to have an accurate picture. We should have an accurate picture. We just got it this morning and we can't even call the area director because it's not even 9 a.m. in California. So I move to suspend this hearing until the director is called, until this afternoon, or however fast he can get over here. [The prepared statement of Hon. Carolyn B. Maloney follows:] [GRAPHIC] [TIFF OMITTED] T1177.003 [GRAPHIC] [TIFF OMITTED] T1177.004 Mr. Miller. This is a regular scheduled meeting that's been having full notice since the GAO was mentioned in this. I just found out about it early last evening. We have a number of issues we want to talk to GAO about other than this one. So I don't feel we should be suspending this hearing at this time but we should proceed to complete this hearing. And we will certainly be having another hearing where we can discuss this and certainly we will have a chance to discuss it. I called Director Prewitt, or we talked earlier this morning when I--to let him know about this--and we've shared this document, nonredacted version, with both you and with Director Prewitt in the Census Bureau. So we'll proceed. Mr. Ryan. Mr. Ryan. Mr. Chairman I wasn't planning on an opening statement but I think I will now. I wanted to hear the testimony from the witnesses, but you know we invited all these gentlemen up here to testify, I don't want to waste their time. I'd like to move forward with the hearing and hear the testimony. But let me point to what my colleague from New York was pointing at in a misleading comment. I'll read this paragraph. It's very clear there's just a simple typo in this paragraph. Here is the e-mail message from the mid-level census employee to the LCO administrators: I will try to get the DS--or the D- 333D report to you on a daily basis. However, this report must and can not be shared with any GAO representative. That's the original version. This a report must be shared with anyone else except for the management and staff. It's a typo. It says this a report that must be shared with anyone else except the management staff. All the chairman added into this was the grammatical correction with the proper grammatical corrections, which is, this is a report that must not be shared with anyone else except the management staff. It's obviously extrodinarily clear from the preceding sentence what the goal was, what the intent of the e-mail was, which was, ``However, this report must and can not be shared with any GAO representative.'' Why this is so alarming is because Congress is in the midst of overseeing the current census as it's unfolding right now. We're in the heat of battle right now. We're in the midst of sending enumerators to streets. In my area I'm very concerned about the people who have P.O. Boxes, didn't get a form, and we have to rely on enumerators going to their door to collect information from these people. We have towns throughout rural Wisconsin that never got a census form, that are hoping an enumerator comes by to get their census. So I've got some real concerns. I want to hear from the GAO on these points. Mrs. Maloney. Will the gentlemen yield for a question? The gentleman points out what he describes--the chairman's editing as a typo. When the chairman added ``is,'' this is a report that must--the way it reads, that must be shared. And the chairman added ``not.'' Must not be shared. But my question is this. My question is this. Seriously, why is not---- Mr. Miller. Have a straight face here. Mrs. Maloney. I do think this is rather humorous because in the sentence before, it says, however, this report must and can not be shared with any GAO representatives. In one case the chairman adds ``not'' and says it's a typo. I question you why is the first sentence not a typo and that ``not'' should have been out of that sentence. So I mean, I think it's very important that we get the chairman here. I move to suspend this hearing until we hear from the chairman. Mr. Ryan. Reclaiming my time. Carolyn, I have the time right now and I'll reclaim it. There's some other typos in here. I think the person who wrote this may have to go back for some grammatical lessons. But I will try to get the D-333D report to you all on a daily basis. That's not very good grammar right there. However, this report must and can not be shared with any GAO representative. That's pretty darn clear. ``This a report that must be shared,'' that's bad grammar. I think it's extraordinarily reasonable, extraordinarily reasonable, to assume that the person who wrote that sentence was intending to say this is a report that must not be shared with anyone else except the management staff. It's extraordinarily clear. I think anybody who looks at this can tell that GAO is being thwarted in their investigation, which is the auditing arm of Congress to investigate, to examine, to oversee the census as it is transpiring. That's what the role in Congress is doing. That's what the whole point of having GAO is. That's what oversight is, to oversee the census. We're at a critical time. So I think we should just, here with the GAO, move on with the hearing. But I agree with my colleague. Mrs. Maloney. I have heard from the director. Mr. Ryan. We should have the director here as well. Mrs. Maloney. I would like to---- Mr. Ryan. I yield back the balance of my time to the chairman. Mrs. Maloney. I would like to respond that when you are and the chairman are alleging some type of conspiracy to hide some information, all of the information contained in the memo is available to the GAO. And let me repeat that again: All of the information is available to the GAO, since the GAO has direct access to the Census Bureau's computers and management system. And since we have heard from the director I would like to read his response into the record and place it in the record. Mr. Miller. We'll place it in the record and make copies available. I think we need them. Mrs. Maloney. I think the director, since you have alleged concealing information, has the right to have his response read to the people in this room that are here listening to your misleading statements. Mr. Miller. Now, Mrs. Maloney, they're not misleading. Those are clear facts. But go ahead, I give you permission to read it. Mrs. Maloney. This is addressed to the Honorable Dan Miller, chairman of the Subcommittee on the Census, and it is carbon copied to myself. And it states it's from director Kenneth Prewitt: As per my conversation with you at 9:07 a.m. This morning, here are the facts: 1. There is no policy that exists at the Census Bureau that states information should not be given to GAO when requested. 2. On Tuesday evening at about 6 p.m., senior Census Bureau staff voluntarily faxed to senior GAO staff the check-in numbers along with workload of housing units completed, number of Assignment Areas and status of work in those AAs as of Monday night. No further data were requested by GAO. 3. I bring to your attention that all NonResponse Follow-Up workload check-out data, and more, exist in our real-time data base system known as Cost and Progress to which GAO has complete access. 4. The Census Bureau does not place expectations on lower and middle management of Field Operations staff to fully understand our standing policies with respect to oversight and access and therefore are tasked to not immediately process the request but to report to upper management when requests for specific data are made of them. So I'd like to place this in the record. Mr. Miller. No objection to putting it in the record. [The information referred to follows:] [GRAPHIC] [TIFF OMITTED] T1177.005 Mr. Miller. Two months ago we had a hearing that I was very concerned about access to the information both by the monitoring board, GAO, Inspector General, and our committee and staff. It's been a real concern when you receive a memo that very specifically and explicitly says do not share it with the GAO. I guess that's not important to you. I think it's a serious question, a serious problem if they're denying something to GAO. The data is not important because it shouldn't be that critical, what I see of the form. But the fact that someone--this is an area manager who is over a number of local census offices--would put this in writing, we need to look into it. Again, I found out about it early yesterday evening. I was in a markup and I found out about it. So we bring it up today since GAO was here. Mr. Davis, do you have an opening statement. Mr. Davis. Well, I have one. I'll just ask that it be inserted in the record. But it seems to me that we've had lots of conversation in the last few weeks about missing e-mails and things that can't be found and the manufacturing of memos, and it would just seem clear to me that anything that could be official would be that which is signed by the Director of the Census Bureau. It would be that which is communicated in a very direct manner to the committee, specifying that it's to the chairman and ranking member. Anything other than that, anything less than that, I'd call it hearsay. We don't know where it came from. We don't know whose it is. We don't know if it was authorized or if it was not authorized. We don't know anything about it. All we know is that it's something. So I would want to just take Dr. Prewitt's memo as the official position of the Bureau. And also I'd say it's inconceivable to me that any agency or any department of this government would attempt to hide something or prevent the Government Accounting Office--I mean that's--I mean that's like hearsay. Could you imagine that? Trying to prevent the GAO from having access to some information? I don't think anybody at any real level would attempt to do that, Mr. Chairman. So I'll just take Dr. Prewitt's memorandum as the official position on this matter. [The prepared statement of Hon. Danny K. Davis follows:] [GRAPHIC] [TIFF OMITTED] T1177.006 [GRAPHIC] [TIFF OMITTED] T1177.007 [GRAPHIC] [TIFF OMITTED] T1177.008 [GRAPHIC] [TIFF OMITTED] T1177.009 Mrs. Maloney. If you could yield to me. Mr. Davis. I yield to the ranking member. Mrs. Maloney. I think that it's important that we have Director Prewitt come and speak for himself. And I agree with my colleague, Mr. Davis, who expressed his--my sentiments, too, so well. And let's bring in the person who wrote the memo to see whether your editing is correct or incorrect. I'm sure they can edit their own memos and interpret their own memos. The way you edited it changes the meaning of it. So I feel that it's important that we hear from Dr. Prewitt. And I for one would stay over and later on today after this hearing, or come back tomorrow or Monday, but I think this needs to be addressed and we need to get the director in to hear his point of view and the area manager to interpret his own memo. I'm sure he can edit his own memo. Mr. Miller. This has been a long-scheduled hearing with the GAO and we're doing this on a monthly basis. We're going to proceed with the hearing. Mrs. Maloney. I'm for proceeding, but let's meet tomorrow too. Mr. Miller. We're going to be looking into it when we get all the information and the people can come to testify. We'll be having a hearing on clarifying this. But to start trying to, you know, say all of this is edited, it is very clear this report says this report must and can not be shared with any GAO representative. Now, I don't know if you can read it in clear English or not. Now, my statement added some grammatical corrections in brackets. In brackets. And I want to read this again so you can hear. Listen clearly, Mrs. Maloney. ``However, this report must and can not be shared with any GAO representative.'' To me, that's clear English. And I'm reading it directly. Now, I think we have a serious problem because this indicates something that's an unofficial hidden agenda. I know the official statement. And I'm glad that's what the official statement is and I hope that's the case. But if there's some hidden agenda that suggests some--is a pervasive problem out there, because we had a problem a couple months ago and we thought we had it cleared up. But this is very concerning to me and it should be very concerning to you because, as Mr. Davis said, if this is a policy, not sharing with GAO, there is something that cannot stand. We'll begin with the hearing. Mr. Mihm, if you would stand, and if Mr. Goldenkoff and Mr. Hite would please stand, raise your right hands. [Witnesses sworn.] Mr. Miller. Thank you. Please be seated. Let the record show that Mr. Mihm, Mr. Hite, and Mr. Goldenkoff answered in the affirmative. Let me officially welcome you. Mr. Mihm, you've testified a number of times before our committee and obviously have been working very hard in this and we appreciate that. Let me just, for everybody's background, Mr. Mihm is Associate Director for Federal Management and Work Force Issues at the U.S. General Accounting Office. Since 1993 he has managed GAO's efforts on the Government Performance and Results Act, the--and related results oriented management initiatives. Chris also is responsible for GAO work and public-private mix issues and the conduct of the 2000 census. He's appeared as a witness before Congress, congressional committees, on numerous occasions to discuss Federal management reform issues, and has been actively involved in working with committees across Congress to show them how GPRA can be used to improve congressional decisionmaking. Prior to assuming his current position, Chris managed GAO's reviews of the 1990 census that has identified the actions that the Census Bureau needed to take to have a more accurate and less costly census in 2000, and reviews the effectiveness of the Resolution Trust Corp., the Federal agency responsible for resolving the Nation's savings. We are fortunate to have somebody that was involved in the 1990 census, since this was your background and knowledge and we appreciate it. I believe you have an opening statement. STATEMENT OF J. CHRISTOPHER MIHM, ASSOCIATE DIRECTOR, FEDERAL MANAGEMENT AND WORKFORCE ISSUES, U.S. GENERAL ACCOUNTING OFFICE, ACCOMPANIED BY ROBERT N. GOLDENKOFF, ASSISTANT DIRECTOR, AND RANDOLPH C. HITE, ASSOCIATE DIRECTOR, GAO Mr. Mihm. Yes, sir, thank you, Mr. Chairman. And Mr. Chairman, Mrs. Maloney, Mr. Ryan, Mr. Davis, we are once again very pleased to be here before you today. I'm fortunate to be joined as usual by my colleagues, Randy Hite and Robert Goldenkoff. Overall, the initial Bureau data on the census is encouraging at this point for nonresponse followup. Major operations are reportedly proceeding on schedule and generally performing as planned. Particularly noteworthy, as you covered with Director Prewitt in the hearing last week, is the 65 percent initial response rate which, in matching the response rate from the 1990 census, surpassed expectations. While the overall response rate was very encouraging, the Bureau was unable to close the gap that has existed between the questionnaire response rates between the short and the long forms. As shown in table 2, and this is on page 5 of my written statement, during the 1990 and 2000 census cycles, questionnaire response rates were higher for the short form than for the long form, and this gap has widened over time. The 2000 census 12.5 percentage point differential response rate was twice that of the 1990 census. Also shown on page 5 of my statement is that the Bureau anticipated a 6.2 percentage point differential between the short form and long forms for 2000. However, the actual difference was much larger because the response rate to the short form was higher than anticipated, while the response rate to the long form was only somewhat lower than anticipated. The higher-than-expected response rate for the short form suggests that the Bureau's efforts during the 1990's to boost response rates by streamlining and simplifying the questionnaires, and in particular the short form, were largely successful. Local response rates are important because they determine staffing requirements as well as the scope and the cost of the Bureau's field followup operations. As of April 18, the response rates at the local census offices ranged from 39 to 80 percent, as seen on the board that's on our right over there. That's also on page 6 of my written statement. About 69 percent of the local census offices met or exceeded the Bureau's expected response rates for their type of office, meaning of course that 31 percent of the local offices did not achieve their expected response rate. This 31 percent represents 157 local census offices, generally covering suburban areas, small- and medium-size cities, towns and/or rural areas. Mr. Davis, I regret to note that the lowest mail response rate, unfortunately, was in Chicago where we traditionally have had a very difficult time taking the census in recent years. The Bureau's nonresponse followup workload is about 42 million housing units, which is 4 million fewer housing units than anticipated due to the high mail response rate. The Bureau has scheduled about 10 weeks to conduct this followup. As shown also on the next board on my right, however, you can see that the Bureau needed, in both the 1980 and 1990 censuses, more time to followup with fewer units. So you can see in 1980 and 1990 there was a much lower workload on followup than we have in 1990 and yet much more time was devoted to followup. According to a Bureau official, as of May 8 the Bureau had completed about 17.4 percent of its nonresponse followup workload, a very promising start. However, it's too early to tell if this pace will continue. And in doing so--and I think this is very important--the Bureau needs to ensure that it collects as complete of data as possible and limits the inappropriate use of proxy data. This is important because this is a problem that the Bureau has had in the past, both in 1990 and of course more recently during the 1998 dress rehearsal. That is, much higher rates of proxy data than was anticipated or desired by the Bureau. One factor that is clearly helping in doing followup is that the Bureau appears to be very well staffed. The Bureau has hired over 416,000 enumerators as of May 4 and the Bureau hired many more enumerators than open positions in part in anticipation of high turnover rates, which during the 1990 census were about in some cases 100 percent. Within the next few days, data should become available that will begin to show if turnover is a problem for the 2000 census as well. In addition to hiring a sufficient number of enumerators, officials at most of the local census offices we contacted believed that they have enough bilingual enumerators to followup with specific population groups, a key concern in enumerating traditionally hard-to-count populations. The Bureau's success in hiring is due in part to its keeping tabs on the progress of local census offices and taking quick action where it experienced recruiting problems. As we noted in our December report to the subcommittee, such monitoring of the recruiting process and responding rapidly to any difficulties was a key to addressing the Bureau's staffing requirements in this very tight labor market. For example, the Bureau responded to recruiting challenges at local census offices by increasing wage rates, including, as Director Prewitt mentioned last week, at the Tampa office. Since January, the Bureau has increased wage rates at 31 offices and, of these, 11 offices have had their wages increased since we last appeared before you in April. As we noted in our earlier work, higher wage rates help make the Bureau a more competitive employer when it encounters recruitment difficulties in such areas as seasonal resort areas. While hiring, too, appears to be going well, the early stages of nonresponse followup were not without some operational challenges. These challenges included a programming error that caused the omission of surname information from nonresponse followup address registers as well as several training and supply glitches. Let me touch on each of these briefly. First, in regards to the surname problem, as you know from Director Prewitt last week, the Bureau discovered that surnames had been inadvertently omitted from the nonresponse followup registers. According to the Bureau, this surname information is important to help enumerators collect data from housing units and situations where questionnaires may have been misdelivered in multiunit structures and in rural areas with clustered mail boxes. To remedy the omission, the Bureau decided to print supplementary address listings that contain the surnames which were then added to the address registers that had already been produced. Enumerators were then to receive training on how to most effectively use the surname address list. However, we found that in 8 of the 12 local offices where we observed enumerator training, that enumerators' training material did not include that supplementary surname address listing, and for most of the offices the trainers were not aware that they were to receive these listings and provide training on them. Later, of the 27 offices we contacted following our observation of training, officials at these offices said generally said that the Bureau notified them of the surname problem and they had taken action to inform their enumerators. Second, in regards to the training, trainers at the 12 offices where we observed training were generally prepared and used relevant examples of situations that they may encounter and how to handle those situations. For example, in the Los Angeles office, the trainers discussed how to handle language difficulties, uncooperative residents, and potentially hazardous situations such as vicious dogs. Nevertheless, at several local census offices, parts of the training were incomplete and key materials were lacking. My prepared statement mentions the problem at Las Cruces, NM, where a video on how to take enumeration was not available. Perhaps more important is that at 5 of the 12 local census offices we visited, enumerators did not get a chance to perform a practice enumeration with actual address registers because the registers were not ready. As a result, the enumerators missed an important opportunity to have on-the-job training and as a group discuss their field work experiences prior to working on their own. One factor that will add to the nonresponse workload is the need to followup on households that are on the Bureau list but does not receive a census questionnaire from the Postal Service, an issue that had been a particular concern of Mr. Ryan. The Bureau mailed out about 99 million questionnaires to housing units in mail-out/mail-back delivery areas. Of these 99 million, the Postal Service was unable to deliver about 11 million questionnaires. These are known, as you know, as undeliverable as addressed, more commonly as UAA questionnaires. And there were, as my written statement details, a variety of reasons as to why these UAAs may exist. Preliminary numbers indicate that of the 11 million UAAs, the Bureau has successfully redelivered about 1.6 million of these during the mail-back phase. The remainder are to be included in nonresponse followup unless the Bureau has other information that shows that those addresses were not accurate. Finally, let me briefly mention the status of the Bureau's data capturing processing. We are pleased to report that as of April 30, the census was processing questionnaires at a rate that will meet the Bureau's May 26 deadline for completing the mail-back questionnaire processing. In addition, the system development contractor has prepared a master plan and adopted an appropriate risk-based approach to modifying the hardware and the software configurations. And furthermore, the contractor is progressing according to plans. Important developments remain, however. Many more detailed supporting plans for those events have not been completed, but as I noted the Bureau and the contractor are taking appropriate steps. The key now is for the Bureau and its contractor to complete the plans and to continue to effectively implement them. In summary, at this early stage of nonresponse followup, the 2000 census appears to be generally on track. However, the Bureau recognizes that significant challenges lie ahead and as the Bureau continues its field followup efforts, it will be important for it to maintain staffing levels, maximize enumerator productivity, monitor the collection and limit the inappropriate use of proxy data, and quickly respond to operational problems. On behalf of the subcommittee, we will continue to track the progress, both here in Washington and throughout the country, that the Bureau and its local census offices are making in completing nonresponse followup. Mr. Chairman, Mrs. Maloney, Mr. Davis, Mr. Ford, this concludes my statement. My colleagues and I would be pleased to answer any questions you may have. Mr. Miller. Thank you Mr. Mihm. [The prepared statement of Mr. Mihm follows:] [GRAPHIC] [TIFF OMITTED] T1177.010 [GRAPHIC] [TIFF OMITTED] T1177.011 [GRAPHIC] [TIFF OMITTED] T1177.012 [GRAPHIC] [TIFF OMITTED] T1177.013 [GRAPHIC] [TIFF OMITTED] T1177.014 [GRAPHIC] [TIFF OMITTED] T1177.015 [GRAPHIC] [TIFF OMITTED] T1177.016 [GRAPHIC] [TIFF OMITTED] T1177.017 [GRAPHIC] [TIFF OMITTED] T1177.018 [GRAPHIC] [TIFF OMITTED] T1177.019 [GRAPHIC] [TIFF OMITTED] T1177.020 [GRAPHIC] [TIFF OMITTED] T1177.021 [GRAPHIC] [TIFF OMITTED] T1177.022 [GRAPHIC] [TIFF OMITTED] T1177.023 [GRAPHIC] [TIFF OMITTED] T1177.024 [GRAPHIC] [TIFF OMITTED] T1177.025 [GRAPHIC] [TIFF OMITTED] T1177.026 [GRAPHIC] [TIFF OMITTED] T1177.027 [GRAPHIC] [TIFF OMITTED] T1177.028 [GRAPHIC] [TIFF OMITTED] T1177.029 Mr. Miller. Yesterday, late yesterday, we e-mailed you a copy of this document. It originated from an area manager and was directed at a number of local census office managers. Could you explain what an office manager is and what is the significance of a memo of this nature coming from someone with this title? Mr. Mihm. The area managers are the critical link between the local census offices, the 511 of those, and the regional offices that are the permanent structures that the Bureau has established. These are often, in fact, generally temporary employees but have been relatively, in the context of the census, long-term temporary employees. They oversee between anywhere from half a dozen to as many as a dozen different local census offices. And I can tell you from my personal experience in 1990 and 2000 that these are positions of responsibility and authority. I have seen myself where local census managers appropriately defer to the judgment and instructions and guidance that they get from their local--from their area manager. So they are a key link in the responsibility chain that the Bureau has established to manage the census. Mr. Miller. Can you please give me your reaction and discuss the implications for GAO's efforts to effectively carry out your mandate to help oversee this census? Mr. Mihm. As you and Mrs. Maloney mentioned during your exchange on the opening comments, the actual form that is referred to here, the D-333D, contains information that we routinely get access to. For us the more important issue is the type of message that such a stark statement about ``you must not'' and ``must and can not be shared with any GAO representative'' sends. That is not typically something that we see and certainly do not like to see, and certainly creates an environment in which--it sends the message that there are areas that are off limits to discussion with the General Accounting Office on behalf of the Congress. That is not in our view a very healthy development and is not something that should be occurring. Mr. Miller. Have you shared this document with anyone else at GAO and are you at liberty to discuss their reaction or how GAO plans to address this serious issue? Mr. Mihm. We had--late yesterday after I received this, we--I discussed it with some of the senior management in GAO, one of the Assistant Comptrollers General. At the General Accounting Office this morning, we talked in more detail with the one of the associate general counsel about this, and they shared the concern that I had just laid out, that it is a disturbing statement both in the way it's written and its starkness. One of the things that's disturbing about it, Mr. Chairman, is that there's no context around it. It's not even as though it may you cannot and must not share this with GAO without approval or without clearance. Such language would be acceptable and within the normal bounds of the way we relate with the Bureau. The message that it sends that is disturbing for us is again, that there are areas in which it is inappropriate to be engaging in with the GAO, that we're some sort of oppositional force that needs to be cornered off. And notwithstanding the good relationships that we have with Bureau headquarters and the openness that we have with them, there is a vast network of temporary employees across the Nation. Having this message out there is, as I mentioned, is not a very healthy development, and that was a view that was shared at the senior levels. One of the things that I know in communications with your office late yesterday and even earlier today that they have asked us to be prepared for is to write to Director Prewitt, and ask him for any copies of any other memos or instructions or guidance that may be floating around that would comment in any way, either in a positive or negative way, about GAO and its access to individuals and documents. Mr. Miller. Well, I would like GAO to thoroughly investigate this matter. In your tenure of employment at GAO--how long have you been with GAO? Mr. Mihm. 17 years. Mr. Miller. 17 years. Are you--are you aware of any instances of where a Federal agency has made it their policy to withhold information from GAO? Mr. Mihm. In many cases, we have to work through access issues with agencies. And in my experience in the agencies that I've dealt with, I have not seen a statement as stark as this. Mr. Miller. What levels of access to Federal data is GAO entitled to under the law? Mr. Mihm. Our enabling legislation is fairly open-ended in that we, on behalf of the Congress, we can basically follow the Federal dollar wherever it goes. And so there is very little that we do not have access to both in terms of individuals and in terms of actual documents. Mr. Miller. Two months ago I raised at the hearing, at a hearing with Director Prewitt, concerns for access to information. I raised the question, what is he trying to hide? Because the monitoring board was having problems. We've seen a memo that was applying to the Inspector General, the--our staff and you. Now there were some meetings afterwards at the staff level and things, I thought, were moving in the right direction. How is your access and how has it been in the past couple months? Mr. Mihm. Well, as--let me split that into two ways. First, in terms of access to routine operational data of the census, because of--as I've mentioned before, because of your efforts and the efforts of this subcommittee, we were able to reach an agreement with the Bureau that gave us access to the computer systems that Mrs. Maloney mentioned in her statement. And so we do have access to a wealth of operational information that's available in the census, including much of the information that is in the D-333D--I think they need to get a more elegant name for that form. In terms of the field operation, it's always a stickier issue because we are taking people's time there and we are very, very sensitive to doing what we need to do in order to provide or assist the Congress in its effective oversight without being a burden on the census. But we are generally able to work through any of those issues that we've had with the Bureau. So there is no outstanding document or request or individual that we have asked for access and have not gotten access to. However, I should add, there are 10, 15 or 20 of us, depending on how it's counted, that are monitoring the census on behalf of the Congress, the GAO staff that are monitoring the census. We are dependent upon people in headquarters and in the field in being forthcoming with us and not just answering precisely the question that's being asked but understanding what the idea or the thought behind the question. We generally have that relationship with the headquarters--they're very willing to work through issues with us. That's a bit of what's disturbing about this memo, is the notion that there is something that would be out there, information that would be out there that would be known but not shared unless we by some chance ask for the very specific question to give me the D-333D. So that's the essence of why it's disturbing. Mr. Miller. Yeah. Obviously it's of concern if there's--I mean, what the policy out of headquarters is that Dr. Prewitt talks about is one thing. The concern is, is there some other hidden agenda--and that's obviously of concern to you. And someone in an area of management that is a high enough level, that is pretty serious. You deal with area managers. I'm not sure this is not one from the local area, but they know who GAO is. Mr. Mihm. I'm not sure how much they know how GAO is, but yes, sir, we do deal with them. When I, for example, when I was out observing the Bureau's enumeration of the homeless, I was accompanied by an area manager who was responsible for a number of LCOs. The LCO manager asked her questions about the operation and deferred to her judgment. Certainly the enumerators were asking questions of her while we were out and deferred to her judgment. As I mentioned, these are positions of some authority and responsibility. These are not--area managers--are not inconsequential in the Bureau management scheme. Mr. Miller. Well, this memo was just e-mailed yesterday so we only found out about it hours ago. So I think that you're going to investigate it, we're going to look into it, of course, and we want to get Director Prewitt's comments on that. So with that, you know, this is a very troubling development and I'm very concerned and we need to get to the bottom of this issue. So I now want to switch over to some questions about your report. But let me at this time switch over to Mrs. Maloney. Mrs. Maloney. Thank you, Mr. Chairman. It's good to see you again. Mr. Mihm. Thank you, ma'am. Mrs. Maloney. When did you, Mr. Mihm, get a copy of this memo? Mr. Mihm. Late yesterday. Mrs. Maloney. Late yesterday you got it. Mr. Mihm. Yes, ma'am. Mrs. Maloney. Did you ask census about it or did you contact them late yesterday? When you say yesterday, about what time? Mr. Mihm. Probably around 4 p.m. It was faxed to me so I can confirm that, but as I recall it was about 4 p.m. Mrs. Maloney. Did you call the Census Bureau and ask them about it yesterday? Mr. Mihm. Not at headquarters; no, ma'am. What we did, because we were still working on other things, including getting our statement up here to the committee, one of the things we did do is at 6 p.m. yesterday I called my colleagues that were out in the field--and at that time the only census regions that were still open were out in Seattle and Los Angeles--and asked them to request copies of the D-333D. Mrs. Maloney. OK. And did you inform the minority staff? Mr. Mihm. We talked to---- Mrs. Maloney. Democratic staff. Mr. Mihm. We talked to the majority staff. And let me go back also to the earlier question, when I came to the hearing, I made sure that that census officials that were here had received a heads-up on this; and in terms of the minority staff, we had talked to majority staff this morning to make sure that the minority staff would be notified of this, and then also talked with some of your staff when they came over early this morning. Mrs. Maloney. I would just like to point out to the public that I was not informed about this incident until 10:15 this morning. So I didn't hear about this incident until then. I would like to ask a few questions for the record about the information in the D-333D form which was attached to the memo the majority has circulated. It contains information about the progress of nonresponse followup at some local census offices; correct? Mr. Mihm. Yes, ma'am. Mrs. Maloney. And is this information--is this information that you would normally have access to? Mr. Mihm. It's through cost and progress; this is information that we have access to. It's workload, cases completed, cases uncompleted, percent done. It's fairly general; the generic operational status issue. So, yes, we do have access to this information. Mrs. Maloney. So you do have access. Access is not denied to you on any information in the D-333D. So how do you gain access to this information? Mr. Mihm. This is through the automated cost and progress reports that we obtained access to now a couple of months ago, as a result of the intervention of this subcommittee. Mrs. Maloney. And you get this information how quickly? Mr. Mihm. There's always a bit of a lag time. This information; the D-333D is generated daily out of the regions. There is a lag between when it comes in, and when it gets into cost and progress. The lag is generally 3, 4 days in some cases. Mrs. Maloney. So you have access. The longest lag time would be 3 days. Could you get it possibly within 1 day? Mr. Mihm. It's certainly possible. Mrs. Maloney. So it's possible that you have access to all of this information between 1 to 3 days and you have direct access. Mr. Mihm. Yes, ma'am. Yes, ma'am. Mrs. Maloney. OK. I think the facts speak for themselves on that one. I would like to place in the record a letter from controller David Walker. [The information referred to follows:] [GRAPHIC] [TIFF OMITTED] T1177.030 [GRAPHIC] [TIFF OMITTED] T1177.031 Mrs. Maloney. And in it he warns of the disastrous results for GAO of the 8 percent budget cut contained in the legislative branch appropriations bill which recently passed the subcommittee. And I will be sending a letter in opposition to this cut. And he warns that this is Draconian and that it would cripple GAO and seriously impair GAO's ability to oversee the executive branch, and that this is a serious cut. Would you agree? Mr. Mihm. Yes, ma'am, and not just because he's the Comptroller General. I know the status of GAO's budget is subject to very sensitive negotiations between the top of our agencies and Members and leadership up here on the Hill. But it is an issue of some concern for us. We work very hard to support the Congress in its oversight and decisionmaking matters. And I know the Comptroller General and all of us feel very strongly that the Draconian cuts that at least some are contemplating would cause real damage to our ability to support the Congress. Mrs. Maloney. I would conclude by stating that the proposed cut by the majority will do far more serious harm to GAO's ability to oversee the census and every other executive branch than access to the D-333D which allegedly some area manager said that you should not have access to something that you already have access to. The D-333D you have access to within a day, 3 days at the least. So the most serious effort to impair GAO's ability to look at the census and get the right answers is the majority's effort, which I hope to be part of a coalition in a bipartisan way to stop, to cut and hamper your budget. Mr. Chairman, I must note that this meeting was delayed by 1 hour due to votes on the floor. I have a bipartisan Women's Caucus meeting at this point with some of the leadership in your party, some of the leadership in my party, and I regret that I have to go to this meeting, and I will be back as soon as I can. I did not know that we would be delayed for an hour and I hope you understand. Mr. Miller. I understand. By the way, I do agree that, you know, we should not cut GAO to that extent. It's a process that we go through, this appropriations process. I think by the time the bill goes to the President, hopefully everybody will be happy because we--as you recognize, it's a very critical issue. Mr. Davis. Mr. Davis. Thank you very much, Mr. Chairman. Mr. Mihm, as you've been involved in doing your work in relationship to looking at the census, are you aware of any instances where you have attempted to receive data, information from the Census Bureau, and have not been able to get that? Mr. Mihm. No, Mr. Davis, our concerns have always been with the speed in which we're able to get the information. We have not had in my experience, now going back over a decade, a situation in which we have asked for something that we need and the Bureau has said no. Our concern is in getting it in a timely format that we can support congressional oversight. Mr. Davis. I know that we've had a great deal of discussion about the infamous memorandum. But is there any way to determine who the individual is or was or---- Mr. Mihm. I believe so. In the unredacted version there's an e-mail address. In discussions that I had with Bureau officials before this hearing they indicated that they were aware of the source. Mr. Davis. So it's going to be fairly easy to find out what the person had in mind or---- Mr. Mihm. I expect sir, yes. And following up on the chairman's earlier instructions, we will be talking to him or her and asking them what was their intent, what was the source of their anxiety about sharing something with GAO, what was the source of their instruction, if they felt they had some, to do that? And are there other documents or pieces of information out there that that individual has instructed census officers not to provide to GAO? Mr. Davis. You mentioned the discrepancy between return of the long form versus the short form. Mr. Mihm. Yes, sir. Mr. Davis. Has there been any attempt to just sort of cursorily determine why the difference? Mr. Mihm. Well, the general difference that we've seen, and as I alluded to earlier, is that the Bureau worked very hard throughout the decade, appreciating fully that they would have a harder time obtaining public cooperation this time, to streamline the census form and to simplify it. It's a much more user-friendly document as a result this time around. They also had the ambitious national ad campaign that we've discussed before, worked in partnerships with a number of local governments and other organizations. And so the Bureau was anticipating a much harder time obtaining public cooperation this time around. Then, in direct response to your question, the Bureau did much better on the short form response rate than they had anticipated and were marginally off on the long form response rate from what they anticipated, actually just a couple of percentage points. And so that shows a widening gap when in reality a lot of that gap is a function of how much better they did on the short form as opposed to a meltdown on the long form. Mr. Davis. Do you anticipate any further delving into trying to find out if there might be other factors contributing to the discrepancy? Mr. Mihm. Yes, sir. As we move forward both in the intermediate term and in the long term, as we begin to think about 2010, it is certainly in our minds what Director Prewitt mentioned at either the last hearing or the hearing before last, that he does not anticipate having a long form in its current form in the 2010 census. We'll be looking at how else can the Nation get that detailed level of demographic data. The Bureau's preferred option is American community survey. We'll also more generally be assessing how the partnership campaign worked, how the ad campaign worked, and others are doing similar assessments, to really see if we've turned the corner on public cooperation with the census. This census reversed a 3-decade decline in public cooperation with the census, and that's one of the more noteworthy stories that has come out of this census. Mr. Davis. I would hope that we'd be able to look at and try and determine whether or not some of the public directions that people were receiving from high-profile officials suggesting that they might not want to complete the long form, whether or not that had any impact on the level of response that we got. You mentioned Chicago being down. Mr. Mihm. Yes, sir. Mr. Davis. How far down? Mr. Mihm. As I mentioned, I regret to report that they are at the--the response rate for the Chicago west office was the lowest in the country, between--which was between 35 and 39 percent. Let me hasten to add though, Mr. Davis, that even though that was lower than anticipated, both given the type of office and historically--and certainly, I know, much, much lower than the Bureau hoped for--that is, like other census offices up and running on nonresponse followup, is reporting progress in addressing their workload, and is making progress toward completing the census. Mr. Davis. When we get down to further analysis, do you know if we're going to be taking into consideration differential community profiles so that information could be used in the early planning and that might prevent us from being in this position? I mean, I take the position that I expected it to be down in---- Mr. Mihm. Yes, sir. Mr. Davis [continuing]. Those areas, and especially given the fact that some of them are some of the most poverty- stricken areas in the country. And I think that there is a direct relationship between poverty and how people respond to some public activity such as filling out forms and that kind of thing. Mr. Mihm. Yes, sir. There is a traditional categorization, called hard-to-enumerate areas, which include certainly the areas that you're describing where there's severe and long-term poverty, areas where there may not be a lot of English-speaking residents, areas where traditional family structures may not exist, a lot of single-parent families. In addition, a lot of these factors overlap, of course, in many parts of the country. The Bureau has a data base on that and had instructed its local census offices to develop hard-to-enumerate plans. We're now gathering samples of those plans, going to be reviewing those and then following up with the Bureau, including in Chicago, to make sure that the special enumeration procedures that had been designed and intended for these hard-to-enumerate areas are executed and executed effectively. Mr. Davis. Thank you very much Mr. Chairman. I would have just hoped we could have made use of some of the information that was given earlier--that is, the Bureau in terms of the knowledge that some people who have worked with these communities have generated over the years--when we suggested that greater use be made of local-based community organizations to be involved in the process. Mr. Miller. At another time I would like to find out what kind of community support you had in Chicago, because I always thought Chicago had a very proactive--not only the government spent a lot of money, the city government, but I was hoping other organizations. Mr. Mihm. Mr. Chairman, if I could add 10 seconds on that. The Chicago Sun Times was one of the newspapers that--exactly on your point, sir--that had been most aggressive in both promoting the census and in giving census coverage. So one of the things that we will be doing is going into a number of communities where these partnerships seem to have worked, to try and develop a set of best practices as we move forward to the next census. Mr. Miller. For Chicago, you mentioned their rate being low. How did it compare to 1990? Mr. Mihm. The 1990 rate I don't know. Mr. Goldenkoff. I don't have that offhand. Mr. Mihm. We don't have that offhand. Mr. Miller. Before I call Mr. Ford, I just want to recognize someone who's with us here today and that's the cochairman of the Census Monitoring Board, Ken Blackwell. Nice to have you here, Mr. Blackwell. Appreciate it. Mr. Ford. Mr. Ford. Thank you, Mr. Chairman. Thank our witnesses our panelists for being here this morning. I too, Mr. Chairman, am concerned. I don't know if I'm quite appalled by the e-mail that we've learned about this morning. I just want to walk back through one more time with the panelists regarding this, because I want to ensure that you are receiving all of the information that you request. And I would even ask the chairman, if he feels free, if he wants to comment at all, to please feel free. Do you believe that it may be common with GAO--and I want to thank you all for all the work that you're doing and the response rate to the congressional offices and committees throughout this Congress and I'm one that believes your budget ought to be raised. I also hope my friends on the other side of the aisle don't cut the Capitol Police staff as well, and hopefully we can persuade them or dissuade them of that motion. But it is your experience and perhaps some of your colleagues' experience as you believed when they make requests of these agencies, for perhaps senior level people to actually have reviewed the materials before they send them to you? Would that be unusual, for instance, if you in your office, if you get a request for something and you ask a subordinate to followup and track down information? I know, at least in my office, when letters go out I want to take a look at them before they go out, or any correspondence from the office. I would probably say to my staff folks, as I do, You're not allowed to respond for me or the office until I get a chance to take a look at it. Because it's my name on the door. I would imagine that probably would not be uncommon at GAO, nor would it be--nor would you expect it to be unusual at a local census office. Mr. Mihm. It's---- Mr. Ford. I'm asking you to speculate, I understand. If we were in a court of law you wouldn't have to respond. Mr. Mihm. It's not uncommon. You're quite right. And especially in regards to the census, where you have a vast network of temporary employees, for headquarters to want to take a look at what's given to us before it's delivered over. And so that's quite common. Randy Hite, my colleague also deals with dozens and dozens of agencies. So, Randy, I don't know if you have a perspective as well. Mr. Hite. I could add one point and that would be it's tremendously variable across agencies. The notion of having requests for information reviewed at higher levels in some agencies like the FBI, for example, that is very, very common. And our experience in dealing with them is very similar to our experience in dealing with the Census Bureau. But I could name probably a dozen agencies where the decisions about whether to release information to us are made at a lower level. So it's going to be a reflection of the policy that the institution has established, the amount of authority that they want to empower individuals at lower levels to execute in dealing with GAO. So it's going to vary from organization to organization. Mr. Ford. I ask this, Mr. Mihm, fully understanding you may not be able to respond, but do you think it was--that the memo or the e-mail that we received is, or the contents of it suggest--or telling those of you at GAO and those of us in this Congress, because the language used by my chairman is somewhat strong--do you believe it's an attempt on their part to hide or conceal information from the GAO? And I know that the question was asked. It's been asked already. But I wanted to ask just a little more directly: Is this an effort or an attempt to--perhaps you see a pattern on the part of the census of not fully divulging and disclosing information that the GAO has requested at the behest of this committee and others here in the Congress? Mr. Mihm. The second half of your question, sir, is a little bit easier to answer as to whether or not it's part of a pattern. As I mentioned, we have not had access problems to the Bureau which we have not been able to resolve. I mean, this subcommittee and the chairman have been very helpful in making sure that we have the access that we need. In terms of whether or not this is an attempt to deny us access in this particular case, I can only conclude based on what it says, which is that we--that it's a report that must not and can not be shared with the GAO or with any GAO representative. The concern I was alluding to earlier is the chilling effect or the broader message that this may send. An area manager is responsible for several LCOs and for overseeing several LCOs. And this sends a pretty powerful message that there are areas that are out of bounds for the General Accounting Office and areas that you don't talk to the GAO about. And as I mentioned, Randy and I are amply supported by colleagues back at GAO, but we're still a small team. So we are critically reliant on people to be forthcoming with us and not walling us off or parsing what they say before they talk to us. If they do, it hampers our ability to support the Congress in oversight and is just not--it's not a healthy relationship to have with an agency if that's the way it's operating. Mr. Ford. I can trust--I would imagine, Mr. Chairman, we're going do have those before the census here sometime soon to explain this. I share your concern and desire to get to the bottom of it. I appreciate your responses, Mr. Mihm. Can I just ask one or two more questions, Mr. Chairman? I think I have just a little time left. One of the things I was concerned about when I was home during our recess recently, Director Mihm, I had a chance to meet with some of the enumerators after their--one of their first 2 days out in the community. And one of the questions was posed to those of us who had several of the regional directors and other local directors there and some other elected officials, there are those that were concerned about their safety because of the comments that had been made by some of my colleagues here in this Congress and even some of those running for other offices. Do you have any reports of hostility or refusals to answer questions, or doors being slammed in the face of enumerators? What type of data do you have on that? Mr. Mihm. We don't have any data on that sir. I talked to a very senior Census Bureau official about this issue directly yesterday afternoon and asked him if there was information suggesting both in a general sense if there was a response problem because of the attention the census has received this time around, as well as whether he was seeing particularly an escalating hostility over what they've experienced in the past. And he said that the information was not available. There's always some isolated incidences around the country, but at this point they appear to be isolated incidences, anecdotes rather than data. Mr. Ford. What type of training or other precautions are there for enumerators who may go into high crime areas, that you know of? Mr. Mihm. The Census Bureau has a verbatim training approach. So the training is the same that's given across the Nation for areas that are considered to be high crime or areas where there may be some sensitivity about the safety of enumerators. They will often do team enumeration. That is, sending more than one enumerator in. They have also in the past, and would certainly this time around, have used very targeted enumeration where they will go in and basically bring in a very large team of census enumerators and crew leaders and try to enumerate a particular neighborhood, all together, on one particular Saturday morning, for example. So that the Bureau is very, very sensitive and takes steps to protect the safety of their temporary workers. At the same time, they realize that you're dealing with millions of Americans that for one reason or another--in this case, over 40 million households--for one reason or another have already decided not to respond to the census. So there's always going to be some episodes of hostility that they run into. Mr. Ford. The mail response rate was, I guess, roughly 3 or 4 percentage points better than predicted. If I'm not mistaken, you might have suggested that the estimated costs of approximately $34 million for each point, which gives us--puts us at about $140, $136 million savings--and didn't go to one of the schools in Florida, Mr. Chairman, I can barely add--but I think about $136 million that can be devoted for other purposes. As much as I would like to see that money come to my great State, that the chairman has suggested, I think it's a good idea that perhaps we use this money for some of the hard- to-count areas. And I was curious to know if you thought that that might be a worthy idea or a noble idea. Mr. Mihm. Certainly, as Director Prewitt committed in his discussions with the chairman at the hearing last week, the Bureau will be using any surplus that may have accrued due to the higher-than-anticipated mail response rate in large part to help with the enumeration of hard-to-enumerate areas. He also cautioned that they have to take a look at how the productivity assumptions play out to make sure, first, that that surplus actually exists, and then the size of the surplus. But as we've reported before, the productivity assumptions were rather conservative that they used. Overall, the Bureau has committed to doing exactly what you're suggesting in large measure and we would endorse that. Mr. Ford. Just take 1 second, Mr. Chairman. I have some members from the Tennessee Homebuilders Association who are sitting back in the back and have been patiently waiting on me. I want to thank them for taking time out of their schedule to come over to our committee. This is important to our district, obviously, and some of the efforts as we look to expand our economy back home. And I thank you, Mr. Mihm, and really all of you for being here today. I can assure you that the capacity of this committee to get to the bottom of investigations or to investigate to make sure we get to the bottom of things is endless almost. So be assured we'll get an answer to this question for you. We'll get back to you soon. Mr. Miller. Thank you, Mr. Ford. One of your questions, your last question, I was agreeing with you. As you know, I sent a letter to the director that we should program this money to go to hard-to-count areas in Chicago or wherever. Mr. Ford. As long as you don't send it to the Florida football team. Mr. Miller. I have offered it, since I sit on that appropriations subcommittee, to help any way we can if we need to do any reprogramming. Director Prewitt said it wasn't necessary. I think we are going to save some money; that, as I said before, we want to do everything we can to count everybody in this country and especially in those hard-to-count areas. So I'm glad we have some more flexibility, whether it's more advertising, hiring more enumerators, whatever possible. I want to make sure, by the way, that we have officially included in the record the memo on this, the redacted one that removes individual names. So, without objection, I'll include that. [The information referred to follows:] [GRAPHIC] [TIFF OMITTED] T1177.032 Mr. Miller. I also want to--Mr. Ryan had to leave before the questions and he had some questions which, as you know, we will submit questions in the next 2 weeks, but I want to make sure that I include Mr. Ryan's. He had some concerns about post office box issues. And so if you would respond to those and make those part of the official record. Mr. Mihm. Absolutely. [The information referred to follows:] [GRAPHIC] [TIFF OMITTED] T1177.033 Mr. Miller. As you say in your testimony and show in the charts, that despite the better-than-expected mail response rate, the Census Bureau still is attempting to count more people in less time during the door-to-door phase than they did in 1990. Is GAO seeing any evidence of pressure for enumerators to get out of the field early, and if so, what are the implications of this? Mr. Mihm. First in the thesis of your question, sir, as you can see from the chart, it took the Census Bureau about 14 weeks in 1990 to enumerate 34 million nonresponse households. They're now planning to do 42 million in about 10 weeks. So there's a much greater effort that they're expecting this time around. They have many more people that are hired and on the ground which is certainly helpful. The Bureau sets and often attains very ambitious goals for nonresponse followups, schedule completion. As I mentioned, within the first 10 days or so, they're at 17 percent completion rates. And that's a very promising achievement. The concern, of course, is always to make sure that there's not a cost that comes along with this. We saw during the dress rehearsal that the Bureau got out of the field in some locations on time and other locations actually early. And there was a bit of a celebration on that. At that point, and in front of a hearing in the Senate, I cautioned that before we join that celebration, we needed to see data on proxy. And that is the extent to which we were getting nonhousehold data. Indeed it wasn't until much after the fact that we found out that the studies were showing that the amount of proxy data was much, much greater than the Bureau had hoped; in some cases double the 6 percent nonresponse universe of proxy data. That's always a concern. It was a concern in 1990 when they had high levels of proxy data. And it gets back to the point, sir, that you and Mr. Davis have been raising about hard-to-enumerate areas. West Manhattan in 1990 had something like a 42 percent proxy rate for their nonresponse universe. And so basically approaching half of the nonresponse workload that in one LCO in a hard-to-enumerate area, that was enumerated using proxy data. Data from 1990 also shows that African Americans are enumerated at higher rates using proxy data than White Americans. So this isn't just a general issue of data quality, the proxy data; it cuts on lines of the differential undercount and the overall accuracy of the census. Mr. Miller. Well, proxy data, they're supposed to make six attempts, three in phone and three in person, to try to contact that household. And then they start talking to neighbors or whatever other source they have. What is an acceptable level? We've talked about this before but, you know, we're going to be approving that here in the next month or two. I mean, what are acceptable levels of proxy data? And at the specific local census office area. Mr. Mihm. It's hard to say. I wish I could give you a definitive answer on that. The Bureau's goal is traditionally to keep it at or below 6 percent of the nonresponse universe. Obviously in some or perhaps many offices they reach that. However, there are more than a few offices in which they do not reach that in which it balloons up quite significantly above that 6 percent nonresponse universe. And as I mentioned, it gets to be a particular concern when it's in offices that also have the hard-to-enumerate populations because you just aren't having proxy data but you're having proxy data disproportionately in areas where we've had traditionally highest undercounts. Mr. Miller. Just a couple of questions about money. The budget predicted the 61 percent mail response rate and follows along with Mr. Ford's comments. Director Prewitt had said earlier that they could sustain a lower-than-expected response rate of, say, 58 or 59 percent without coming back to Congress for additional funding. With the significantly better-than- budgeted short form mail response rate and a long form response rate only 1.8 percent rate lower than expected, do you believe the Bureau should be able to cope with this shortfall? Mr. Mihm. There is---- Mr. Miller. Money. Mr. Mihm. The short answer is yes. Nothing that we've heard from the Bureau suggests that the incremental amount that they were off on their anticipated long form mail response rate causes them undue or severe difficulties. In fact, I think one of the things that Director Prewitt has pointed out is that at least some of the surplus from the higher overall mail response rate may have to go to addressing if they have lower productivity in the field, because they have more-than- anticipated long forms to followup on. So it should not be a telling problem for them. Mr. Miller. Let me--I was talking with Congresswoman Meek this morning, and Congresswoman Meek and I introduced legislation, which we ran into some roadblocks in different areas, and that was to allow people to work for the Census Bureau that may need to get waivers. You make a statement in your report--I was glad to hear that, but would you expand upon it--in the hard-to-count areas to get people to work for us, but not affecting their Medicaid or without affecting their temporary assistance for needy family programs. I see a large number of States, according to your report, have made, you know, waivers and such. Would you expand upon that? And we've talked about it before, but is the ability to hire people in the hard-to-count areas, whether it's the language issues or people in some of Mr. Davis' very difficult- to-count parts, would you expand about the hiring abilities there and what the States have done? Mr. Mihm. Yes, sir. Dealing with the second part of the question first, the ability to hire in each neighborhood is very important for the Bureau. They've really put a great deal of effort into that. They recognize that people want to enumerate in neighborhoods that they're comfortable with and people want to be enumerated by people that they're comfortable with. So it's not a matter that if they're having a recruitment problem in an inner city area or a suburb that you can just kind of parachute in a bunch of enumerators from other parts of the city or area. The Bureau will do that as a very last resort if they have to. But their clear preference is to get people from local areas. As we mentioned in the written statement, the Bureau has continued to work with State governments in order to get them to grant exemptions. And since our March 14th testimony there is--New York, South Carolina, Virginia, Indiana, Kansas, California, Alaska, Idaho, have all enacted State legislation providing some sort of relief from State requirements for people on assistance to work on the census. Interestingly, also the Bureau data is showing since 1997 over 15,000 welfare-to-work hires have come to work on the Census Bureau for the census. And so there's a real effort that the Bureau makes to get people from the local communities. And as part of that, they have been working very hard, consistent with the legislation that you and Mrs. Meek have had pushed in order to get exemptions from any disincentives that may be out there. Mr. Miller. So the overall evaluation of hiring enumerators in the localized hard-to-count areas, how is that going? I mean, Director Prewitt said it is going well. Mr. Mihm. Yes, sir. Largely it's going quite well. Nationally they are front-loaded, and that is, they hired more people than they actually had positions for. So they're splitting the positions actually into kind of two different positions. They have been successful in virtually all areas of the country. In the discussions that we've had with local census managers they are quite comfortable that they are able to get enumerators and other office staff with the proper language skills for that. In one area in Chicago, for example, a local census manager was telling us that there was Lithuanian, Polish, Hispanic, and Chinese neighborhoods, and were able to get individuals that could cover all of those neighborhoods. So they are able to get people with the proper language skills at this point. As I mentioned, it is within the next few days that we will begin to start seeing some of the first turnover data and that will be critical to telling all of us--and obviously including the Census Bureau whether turnover is at historically high rates at some of these hard-to-enumerate areas. Mr. Miller. I want to enter something else in the record and then call on Mr. Davis. At last week's hearing, Dr. Prewitt testified that a Bureau employee mistakenly faxed a document listing the names and Social Security number of job applicants to the home of a private citizen instead of another census office. The private citizen, a constituent of Representative Tom Coburn, then turned over the document to the Congressman. It was insinuated in the hearing that Congressman Coburn might have distributed the document to the media. That was contrary to my understanding of the events and I stated so at the time. Since then, Congressman Coburn has written me a letter detailing exactly what happened. His letter, addressed to me personally, says, ``I want to assure you that I have not distributed the document to anyone in the media. I permitted only one reporter to see any portion of the document. I was careful to conceal its confidential contents. The reporter was allowed to merely confirm that the document was in fact from the Census Bureau.'' I would like to enter this, his entire letter, at this time into the record. Without objection, it will be entered. [The information referred to follows:] [GRAPHIC] [TIFF OMITTED] T1177.034 Mr. Miller. Mr. Davis. Mr. Davis. Thank you very much, Mr. Chairman. We've indicated that in spite of there being some areas, especially inner city hard-to-count areas where the response is less than expected, there have been others where the response has been better than expected, and overall it is my understanding that the response has been better in those areas than expected. Are you aware of any effort to take a look at the different experiences, coming from areas that were similar, to look at what may happen or what might have happened in one that did not happen in another? Mr. Mihm. Yes, sir. In fact, we're doing an evaluation of that right now. We'll be in the field shortly. Looking at a selection of communities that had much higher-than-anticipated mail response rates, both higher than anticipated this time around and also as measured by doing much better than they did in 1990, the Detroit and surrounding area seems to have done very well. One of the things that I should have mentioned earlier is that one of the interesting stories that's coming out of the census this time around is that when you look at the expected mail response rates adjusted by type of office and type of area, actually the urban areas and the hard-to-enumerate areas within urban areas did better than expected this time around, even though in an absolute sense they had lower mail response rates. Where the Bureau did not do as well as expected are in some of the suburban and nondensely populated areas including some of the more rural areas. And so that's the challenge. I don't know if, Robert, if you have anything in particular that you want to add to that. Mr. Goldenkoff. No. Mr. Davis. I know that there's the 10-week schedule to followup on the nonresponse. Are we aware of any conversations in the planning in the efforts that this might be lengthened in some of the areas that were experiencing particular difficulty? Mr. Mihm. Yes, sir. The Census Bureau in both their operational plans, and Director Prewitt personally to this subcommittee in other statements that he has made, has committed that he will not leave the field until he's done in any particular area. The 10 week is certainly a planning deadline and that they are doing everything they can to achieve. But the last time in--the last time being in 1990, in parts of New York City they did not close out until August. And, you know, while no one at the Bureau says they anticipate being in the field in August, if that's what happens I'm quite certain that they will continue to enumerate as long as they need to in the field. Mr. Davis. Do you know if there's a projected cutoff level? I'm saying if there is any determination that, say, we've gotten up to an 85 percent response--which is OK, you know-- we've--this is about as well as we're going to do, we've done everything that we can think of, that we're going to cut here and move ahead? Mr. Mihm. There better not be. The closest thing that they have to your question though, sir, is that the Bureau does have what they call closeout procedures. Closeout is done not on a whole office level, it's done on parts of offices as the enumeration in each area covered by a local office reaches the 95 percent threshold--then they will make one final attempt at the final 5 percent irrespective of whether or not they've been to them six times or not. And so they will make one final attempt and get as complete data as possible. And if they can get complete data from the household, great; if not, they get it from proxy. And if they even can't get it from proxy, then they have algorithms which impute the persons and characteristics of those persons. But that's the 95 percent threshold. Mr. Davis. So communities can be pretty comfortable that at least up to that point, there's going to be all of these efforts put forth to try and make sure that that threshold has in fact been reached. Mr. Mihm. The Bureau procedures certainly call for them to do that. And we will certainly continue to be monitoring that both at the headquarters level and, importantly, all across the country including sir, Chicago, to make sure that those procedures are actually followed. Mr. Davis. Well, I thank you very much. Mr. Chairman, I also want to just thank you for the position that you've taken relative to the utilization of unexpended resources in areas that may in fact have some difficulty in counting. I think that's the only way to really go in terms of trying to make sure that across the board we get an accurate, or as accurate account as we can, from all of the people in this country no matter what type community they may live in. Mr. Miller. Thank you. I agree. You know, it's always been the challenge from day one to make sure we get the undercounted completely counted. I have two short questions. Does GAO have a plan to evaluate performance of the local census office level and, if so, would you describe the plan and the criteria you would use to look at performance? Mr. Mihm. Yes, sir. There's actually a several-pronged approach that we have. We've been working to develop a list of local census offices as a subset of the 511 that are those that are likely to be most challenged in taking the census based on criteria such as the hard-to-enumerate populations, mail response rate, progress in recruiting, and now progress in staffing and possible turnover. We've had a first set of contacts with 27 of those offices. We anticipate further contacts. We are certain that we will be making further contacts with them and will be actually planning to be on location as these offices get toward the end of nonresponse followup so we can monitor personally and directly the closeout procedures that Mr. Davis and I were just discussing. Separately, what we'll be doing is a sample interview of LCO managers. We realize the managers are an incredible resource that's only there for a short time. Once the census is over these people go away and we've lost the ability to tap their knowledge about what worked well, what didn't work well, what are some of their ideas for the next census. These are the people that have been living, breathing, and dying with the census over the last few months. So we'll be interviewing a sample of those managers. Mr. Miller. Thank you. I hope you all have visited your local census offices. I'm impressed at my local census office, the local manager, and the assistant, and seeing their operations. So if you haven't, I would encourage you to make a chance just to stop by and visit. I've been impressed. Because you're right; these are people that are not going to have job in a few months, at least with the Census Bureau. And I think they do a very good job. Let me ask one final question and this is a question I've asked--a couple Members have asked me--so let me just clarify it. And that is, that after the deadline for the forms to come back, they're still coming in. And we talked about it last week with Director Prewitt. Would you explain to me what happens to those forms? Because Members are telling me, well, we're having people say, well, I sent my form in, even though it was late, and they still insist on asking those questions. So you may have that same thing. Would you explain that issue? Mr. Mihm. Yes, sir. The reaction that some of the constituents are giving to your colleagues is a fortunate by- product or perhaps or maybe unfortunate by-product of a fortunate situation. And that is that the mail response to the census continues to trickle in. I recall just a couple of weeks ago, or I guess it's less than that, a week or so ago, seeing the beginnings of a census add on TV and slapping my head going oh, my gosh, what's this doing now, the mail response form is-- or mail response portion is over--and then hearing the tag line which was, cooperate with the Census Bureau employee, if you haven't responded, when they come to your house. So in other words, the census is continuing to advertise, they're still up on the air. Now the focus is telling people to cooperate with the census takers. The effect of that, at least in part, is that they're still getting census forms mailed in. So what happens as a result is that these people are too late to be deleted from the nonresponse workload, so they will have a census enumerator go to their house. The census enumerator will explain to them, ``I understand that you said you've sent in your form, I still need to get a form from you.'' The reason the Census Bureau requires that is that they can't be assured that they get every form that's mailed. They can't be assured that everyone is, to put it delicately, recalling correctly that they mailed back the form. So they will still try and enumerate these individuals. They then have a series of procedures, algorithms, that they go through to determine the more complete form and the form that the census will actually accept. When they get an enumerator form versus the mail-back form, generally it is the mail-back form they would accept. And after that, it deals with the more complete form and higher popultion count. Mr. Miller. Is it not feasible to kind of update their lists if they get some more in the mail in late April or something like that? Mr. Mihm. They did it once. The initial mail response rate, or I guess the ``preinitial'' or the first cut of the address list for nonresponse followup, was April 11th, what is now defined as the initial mail response list was the list on April 18th. That was then sent to local census offices where they were to have marked through on the enumerators' registers. The Bureau---- Mr. Miller. It's not feasible to do it anymore. Mr. Mihm. That has been the Bureau's conclusion. We haven't looked at it and see if they could do that, sir. Mr. Davis. Well, it's just interesting, that line of conversation, because on Saturday of this week, a group of elected officials and their volunteers are going into the area that we talked about earlier, simply asking people to cooperate with the census takers and asking them to be available and be ready to not close the door, to not freeze them out. And right now there's a decent level of excitement about it. And we're looking forward to seek, you know, the extent of the impact that it will have. And we've had that kind of cooperation certainly from the local census people in the area. It hasn't been any difficulty at all getting information, working with them, and we hope this is going to generate---- Mr. Mihm. Mr. Davis, if I can suggest that the experience from 1990 and the experience from here in 2000 already suggests that those types of messages sent by community leaders, including elected leaders, ministers, and others, are very, very powerful. There is no substitute for community leaders telling people that it's not just appropriate but it's important to the community to respond to the census. There is no substitute for that. Mr. Miller. Anything else Mr. Davis. Mr. Davis. No. Mr. Miller. Mr. Ford, do you have anything else? Mr. Ford. No, Mr. Chairman, I don't have any other questions. I would only thank the panelists again and can assure them that I would imagine we're going to ask some folks to come before us from the census to answer some of these questions. Just one last thing, I'm sort of getting explained to me this chart here. I'm a little confused by the last number, or I should say the last column there, or the tallest one, the furthest--the closest to you on the right there. If the 23--or the 1980 number, the 1990 nonresponse followup, can you just explain to me why it goes down like that, maybe a little slow. Mr. Mihm. Why the timing goes down? Mr. Ford. Right. Right. Mr. Mihm. The schedule that the Bureau set calls for them to be able to enumerate these 42 million households in a 10- week period, where it took them 14 weeks to do this back in 1990. And that's the line that slopes down there. They believe that they can do this through additional hiring that they're doing this time around. And the early data certainly suggests that they've got people on the ground. As I mentioned, they have 460,000 people that are out there knocking on doors. They believe that they can do it through better targeting or faster action in areas where they are having either recruitment problems, including raising the pay rates, which we have suggested is important when they have staffing problems. They believe that they can do it through better and closer management of local operational problems. All of those are going to be key in order to making this very ambitious timeframe. Mr. Ford. Thank you. And, Mr. Chairman, I yield back whatever time. Mr. Miller. Thank you. Thank you again for being here today. I appreciate it. And you know, we really rely on you providing that objective nonpartisan information in oversight on this very critical issue. I ask unanimous consent that all Members' and witnesses' written opening statements be included in the record. Without objection, so ordered. In case there are additional questions the Members may have for our witnesses, I ask unanimous consent for the record to remain open for 2 weeks for Members to submit questions for the records and that the witnesses submit written answers as soon as practicable. Without objection so ordered. Meeting adjourned. Mr. Mihm. Thank you, sir. [Whereupon, at 1:05 p.m., the subcommittee was adjourned.]