[Senate Hearing 112-676]
[From the U.S. Government Publishing Office]






                                                        S. Hrg. 112-676

                    CENSUS: PLANNING AHEAD FOR 2020

=======================================================================


                                HEARING

                               before the

                FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT
                   INFORMATION, FEDERAL SERVICES, AND
                  INTERNATIONAL SECURITY SUBCOMMITTEE

                                 of the

                              COMMITTEE ON
               HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS
                          UNITED STATES SENATE


                      ONE HUNDRED TWELFTH CONGRESS

                             SECOND SESSION

                               __________

                             JULY 18, 2012

                               __________

         Available via the World Wide Web: http://www.fdsys.gov

                       Printed for the use of the
        Committee on Homeland Security and Governmental Affairs



[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]








                  U.S. GOVERNMENT PRINTING OFFICE

76-063 PDF                WASHINGTON : 2013
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing 
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC 
area (202) 512-1800 Fax: (202) 512-2104  Mail: Stop IDCC, Washington, DC 
20402-0001




        COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS

               JOSEPH I. LIEBERMAN, Connecticut, Chairman
CARL LEVIN, Michigan                 SUSAN M. COLLINS, Maine
DANIEL K. AKAKA, Hawaii              TOM COBURN, Oklahoma
THOMAS R. CARPER, Delaware           SCOTT P. BROWN, Massachusetts
MARK L. PRYOR, Arkansas              JOHN McCAIN, Arizona
MARY L. LANDRIEU, Louisiana          RON JOHNSON, Wisconsin
CLAIRE McCASKILL, Missouri           ROB PORTMAN, Ohio
JON TESTER, Montana                  RAND PAUL, Kentucky
MARK BEGICH, Alaska                  JERRY MORAN, Kansas

                  Michael L. Alexander, Staff Director
               Nicholas A. Rossi, Minority Staff Director
                  Trina Driessnack Tyrer, Chief Clerk
            Joyce Ward, Publications Clerk and GPO Detailee
                                 ------                                

 SUBCOMMITTEE ON FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT INFORMATION, 
              FEDERAL SERVICES, AND INTERNATIONAL SECURITY

                  THOMAS R. CARPER, Delaware, Chairman
CARL LEVIN, Michigan                 SCOTT P. BROWN, Massachusetts
DANIEL K. AKAKA, Hawaii              TOM COBURN, Oklahoma
MARK L. PRYOR, Arkansas              JOHN McCAIN, Arizona
CLAIRE McCASKILL, Missouri           RON JOHNSON, Wisconsin
MARK BEGICH, Alaska                  ROB PORTMAN, Ohio

                    John Kilvington, Staff Director
                William Wright, Minority Staff Director
                   Deirdre G. Armstrong, Chief Clerk

















                            C O N T E N T S

                                 ------                                
Opening statements:
                                                                   Page
    Senator Carper...............................................     1
    Senator Brown................................................     3
    Senator Coburn...............................................     4
Prepared statements:
    Senator Carper...............................................    47
    Senator Brown................................................    49

                               WITNESSES
                        WEDNESDAY, JULY 18, 2012

Hon. Robert M. Groves, Director, U.S. Census Bureau, U.S. 
  Department of Commerce.........................................     5
Hon. Todd J. Zinser, Inspector General, U.S. Department of 
  Commerce.......................................................     9
Robert Goldenkoff, Director, Strategic Issues, U.S. Government 
  Accountability Office..........................................    10
Jason Providakes, Ph.D., Senior Vice President and General 
  Manager, The MITRE Corporation.................................    30
Jack Baker, Ph.D., Senior Research Scientist, Geospatial and 
  Population Studies, The National Academy of Sciences...........    32
Andrew Reamer, Ph.D., Research Professor, George Washington 
  Institute of Public Policy, George Washington University.......    34

                     Alphabetical List of Witnesses

Baker, Jack, Ph.D. :
    Testimony....................................................    32
    Prepared statement...........................................   105
Goldenkoff, Robert:
    Testimony....................................................    10
    Prepared statement...........................................    82
Groves, Hon. Robert M.:
    Testimony....................................................     5
    Prepared statement...........................................    51
Providakes, Jason, Ph.D.:
    Testimony....................................................    30
    Prepared statement...........................................    99
Reamer, Andrew, Ph.D.:
    Testimony....................................................    34
    Prepared statement...........................................   110
Zinser, Hon. Todd J.:
    Testimony....................................................     9
    Prepared statement...........................................    67

                                APPENDIX

Questions and responses for the Record from:
    Mr. Groves...................................................   118
    Mr. Zinser...................................................   124
Statement for the Record from:
    Jerry Cerasale, Senior Vice President, Direct Marketing 
      Association................................................   130
    Lisa Maatz, Director of Public Policy and Government 
      Relations American Association of University Women (AAUW)..   132
    Wade Henderson, President and CEO, The Leadership Conference 
      on Civil and Human Rights..................................   135

 
                    CENSUS: PLANNING AHEAD FOR 2020

                              ----------                              


                        WEDNESDAY, JULY 18, 2012

                                 U.S. Senate,      
        Subcommittee on Federal Financial Management,      
              Government Information, Federal Services,    
                              and International Security,  
                      of the Committee on Homeland Security
                                        and Governmental Affairs,  
                                                    Washington, DC.
    The Subcommittee met, pursuant to notice, at 2:34 p.m., in 
Room SD-342, Dirksen Senate Office Building, Hon. Thomas R. 
Carper, Chairman of the Subcommittee, presiding.
    Present: Senators Carper, Brown, and Coburn.

              OPENING STATEMENT OF SENATOR CARPER

    Senator Carper. This hearing will come to order. Welcome, 
one and all. Today's hearing will examine lessons learned from 
the 2010 census, while identifying issues that show promise for 
producing an even more accurate and hopefully more cost-
effective census in 2020.
    I want to begin by thanking Dr. Groves for his commitment 
to public service and for his willingness to help the Bureau 
navigate through some very challenging times. I must admit the 
news of your decision to leave this post as Director of the 
Census Bureau is bittersweet. It is actually mostly bitter, if 
you want to know the truth. But when you came on board in 2009, 
the Census Bureau faced many challenges that threatened the 
success of the 2010 census, as we will all recall, and you 
along with your dedicated staff confronted these challenges 
head on, and through his very impressive skill set and 
background in these issues related to the census and 
statistics, he helped right the ship through the completion of 
the 2010 census.
    And under your leadership, the Bureau completed key 
operations on schedule, hired nearly 900,000 temporary workers 
at a time when a lot of people needed jobs, obtained an 
exceptional participation rate of 74 percent, and managed to 
report population figures in time to support the redistricting. 
The Bureau has also realigned its national field office 
structure and implemented key management reforms, reducing 
costs by an estimated $15 to $18 million annually beginning in 
2014. Three years after your arrival, Dr. Groves definitely 
leaves the Census, I think, in better shape than when you found 
it.
    However, despite these achievements, the 2010 census was 
still the most expensive in our Nation's history, by far, and 
even taking inflation into account, the total cost of the 
single operation escalated from an initial estimate of about 
$11.3 billion to right around $13 billion. Even more disturbing 
is the fact that with all the modern scientific improvements in 
technological advancements that have been made over the years, 
the framework for conducting the 2010 census was based off of a 
model that we first used in the 1970s. In fact, I have just 
been re-reading part of the Old Testament, and found that King 
David ordered a census. Got in trouble with the Lord for 
ordering it; I do not know why.
    But they were doing a census all this time ago. They did 
not have these handheld--I think it was mentioned, I think, in 
the Book of Samuel about the handhelds. No, not really. But 
pretty much we are not a whole lot further along than we were 
then. Actually, we are, but you get the gist.
    Although the methodology basics of the census have remained 
the same over the past 40 years, the cost of the census has 
decidedly not. The average cost per household was about $97 in 
2010, compared to $70 in 2000, and $16 in 1970. The total cost 
of the 2020 census could rise to as much as $30 billion. This, 
in my view, is just not acceptable. We cannot do that. And it 
is especially not acceptable at a time when we are struggling 
to find solutions to the serious deficit and debt crisis our 
country is currently facing.
    I have spoken at previous hearings about the need for us to 
look in every nook and cranny in the Federal Government and ask 
this question: ``Is it possible to get better results for less 
money or better results for the same amount of money? ''
    The hard truth is that many programs funding levels will 
need to be reduced, and even some of the most popular and 
necessary programs out there will likely need to do more with 
less, or at least more with the same amount of money. The 
Census Bureau, despite the vital and constitutionally mandated 
nature of its work, cannot be immune from that sort of 
examination.
    Today we will look at the Bureau's planning efforts for the 
2020 decennial, and although it is 8 years away, it is never 
too early to start thinking about ways to reduce costs and 
improve quality through more efficient data collection. More 
importantly, we need to make certain that the issues that lead 
to the failures and cost overruns we saw the last time around 
have been addressed and will not reoccur. Taxpayers ought not 
be expected to pick up the tab for them again.
    Looking ahead, the Bureau's research should focus on how 
existing technology can be incorporated into the 2020 design. 
Obviously, the Internet is here to stay, and according to 
experts, an Internet response option could have saved the 
Bureau tens of millions of dollars in processing costs in 2010 
alone.
    Future research should not only focus on how to implement 
Internet data collections, but also how to reap the benefits, 
both financial and otherwise, of it, and other technologies 
next time around. We also need to make certain that people who 
make up our growing and changing country are comfortable enough 
with the security of the data collection methods for us to 
allow for an accurate census.
    Moreover, steady leadership will also be critical in 
reversing a trend of decennial censuses marked by poor planning 
and escalating costs. The 2010 census experienced several 
changes in leadership, as we will all recall, and vast spans of 
time with waiting or with acting or interim directors, further 
putting the operation at risk.
    I think in the 27 months leading up to the census day, the 
Bureau had, as I recall, three different Directors during that 
span of time. Dr. Groves, your departure will undoubtedly leave 
some very big shoes to fill, but it is imperative that we get 
someone just as good in place as soon as possible so that we 
can avoid the operational and management changes and challenges 
that plagued some of our prior decennials.
    We look forward to hearing from our witnesses today who 
will help us to identify ways to best balance the need for an 
accurate census, with a need to ensure reasonable costs for 
this endeavor. Before turning to Senator Brown, I just want to 
say to Senator Coburn, just a real special thanks for your 
diligent work on this issue. We have, I think, done better than 
we might otherwise have because of your strong efforts, and in 
the future we will do better still because of your ongoing 
efforts. Senator Brown.

               OPENING STATEMENT OF SENATOR BROWN

    Senator Brown. Thank you, Mr. Chairman. It is also good to 
have Senator Coburn here, and I think he has spearheaded this 
issue and has some concerns, which is a good thing. I am going 
to give my opening statement, leave some questions for the 
record, and defer to Senator Coburn. I have other commitments.
    The decennial census, due to its size, is probably the most 
schedule-driven project mounted by the Federal Government, as 
we all know, and it was a vital undertaking, as you referenced. 
The results were used, obviously, for redistricting and the 
annual distribution of billions of dollars in Federal and State 
funds.
    The results of the 2010 census were both encouraging and 
discouraging at the same time. Encouraging that our population 
has expanded to become more diverse and that the Census Bureau 
achieved a high degree of accuracy in the 2010 count. As you 
referenced, the discouraging news is that the costs exploded 
and we simply cannot afford that continued cost. I believe, and 
have learned in my brief tenure here, that we cannot continue 
to do the things that we have always been doing, to the 
detriment of the American taxpayer.
    For the most part, the basic model of conducting the census 
has not changed since the 1970s, as you referenced. Using 
methods like the old-fashioned snail mail, et cetera. Last 
year's hearing revealed the Census Bureau must innovate and 
bring the census into the 21st Century. I am encouraged as well 
that Director Groves has recognized this need to change the 
census, beginning many reforms aimed at reducing cost while 
maintaining quality.
    I am hopeful. I know as long as I am here I am going to 
make sure that those innovations and cost reduction efforts 
continue. For example, I look forward to filling out my 2020 
questionnaire online. Additionally, with Director Groves' 
pending departure, it becomes more imperative that we 
institutionalize these necessary changes.
    I also want to voice my support for the continuation of the 
American Community Survey (ACS). It is our country's only 
source of micro-area estimates on social and demographic 
characteristics. The ACS survey is critical to many businesses 
like AIR Worldwide of Boston which founded the catastrophe 
modeling industry in 1987 and utilizes the ACS survey in the 
development of its catastrophe modeling. Eliminating the ACS 
survey would be short-sighted and hinder the ability of the 
Census Bureau to achieve efficiencies in the 2020 census.
    This is the second hearing we have had on the census this 
session, so I want to thank you, Mr. Chairman, for that, and 
Senator Coburn, for your efforts and I look forward to watching 
the testimony. Thank you.
    Senator Carper. Good, you bet. Senator Coburn, welcome.

              OPENING STATEMENT BY SENATOR COBURN

    Senator Coburn. Well, first of all, let me thank each of 
you for being here. I think we also ought to thank President 
Obama for his nomination, Dr. Groves. Being on the other side 
of the aisle, we are always critical, but when we recognize 
great leadership in terms of Dr. Groves and his leadership in 
the Senate, we ought to recognize who nominated him.
    I am particularly interested in this hearing because I 
think there is still lots to do, and doing it now rather than 
what Dr. Groves was faced with means we are going to be better 
at it, we are going to save more money, and the data is going 
to be more accurate.
    I do have some significant concerns with the American 
Community Survey. As a matter of fact, I experienced the 
harassment personally from the Census Bureau when I was sent 
the American Community Survey 3 years ago. I stood by my 
constitutional rights to not answer certain questions in that.
    I recently have had five letters to my office on the 
American Community Survey with people with the same complaints 
as to their constitutional right not to answer questions of 
total privacy that is really nobody else's business. That does 
not mean I do not believe we need an American Community Survey; 
I think we do.
    But I think there are ways, and one of the things I would 
like to explore is, how do we do it and still get the same 
information, make sure it is not skewed, and how we do it with 
a carrot instead of a stick, because one of the problems that 
is occurring in our country today is the undermining of the 
confidence of the rule of law.
    And when you have good patriotic Oklahomans that pay their 
taxes every year, work, follow the rules, follow the laws, feel 
invaded when the government comes in and says, We want to know 
this, this, and this, and they say, I am sorry, I will pay the 
fine, but I am not telling you this, this, and this.
    So there are ways for us to get around it. We know that the 
data is important and we know that the accuracy of the data is 
important, but what we have to have is innovative thinking as 
we approach that, and that also goes towards the 2020 census.
    I have had a conversation with Dr. Groves as well as 
appreciate the work of the Inspector General (IG) and the 
General Accountability Office (GAO). I am a big defender of 
IGs, and everybody knows my reputation of protecting and 
defending the GAO for the valuable work that they give to 
Congress. We could not do anywhere close--and that may be a 
good thing--without that help. So I am appreciative of you 
being here.
    I will save the rest of my statement as we ask questions.
    Senator Carper. Thank you, Dr. Coburn. I have a fairly 
lengthy introduction for our witnesses. I am not going to use 
that. I would second what Dr. Coburn said about thanking the 
President for submitting your name and for your service.
    Todd Zinser. Todd has been before us number of times, the 
IG for the Department of Commerce. How is the Secretary of 
Commerce doing? Any idea how he is coming along?
    Mr. Zinser. I think he is doing fine. In fact, he has 
having a little reception for the staff at Commerce Department 
later this week.
    Senator Carper. Oh, that is good. OK, thank you. We are 
very appreciative of your work and that of your team and 
welcome you back here today. And Robert Goldenkoff, our friend 
from GAO, Director of Strategic Issues, responsible for 
reviewing the 2010 census and Federal Government-wide capital 
reforms. Welcome. Thank you all.
    In certain times we try to limit you to 5 minutes. I am not 
going to do that today. If you go on too long, just drone on 
endlessly, I will rein you in. But short of that, we will let 
you go.

 TESTIMONY OF HON. ROBERT M. GROVES,\1\ DIRECTOR, U.S. CENSUS 
              BUREAU, U.S. DEPARTMENT OF COMMERCE

    Mr. Groves. Thank you. I am really delighted to be here. I 
want to thank both of you personally. One of the greatest 
pleasures I have had during my directorship is working with you 
and I appreciate your interest in the Census Bureau and your 
support for the things we are trying to do. 
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Groves appear in the appendix on 
page 51.
---------------------------------------------------------------------------
    Senator Carper. Did you ever think you would say that to 
two guys, one from Ohio State (OSU), and another from-- 
[Laughter.]
    The guy from Michigan. You never know.
    Mr. Groves. So I want to begin with a little retrospective 
on the 2010. And as you, Mr. Chairman, noted, when I was in 
this room for my nomination hearing in the spring of 2009, the 
forecast was one that the 2010 census was headed for disaster. 
I think what we all discovered, I most deeply, was that the 
team that was assembled late in the decade was much better than 
anyone knew at the time.
    Recently, as you know, the statistical evaluation of that 
census showed that it was one of the best this country has ever 
seen, if not the best. Having said that, there are still groups 
in this country that are very difficult for us to enumerate. 
These are deep social forces that produce these problems. We 
have trouble counting renters, young children, young adult 
males, African-Americans, Hispanics, American Indians on 
reservations. That has been true for several decades.
    On the other hand, we overcount groups in the population, 
owners of homes, older persons, females, and white non-
Hispanics. These are problems that are deep in our society that 
the Census Bureau cannot solve itself, but they are real 
problems and cause us difficulties.
    I am pleased that we returned nearly $2 billion of the 
taxpayers' money and presented the key results on time. But I 
want to note, and I think it is important, I think, for the 
Subcommittee to know that the credit for the accomplishments of 
this really go to the wonderful scientists and managers at 
Census Bureau. They pull off an enormous feat and they deserve 
that glory.
    Meanwhile, we have been restructuring and realigning a lot 
of key functions at Census, and we have one mantra in this. We 
want to get more efficient and take advantage of new 
technologies to do that. With the approval of Congress, we have 
restructured the Census Bureau. We have added a Research and 
Methodology Directorate to inhouse discover the innovations in 
statistical operations that reduce costs.
    We have reduced the number of regional offices from 12 to 
6, simultaneously modernizing supervisory structures and 
software support systems. We have built cross-functional survey 
teams with technical and management resources to find 
efficiencies in our operation. We have new leadership in the 
information technology (IT) Directorate that is making a 
difference already.
    We have given it enterprise-wide authorities, so whenever 
feasible, we are sharing IT services across directorates. We 
are using public cloud services. We have done this several 
times successfully, saving big, big money. Consolidating data 
centers building. We are also building a private cloud 
environment that I have great hopes for.
    Senator Carper. Building a what? What is that?
    Mr. Groves. A private cloud environment inside. So a lot of 
our operations have surge needs and if each directorate said, 
``We are going to take care of ourselves, you have to pay for 
that capacity.''
    Senator Carper. At first I thought you said private club. 
[Laughter.]
    You certainly got my attention.
    Mr. Groves. We have completed an application programming 
interface (API), and we have released this in beta, but this is 
going to allow developers all over the world to produce apps on 
all sorts of mobile devices to access our statistical 
information. Within a matter of weeks, we will be able to show 
you an economic statistics app that will run on Android and 
iPhones and allow you instant access to the most up-to-date 
data on that.
    We are attempting to build a culture of innovation and that 
takes a while, but one of our tools is an annual challenge to 
our staff and the challenge is, Write down your ideas to make 
us more efficient and if they are meritorious, we will do them. 
And we are getting hundreds of proposals each year. There are 
tons of ideas in the Federal Government workers and the Census 
Bureau for doing things more efficiently. We are trying to 
release those.
    We are building better statistics through new analysis. We 
have launched a team devoted to blending together dataset that 
we already have collected to create new information. We can do 
this without launching any new data collections. We do not pay 
much at all for that.
    Through the collaboration with the National Science 
Foundation (NSF), we have established a network of research, 
university research teams working on problems that we face of 
statistical, geographical, and computing nature. And when you 
put all these things together, it allows me to speak to the 
2020 planning effort, because it is in this environment that we 
have tried to launch a new way of planning the next decennial 
census. These are interconnected things.
    We, too, recognize that the rising cost of the decennial 
census cannot be sustained. That is well accepted, both by the 
leadership of Census, but all through the culture now. We get 
it. So we are focused intensely on cost-saving ideas. We have 
built a new 2020 directorate. That was part of the 
restructuring, deliberately to build a new culture for the new 
census.
    It is free to optimize its structure. In about a year, 
about this time next year, the proposed new structure of that 
will be put forward. We have created an executive level 
steering committee that directs cross-directorate 
collaboration, a key thing we are working on. We have 
restructured research teams; we have made them smaller and 
nimbler.
    Every one of the research projects that we are mounting has 
cost efficiency as one of the goals. It is not just quality; it 
is quality and cost at the same time. And we are attempting to 
use many small tests rather than a small number of very large 
tests, and we think we will save money on the planning if we do 
that, and we will run through more ideas that way.
    The key innovations that you should look for are things 
that are on our agenda. You know about these things, but they 
need the kind of nurturing that you talked about just a few 
minutes ago. We aspire to do targeted address canvassing at the 
end of the decade that can save hundreds of millions of dollars 
if we do it right.
    To allow us to do that, we want to continuously update our 
master address list and we are working on that right now. This 
will require new partnerships with local areas and the Postal 
Service.
    Two, we are designing--all of our options are what survey 
methodologists call multi mode designs. We will use mail, 
telephone, Internet, face-to-face interviews, and any other 
electronic response option that is going to arise over the next 
few years that we do not even know about yet. We are doing 
those multiple modes because it is clear, if you talk to census 
statisticians around the world, no one thing works for all sub-
populations. We are a very diverse society and all Internet 
census will not work and all mail census will not work.
    Most importantly, we are evaluating using existing 
administrative records that are held in other Federal 
Government agencies to obtain data about the households who do 
not otherwise respond. This tool can save very large amounts of 
money if we succeed in that.
    And then finally, we talk a lot about program management on 
2020, learning lessons from 2010. We have integrated budget and 
schedule and scope, and soon we will have a fully integrated 
schedule that has what questions need to be answered for what 
decisions at what time in the decade and we would love to share 
that with you. That would be a great way to keep track of our 
work.
    Now, in addition to working hard to save money, I have 
another duty in this position and that is to inform the country 
of the impact of budgets on the scope and quality of what we 
do, and I must note that the Fiscal Year (FY) 2013 House 
Appropriations Bill has major effects on what we do.
    It cut the President's request by about $358 million, or 37 
percent. I must report to you that if that stands we cannot 
conduct the economic census of the United States, scheduled for 
just a few months from now, which measures the health of our 
economy.
    As you know, the bill does not permit any spending on the 
American Community Survey. In addition, the cuts will halt 
crucial development of ways to save money on 2020 and it will 
eliminate most of the remaining 2010 products. I feel obliged 
to tell you that.
    I want to turn to near-term challenges that I think are 
very big. Some of these are more amorphous than others, but 
they are things that we all have to worry about. There are four 
in number that I want to mention. One is a moment's notice will 
tell you that the Census Bureau's mission can be fulfilled only 
with the willingness of the American public to provide data.
    We know this every day. We see this every day. A real 
challenge for us is to link the valuable statistical 
information we provide, people's knowledge of that, to these 
requests for their own answers. And that is a constant 
challenge to us. It is a big educational burden to communicate 
to large, diverse groups of people why it is useful to have 
this statistical information for key decisions and then how 
that information leans on individual decisions to participate.
    Two, we must continue to nurture ties with university 
researchers and technology firms. We are very active in 
discussions with technology firms on a variety of issues: 
Dissemination, data collection, and others. The challenges here 
are greatest with regard to staying up-to-date on mobile 
computing technologies, new features of the Internet, and new 
geographical information technologies.
    Three, some Internet-based data and other so-called big 
data sources are relevant to what we do. The explosion of 
Internet-related data in this country is a real change in the 
work of information agencies. A challenge to the Census Bureau 
is developing real organic access to those kind of data, new 
public/private partnerships, and learning how to combine those 
kinds of data with our traditional surveys and censuses to 
provide better information to the American public.
    And then finally, four, the world of statistics is rapidly 
changing. There are developments in the field of statistics 
that we should use. They are model-based innovations that can 
enhance the quality of what we do and reduce the cost of what 
we do. So a challenge to the Census Bureau is making sure that 
we stay up to date on that domain of knowledge, but also that 
we have access to other data to permit us to enrich our 
estimates with auxiliary data.
    So let me close by saying, it is my fervent hope that the 
oversight from this Committee will act to allow the Census 
Bureau to continue on the path that we have set. Thank you very 
much.
    Senator Carper. Not to worry. Thank you so much for that 
testimony, and again, for your wonderful leadership. Todd 
Zinser, Mr. Zinser, please proceed. Thanks a lot for joining us 
again today.

 TESTIMONY OF HON. TODD J. ZINSER,\1\ INSPECTOR GENERAL, U.S. 
                     DEPARTMENT OF COMMERCE

    Mr. Zinser. Thank you, Mr. Chairman, Senator Coburn. Thank 
you for the opportunity to testify about planning for the 2020 
census and the challenges the Census Bureau faces in laying the 
groundwork for the decennial. In our April 2011 testimony 
before this Subcommittee, we identified seven top management 
challenges facing the 2020 decennial. 
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Zinser appears in the appendix on 
page 67.
---------------------------------------------------------------------------
    They included: One, revamping the cost estimation process 
to make it better; two, using the Internet and administrative 
records to control costs and improve accuracy; three, 
implementing a more effective testing program using the 
American Community Survey; four, effectively automating field 
data collection; five, the continuous updating of address lists 
and maps; six, improving project management and planning; and 
seven, establishing a Census director position that spans 
Administrations.
    Since that testimony, we have issued our final report on 
the 2010 decennial which included 19 recommendations and helped 
us identify the challenges ahead for 2020. We have also issued 
two reports, in April and May of this year, one focused on 2020 
planning and the other on the master address file (MAF) and the 
topologically integrated geographic encoding and referencing 
(TIGER) database.
    The challenges we identified in April 2011 remain 
operative. Our work over the past year and recent developments 
have brought five key issues concerning those management 
challenges into greater focus. The first key issue concerns 
departmental oversight of the Bureau's data collection and IT 
infrastructure projects.
    The Department will need to play a strong oversight role 
early in the 2020 census. Now is the time for the Department to 
assess the Bureau's IT and data collection plans and help 
Census manage operational risk. For 2010, we saw the path to 
escalating IT costs begin early in the decade. For 2020, the 
Department needs to help the Bureau as it develops cost 
estimates, establishes critical path management, and maintains 
a more reasonable cost route.
    The second key issue concerns decennial planning within the 
constrained budget environment. We have already seen during the 
research and testing phase how the Bureau has had to adapt to a 
challenging cycle of Federal Government budgets. In Fiscal Year 
2012, the Bureau canceled 20 of 109 studies that measure 2010 
performance and inform 2020 plans.
    Like all other Federal agencies, the Census Bureau must 
continue to plan within constrained budgets. Providing the 
Department and Congress reliable and transparent budget 
requests will be paramount.
    The third key issue concerns continuity of leadership at 
the Census Bureau. Because the Bureau must operate on long 
planning cycles for decennial surveys, it is difficult to 
maintain leadership with a consistent vision and much easier to 
fall back on old ways and institutional habits. Making the 
nomination and confirmation of a new Census director a priority 
will significantly help the Bureau manage its critical issues 
of budget, design, and survey content, which dictate the 
success of the decennial.
    The fourth key issue concerns modernizing the 2020 census 
with an Internet option, greater data sharing, and the use of 
administrative records. When the Bureau decides how the 2020 
census will be designed, it may need the help of Congress to 
facilitate that design.
    The goal should be significantly reduced labor costs 
through more equitable interagency sharing and more effective 
use of data and administrative records, more automated data 
processing and fewer time consuming, costly, personal 
enumeration visits.
    The final key issue concerns the uncertainty surrounding 
the American Community Survey. As Congress considers whether to 
make responding to the survey voluntary rather than mandatory, 
or eliminate the survey's funding entirely, there are several 
implications that will factor into the deliberations, some of 
which we have identified in our written testimony.
    With respect to making the ACS voluntary, Census research 
in 2002 and 2003 indicated that a voluntary ACS would result in 
a significant reduction in the mail response rate. This 
reduction would then require a more costly survey to obtain the 
same level of reliability--and would also have an adverse 
impact on the data quality for areas of low response.
    With respect to eliminating ACS funding altogether, the 
implications include no longer being able to use the ACS as a 
test bed for the 2020 decennial, the loss of the trained and 
experienced workforce distributed across the country that carry 
out the ACS on an ongoing basis and also support decennial 
operations, and losing the opportunity to use the ACS over the 
course of the decade to build and perfect the IT infrastructure 
necessary to securely use an Internet option for 2020.
    Finally, Mr. Chairman, on behalf of the entire staff at the 
Office of Inspector General, I would like to thank Dr. Groves 
for his leadership of the Census Bureau. It was a personal 
privilege to have served with Dr. Groves and we wish him well. 
This concludes my statement. I would be happy to answer any 
questions you or Senator Coburn may have.
    Senator Carper. Thanks much for the testimony. We welcome 
Mr. Goldenkoff. Please proceed. Thanks for joining us.

TESTIMONY OF ROBERT GOLDENKOFF,\1\ DIRECTOR, STRATEGIC ISSUES, 
             U.S. GOVERNMENT ACCOUNTABILITY OFFICE

    Mr. Goldenkoff. Mr. Chairman, Senator Coburn, thank you for 
the opportunity to be here today to provide an update on the 
Census Bureau's planning and reform initiatives for the 2020 
census. As was earlier noted, the basic design of the decennial 
census, an approach the Bureau has used since 1970, is no 
longer capable of a cost-effective enumeration. Unless changes 
are made going forward, future headcounts could be fiscally 
unsustainable. 
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Goldenkoff appears in the 
appendix on page 82.
---------------------------------------------------------------------------
    The 2010 census was the most expensive population count in 
U.S. history with a total cost of around $13 billion. Without 
reforms, the 2020 census could be even more costly. In fact, 
the Census Bureau estimates that if it uses the same approach 
to count people in 2020 as it did in 2010, it would cost $151 
to count each housing unit compared to 2010s $97 per housing 
unit, a 56 percent cost increase.
    When we last testified before the Subcommittee in April 
2011, we discussed four lessons learned from the 2010 and 
earlier decennials that could help secure a more cost-effective 
enumeration in 2020. They included, one, reexamining the 
Nation's approach to taking the census. Two, tailoring 
operations to specific operations in population groups. Three, 
addressing prior high risk areas. And four, ensuring that the 
Bureau's management, culture, and business practices all align 
with a successful census.
    The Bureau generally agreed with these actions and is 
taking steps to address them. As requested, my remarks today 
will focus first on the Bureau's progress in each lesson 
learned, and second, what remains to be done going forward.
    Overall, the Bureau's preparations are off to a good start. 
However, sustaining those efforts will be a tremendous 
challenge as the Bureau's planning over the next few years will 
take place in an uncertain environment given the extent of the 
Bureau's reforms, budget constraints, and next month's planned 
resignation of the Census Director. As a result, continued 
Congressional oversight throughout the decade will be critical.
    With respect to the first lesson learned, reexamining the 
Nation's approach to taking the census, the Bureau is already 
rethinking the design of the enumeration. For example, the 
Bureau is researching how it can use administrative records 
such as data from other government agencies to help it locate 
and count people, including nonrespondents. Administrative 
records could help reduce the need for costly field operations, 
but the Bureau must first resolve data quality and access 
issues.
    As for the second lesson learned, tailoring operations to 
specific locations and population groups, in 2010, the Bureau 
effectively targeted several activities, including its paid 
advertising campaign. For 2020, the Bureau is considering 
expanding these efforts, for example, to address canvassing. A 
better understanding of value added of key census operations 
would help the Bureau's ability to further target its 
operations and allocate its resources more efficiently.
    On the third lesson learned, addressing factors that led to 
2010s high risk designation, it will be important for the 
Bureau to improve its ability to develop reliable life cycle 
cost estimates and strengthen its information technology (IT) 
management so that shortcomings in these areas do not recur in 
2020.
    With respect to the fourth lesson learned, ensuring that 
the Bureau's management and culture align with the cost 
effective enumeration, we found that the Bureau's early 
planning efforts for the 2020 census were consistent with most 
leading practices for organizational transformation, long-term 
planning, and strategic workforce planning. But in the future, 
additional steps will be needed, including determining how to 
monitor and evaluate its workforce planning efforts.
    In summary, the Bureau is making noteworthy progress in 
addressing these and other lessons learned from the 2010 
census. To help sustain the Bureau's progress going forward, it 
will be important for Congress to continue to hold the Bureau 
accountable for results, weigh in on key design decisions, and 
provide appropriate funding.
    I would also like to take this opportunity to congratulate 
Dr. Groves on his successful stewardship of the Census Bureau 
and wish him the best of luck in his new position at Georgetown 
University. Under Dr. Groves' leadership, the Census Bureau 
mitigated the risks jeopardizing the decennial and delivered an 
operationally successful headcount.
    The Census Bureau embarked on a more transparent and 
results-oriented path, and it developed a vision for a more 
cost-effective approach to collecting and disseminating data in 
the future. Mr. Chairman, Senator Coburn, this concludes my 
remarks. I would be pleased to respond to any questions that 
you or other Members of the Subcommittee might have.
    Senator Carper. Mr. Goldenkoff, thanks a whole lot. I have 
asked Dr. Coburn to lead off the questions. I have got to 
return a phone call real quick and I will be right back. Tom?
    Senator Coburn [presiding]. Well, thank you for each of 
your testimonies. What I noticed across all three testimonies, 
Dr. Groves was talking about the changes he has made in terms 
of IT. Actually, I believe what you said was IT was across all 
sections, all silos, but I also heard worries by IG and GAO on 
the implementation of that. What you all are talking about is 
one of our biggest problems in the government, not just at 
Census.
    So first of all, Dr. Groves, I wanted to correct an 
omission I made. You really were successful, through your 
leadership, but also the quality of people that you have 
working for you. And so I want to publicly recognize them, 
because what we saw was under great leadership they shone. They 
stepped up. So that reflects great on those employees, but as 
well your leadership.
    So tell me, if you would, hearing the concerns about IT 
transformation and the watching of it, it is a sinkhole. Just 
so you all will know, we spend $64 billion a year on it in the 
Federal Government and half of it is wasted, fully half, $32 
billion. The American people need to hear that. $32 billion a 
year goes down the drain in IT projects that do not work.
    So would you give me some assurance, first of all, of where 
they are worried and what is in place, and how are you going to 
address their concerns on the IT portion of it?
    Mr. Groves. Well, worry is not an ill-placed emotion, I 
think, because it would motivate the kind of scrutiny of 
changes that are coming in now. I must admit I am rather 
optimistic. There are features of the environment on the IT 
side and the leadership side that were not there 10 years ago. 
OK?
    So a move for more centralization of authority, of CIOs is 
afoot. That seems inevitable that will not stop. And the reason 
I do not think it would peter out is if you just look at 
Census, 10 years ago each of our directorates really had their 
own IT set-up, they had their own hardware, they had, many 
times, unique software systems.
    What we are doing in our data center is virtualizing 
servers, basically. So you may have had your own servers that 
did your stuff and I had mine. Now we get the same services, 
but we are actually sharing servers in a completely--in a way 
that does not affect our productivity. That service provision 
is a change, I think, and a good change.
    And then the hard thing is going to be the next 
developments, the new things that we do. How do you avoid the 
problems of the past where you spend years developing 
requirements and then years more working on system development 
that actually is so late in delivery that it does not solve the 
original problem?
    Agile programming environments are good. We are using that 
wherever we can. The very basic lesson of keeping the user in 
the loop all the time, I think, we have gotten, but it is 
counterculture still. So pressure in that regard, I think, is 
needed throughout.
    I am hoping we have turned a corner, both because of these 
administrative things, the technical nature of the hardware 
environment, and then our software developments, but scrutiny 
will be required.
    Senator Coburn. So there is good reason for us to watch 
that closely?
    Mr. Groves. Absolutely. And, calling the leadership on 
this. Our leadership is attuned to this problem now. I think we 
talk about it a lot. We are talking openly about it as 
something that we want to do better on. I think that is real 
healthy, but calling us to account on this is a useful thing, I 
think.
    Senator Coburn. So you have got this in place and moving 
forward and you are leaving next month. That is what I hear the 
IG and the GAO say. OK. How is it going to go forward without 
you there?
    Mr. Groves. Well, we have good leadership. We have a 
wonderful new CIO. He has brought in several key, high-level 
scientists who are rolling up their sleeves and working with 
the program areas. They have built the trust or are building 
that trust.
    I think the other way you could keep track of this is that 
for 2020 especially, we are going to be pretty transparent on 
deliverable dates and key intermediate deadlines. That will be 
out there and we are hoping to establish an ongoing briefing of 
your staff, so too in the House, to tell you where we are on 
things. And that should be a wonderful vehicle that would allow 
you to catch things that are off-track early, if they go off-
track.
    Senator Coburn. Any comments on what you have heard, either 
of you? Both of you raised it as a concern.
    Mr. Zinser. Sir, I do think that the culture at the Census 
Bureau has greatly improved since we expressed concerns over 
the handheld computers. That problem was hidden, possibly from 
the Director and definitely from the Department.
    That is not going to happen in the future the transparency 
mechanisms that have since been built in and the increased 
oversight will help in that regard.
    Senator Coburn. Well, it also was a problem. We did not 
know what we want, so we created a cost-plus contract and we 
kept changing what we want. Plus, in my personal opinion, we 
had poor performance by the supplier. So I am in adamant 
opposition to all cost-plus contracts other than pure research, 
because if we do not know what we want, then the first thing we 
ought to do is go research what we want and create specifics 
for that. Anything else?
    Mr. Goldenkoff. Well, I agree with what has been said so 
far. What I would add to that is that the management processes 
need to be in place before any contracts are let. I think what 
you were alluding to is that for 2010, the specifications for a 
lot of these systems were not matched to the operational 
requirements and that led to some of the problems that came 
about later in the decade.
    Senator Coburn. We actually showed in a Subcommittee 
hearing how you could get an iPhone to do exactly what we paid 
almost $500 million to get somebody to develop. So, I mean, 
that is a governmentwide contract problem. From the testimony 
we had earlier today from Danny Werfel at the Office of 
Management and Budget (OMB), the Comptroller, I think we are on 
the way to starting to change some of that culture within the 
government. So I think that is good.
    Mr. Goldenkoff. And it is a cultural shift that needs to 
take place. It is one thing to talk about the implementation of 
agile programming or, we saw in 2010 the use of IT management 
leading practices here and there, but it has to be implemented 
organization-wide. There needs to be that executive level 
oversight and that is one thing that we did not see happen 
during 2010, but we are seeing some important shifts moving 
forward.
    Senator Coburn. One thing that caught my mind during all of 
your testimony is getting this administrative data from other 
government agencies. Who in the government knows what we need 
to do to enable that for you? Because there are certain laws on 
the books that restrict some of that? Somebody needs to be 
charged with an assessment of how do we create an activity?
    Or if it requires legislation that allows the Census Bureau 
to have data that otherwise we would not want them to have 
through previous legislation? Does anybody know where that list 
is so that we could actually start thinking about how we can--
if it is going to require our help to get there, what we need 
to be doing? Because the sooner we get that done--this place is 
designed to not pass laws. That is what our Founders intended. 
They wanted it to be very difficult to change laws.
    So the sooner we get that list of what is necessary for us 
to do, the sooner we can aid the Census Bureau in what I think 
I heard Dr. Groves say could save us millions and millions and 
millions of dollars, plus a lot of time. Anybody have any idea 
of what that list is?
    Mr. Zinser. Senator, we do have some suggestions about 
legislative changes. We think there are sufficient laws on the 
books allowing the Census director to access records and 
administrative data from other agencies. However, there is not 
corresponding legislation requiring the heads of those other 
agencies to provide the information to the Bureau. As a result, 
there are some key legislative initiatives that could help.
    Senator Coburn. Can you address that directly to me in 
response to this question?
    Mr. Zinser. Absolutely.
    Senator Coburn. Put it in writing and where those areas 
are? I think that would be very helpful, and my commitment is 
to try to grease that skid to see if we cannot get it done.
    Mr. Zinser. Yes, sir.
    Senator Coburn. Privacy rights are an important thing in 
this country and we can do that without violating privacy 
rights. But we also do not want fiefdoms to limit our ability 
to save tons of money.
    If we could, can we move on to one other question for Dr. 
Groves? I have never had a hearing where I am the only one 
present, so it is kind of fun. One of the things I heard in 
your testimony was an expansion of what you can do in terms of 
data, in terms of research, in terms of data. And I would just 
put forth a cautionary tone for you there.
    This country faces an enormous challenge in its fiscal 
situation. This Congress has proved it is not up to that. Maybe 
future Congresses will be. But I think it is important to go 
back and look first at what the charge is for the Census Bureau 
before you take on new things. Two reasons I would say that.
    One is the fiscal implications and the dollars spent there 
that might not be able to spend, saving us some money somewhere 
else within your organization. And I will assure you that you 
will not see the House number for the Census Bureau at this 
time. It will not come through. The worst will be is you will 
be where you are right now for at least--for the first 4 or 5 
months, and I think that is pretty well a given by most of us 
looking at that.
    That is not to say that we are not going to see some dollar 
cuts, but you ought to be involved in what those are, rather 
than some of us who actually do not know what we are talking 
about being involved in what those are.
    But the second point is, there is a private sector out 
there and if what you can bring through analysis and research 
of numbers that you have that are already public, somebody 
wanting to make a buck can also do that. And there is a lot of 
private capital out there that is looking for ways to make 
money out of your data. You see it all the time.
    And so, I would just caution, that is just one Senator's 
opinion, because we are going to have trouble doing the basics 
in the future, financing the basics for what we do. So I would 
just caution in that regard. If I could continue for a little 
bit longer, I will be finished with just one more.
    It leads the American Community Survey. As you and I talked 
in our office, there is a real cultural change occurring in our 
country about the invasiveness of that survey. Some of it is 
right; some of it is wrong. And I think it is good to explain 
to people what the utilization of that data is and why they 
ought to participate in it. I do not discourage that.
    But if you look at the Constitution, there is nothing in 
there about the American Community Survey. It is an expansion 
that we have done. And we have done it on the basis of having 
data so we know how to distribute funds. That is one of the 
reasons. That is why it started.
    But I think it is really important that we figure out, how 
do we get this data and still pay attention to this very real 
concern by people that we are violating their privacy? And I 
think we can figure that out. The study says that you had a 20 
percent reduction in participation. Well, there are carrots 
that can increase that, and figuring out what those carrots 
need to be to not skew the data is not rocket science. I mean, 
it will take some work, but it can be done.
    But the point is, I think we ought to pay attention to the 
message we are hearing. This is not a small deal. This is a big 
deal. And the reason you saw that coming from the House is 
because the people's House are feeling all this blowback on the 
American Community Survey.
    So I would love, before you leave, even though we have had 
a conversation about it, Director Groves, and I would also love 
you all's input as well, and anybody else listening to this, is 
how do we accomplish the collection of the data in a way that 
truly protects us? How do we innovate in a way that gets us the 
same data with as good a quality and not rub up against 
personal liberties?
    And I think that is the challenge in front of us, because I 
am not certain that you could not, even though it has been 
tried before, but under the environment we are in today, see a 
real revolt on that. The Internet spreads information. Half the 
time it is true and half the time it is not.
    But people do not go to factcheck.org. What they do is they 
write their Senator a letter, and then you have to say, Well, 
these are not true facts that you are reacting to. So how about 
your comments on that? If all the witnesses would comment on 
this issue.
    Mr. Groves. Well, I understand the reaction. I have talked 
to respondents, as you probably have talked to your 
constituents on this, so I understand the concerns. There are 
technical issues that could be investigated on this. Your 
notion of approaching this not to increase the benefits of 
participation as opposed to just persuasion is logical and we 
know how to do that. We need a little research to make sure it 
works. It is risky, but it could be done.
    The enormous challenge the ACS has is that it publishes 
statistics on really small areas. That makes it slightly more 
difficult than, say, the unemployment survey which is not 
producing such lower level estimates. My reaction to what you 
are saying is to take this on as a serious matter that could be 
investigated in a variety of ways.
    We have been having workshops, by the way, through the 
National Academy of Sciences panel on the 2020 with people who 
are not really users, but more representative of consumers, and 
ideas are coming up out of that suggesting ways that we could 
keep in touch with real people as we do our work in ways that 
would benefit.
    There will be--my knowledge--I worked before I came to 
Census in this very area of how do people make decisions on 
whether to participate or not. I am not optimistic that any one 
thing will work for all people.
    Senator Coburn. No, I agree.
    Mr. Groves. And there will always be, especially in this 
country, a strain of people that say, I do not want to tell 
anyone anything about myself. We will not succeed on them. But 
on large portions of the population, I think it is worth 
investigating.
    So my reaction, if you are exactly right and this is the 
beginning of a wave and this will just get bigger is, it is 
prudent for us to do the research that provides you with the 
trade-off decisions on what would happen if we took this course 
to the quality of the data versus some other course, and that 
would be a proper role for a statistical agency and a proper 
role for the Congress, I think.
    Senator Coburn. Well, I would say the fact that the House 
passed elimination of funding for the ACS is a pretty good 
indication there is a wave out there. I do not know if you 
would concur with that, but that tells me something. And I know 
we need that data. So one of the legacies you could leave as 
you leave is to make sure that is one of the research projects 
that is started and tested and looked at.
    Because I would suggest, with the low rating government as 
a whole has with the American people, that your compliance is 
going to tend to get tougher. The voluntary compliance is going 
to go down. I do not want that to happen. But at the same time, 
we need the data so we need to be on top of that. Any comments 
from either of you on that?
    Mr. Goldenkoff. Well, I mean, there is the inherent 
challenge of balancing the need for timely and accurate small 
area data on the one hand without violating people's personal 
privacy and unduly burdening them on the other hand. I think 
one way, one approach, the Bureau can take, in the shorter term 
is to continually re-examine and assess the need for specific 
questions on the ACS. Are they still relevant? Are they needed 
at that low level of data, for example? Some of them still 
might be; others may not be, and I do not know how often that 
review process is done.
    Senator Coburn. And that also would be a variable based on 
the geographic location you are looking into, too, as well, as 
you try to get these groups that are under-represented.
    Mr. Goldenkoff. Because, I mean, some of those questions 
really had their origins in the census long form back prior to 
2000. Well, here it is 10 years later and maybe it is time to 
revisit the need for those questions at that level of 
granularity.
    Another thing that the Census Bureau might do is look at 
how they are publicizing and communicating what is done with 
this data. Some of those questions on their face, they may look 
a little bit odd and people may question, why does the 
government need to know this, because we know that is 
happening. So maybe a big part of it could be a communications 
issue.
    And now with all the various forms of technology that are 
available, and social media, there might be new and better ways 
of communicating the importance of that information and what is 
done with it.
    And then finally, another approach for the Bureau to take 
is to communicate the cost of non-response--how much that will 
cost for follow-up--that the Bureau will come back and call you 
up on the phone and then knock on your door.
    Senator Coburn. Well, that is the whole point of changing 
the way you do this so that you do not get this non-response.
    Mr. Goldenkoff. Right. And so now in these cost-conscious 
times with taxpayers saying, ``Hey, this is my money being 
spent,'' well, maybe that by itself might be an incentive. And 
then finally, as you said, there might be other incentives out 
there instead of the hammer of this is required by law, which 
probably is important and Bureau studies have shown the 
importance of that. But are there other incentives that could 
work more constructively with folks?
    Senator Coburn. All right. I just have one final question 
and that is on the part of maps and what the Census Bureau is 
doing with maps versus what can be bought in the private sector 
and the justification for continuing to spend money there when, 
in fact, you can probably or, at least, I think it was, Mr. 
Zinser that raised the issue in terms of maybe what is 
available in the private sector can be modified to meet the 
needs rather than spending the money internal in the Census.
    Mr. Groves. I think a good development that is occurring 
because of our work on trying to reduce full address canvassing 
is basically an opening up of collaborations with all sorts of 
entities. It would be good if that were nurtured. The 
relationship between the private sector and statistical 
agencies is a dynamic one, more dynamic than it has ever been 
in my career.
    We are working with several of them on dissemination and, 
in a way, a map is a dissemination of a piece of information. 
So getting that right is difficult because the private sector 
has different needs and goals, but we are working on this. I 
would hope that continues.
    I think part of the reason I am optimistic it may continue 
is our budget pressures are such that we are re-examining 
everything we do and that will continue, I suspect, for some 
years. That is a healthy environment to consider new 
arrangements.
    Senator Coburn. All right.
    Mr. Zinser. Sir, can I just add one thing? I am sorry.
    Senator Coburn. Sure.
    Mr. Zinser. I think your question about improving Census 
Bureau map quality and cost-efficiency is about more than just 
sharing information or using available private sector 
resources. There also needs to be a freer flow of information 
between the Census Bureau and State and local governments. 
Right now, the law is basically a one-way street, except for 
one part of the decade when the Bureau conducts an address 
update operation with local governments. Laws need to allow 
more free-flowing information between jurisdictions and the 
Census Bureau.
    Senator Coburn. Where is the problem? Coming from the 
States?
    Mr. Zinser. The Census Bureau has restrictions from Title 
13 on sharing information. Census shares address information 
with local governments during only one period of the decade; we 
think legislation ought to loosen those restrictions so that 
the Bureau can share data more freely throughout the decade as 
part of the Bureau's continuous updates.
    Senator Coburn. Well, would you add that to the list--
    Mr. Zinser. Sure.
    Senator Coburn [continuing]. That you are going to send to 
me?
    Mr. Zinser. Yes, sir.
    Senator Coburn. Thank you, Mr. Chairman.
    Senator Carper. [presiding]. Sure, you bet. I just want to 
follow up on a little bit of Dr. Coburn's questioning on the 
ACS. One of the things that continues to amaze me is I have two 
sons that are 22 and 23. They are all about social networking 
and have been for a while. And I am amazed at the kind of 
information people are just willing to like bare their souls in 
Facebook and other venues.
    And then when it gets to something like the ACS, they are 
just much more reluctant, maybe understandably so because you 
are turning it over to the government as opposed to your 
friends or those who are thought to be friends.
    But anyway, you said something earlier in your comments, 
Dr. Groves, that I took down and asked Velvet Johnson who is 
sitting over my left shoulder, asked her to note it as well. 
You had asked the Census employees with an eye toward 
innovation and creativity just to think about good ideas and 
smart ways to do business and just submit those. I think you 
said you had gotten hundreds of folks who participated in that.
    And I said, one of the things we need to do on another side 
of our jurisdiction is work with the Postal Service. And going 
forward in our work with the Postal Service we must try to make 
sure that they remain not just relevant in the 21st Century, 
but sustainable on a fiscal basis. They need not just to cut, 
cut, cut. They need to find ways to get better results for less 
money and they need to be able to grow revenues. And so, I 
said, just write that down. We want to make sure that the 
Postal Service is doing the same thing in soliciting ideas from 
their folks.
    I do not know to what extent this ACS issue is a growing 
concern and I think it is reflected by the House vote, but if I 
were in your shoes or the shoes of the person who is going to 
succeed you, I would be making sure we do a lot of focus groups 
around the country in different venues and just add people from 
all different walks of life throughout the country.
    Where do you think the lines should be drawn knowing what 
you need in terms of information? Just get their input. I think 
I would urge that whoever is going to be sitting in your seat 
6, 7, 8 months from now, that they will be doing that. I know 
we have other folks who are in the audience who is from the 
Census Bureau? Just make a note of that if you would. Be sure 
to tell the next Director that there are one or two guys on 
this Subcommittee that think that is a good idea.
    The other thing I want to ask--I have several things I want 
to ask, but every organization I have ever been a part of, 
whether it was the Navy or running the State of Delaware as 
Governor, or working here in the Senate, everything that I have 
been a part of leadership has been the key.
    If I go to a school, the school is doing a good job, more 
often than not because they have a great principal. They can 
have wonderful teachers, but if they have a lousy principal, 
they are not going to be meeting their potential. The 
leadership is just so important. The same is true here with the 
Census Bureau.
    One of the things we ask all in trying to help identify a 
talent pool for people to fill this position several years ago, 
we had a panel in. I think they were people that had previously 
been Census Bureau directors, and we asked them to basically 
give us, not at that time, but to just give us a couple good 
names that we would then share with the Administration. I also 
asked them to tell us what kind of skills and talents we ought 
to be looking for, or the Administration, our country should be 
looking for.
    I will not ask you all to give me a couple good names, 
although later on I may ask Dr. Groves to give me a couple to 
share with the Administration. But rather, the qualities that 
we ought to be looking for, the President should be looking for 
our country, our Senate, our colleagues should be looking for 
as we move into this century.
    Let me just start with you, Dr. Groves. Just give us some 
thoughts of the qualities, the kind of leadership qualities and 
skill sets we should be looking for.
    Mr. Groves. I view this position as both a managerial one, 
but a technical one, a scientific one. So what goes on at the 
Census Bureau is based on a set of principles that are derived 
often from statistics, but sometimes from other disciplines. So 
pretty high on my list would be someone that understands that 
knowledge domain. Does not mean they have to be the greatest 
scientist in the world, but you have to be able to understand 
technical matters.
    I want to stress that because I think I have heard opposite 
opinions, that what we need in the Census Bureau is maybe an 
ex-general that has run an Army. It is sort of like running an 
Army. Right? That is what you need. Well, I do not agree with 
that.
    Senator Carper. OK.
    Mr. Groves. Now, you do need managerial--
    Senator Carper. How about a Navy man?
    Mr. Groves. I did not want to pick on one branch.
    Senator Carper. OK.
    Mr. Groves. Sorry. But you do need managerial expertise, 
and what kind you need, I think you need a calmness and a 
certainty of decision style. That is, you have to make 
decisions. Decisions need to be carefully crafted, but quickly 
made. Indecision is deadly, I think, in an environment like 
this.
    As Senator Coburn noted, we have some good leaders at the 
Census Bureau, and knowing when to delegate and when not to 
delegate, what is the right decision on that score is important 
greatly. And then I have learned something that I did not know 
before I took this position.
    There is an outreach side of a Director. The ability to 
relate to communities, to understand how real people live is, I 
believe, I would rate that higher than I did before I came in. 
That is pretty important because we are relying on their 
goodwill to do our business, and kind of understanding how 
people process what we do is an important thing.
    Senator Carper. Those are great items to consider. Mr. 
Zinser, do you agree with any of those or disagree with all of 
them, add to them?
    Mr. Zinser. Yes, sir. I would agree that all the qualities 
of a good leader are important. With respect to the Census 
Bureau, what Dr. Groves brought that did not exist before was 
an openness and willingness to subject the Bureau to oversight 
and to communicate with the Department and the oversight bodies 
in a straightforward way.
    I think that the way the Census Bureau operated in the past 
was a little more insular. The next leader of the Census Bureau 
would do well to adopt an open and transparent management style 
like Dr. Groves did.
    I also think that relationships are very important inside 
the Department, with Congress, and with the rest of the 
country.
    Senator Carper. OK, thanks. Go ahead, please.
    Senator Coburn. Just a follow-up question. How are we going 
to assess that? Let us say we have a new one and it is not as 
open and transparent as Dr. Groves. How are we going to find 
that out? Because the problem is, is we all know how 
organizations work, and outside a whistle-blower on a specific 
issue, tell us how we are going to assess that.
    Mr. Zinser. I brought up transparency at the Census Bureau 
because I actually had to write a letter to this Subcommittee 
when I first arrived at the Department asking for the 
Subcommittee's help in getting access to Census Bureau 
information. We were faced with all kinds of barriers with 
respect to operating inside the Census Bureau and accessing 
their information.
    And to the credit of Dr. Murdock and then Dr. Groves, there 
has been a sea change. As a result, transparency depends partly 
on hearings like this. GAO, OIG, and the Department have to be 
held accountable for pushing that kind of transparency.
    Senator Coburn. It is actually a more fun place to work 
when it is transparent.
    Mr. Zinser. That is what I think.
    Senator Carper. Mr. Goldenkoff.
    Mr. Goldenkoff. Well, in terms of the leadership 
characteristics--
    Senator Carper. Let me just say, one fact comes to mind. We 
have some opportunities here as well as we go through the 
vetting process, as we go through the confirmation process. I 
just think nominees come to call on us and that is a good point 
to make. I would say, By the way, did you know? All right, 
please.
    Mr. Goldenkoff. We listed some attributes in our written 
statement that we submitted for the Record, but just building 
on some of those in thinking about attributes that help create 
a high performance organization, those are things that are so 
essential. And I think to Dr. Groves' credit he certainly 
exhibited and applied a lot of them and thought more at a 
corporate level, and brought a lot of stewardship and 
innovation to it.
    I think in the past where the Bureau was not as successful 
is when they applied more of a scientific or ``if 
statistician's approach to solving problems. As if we can just 
do one program a little bit better, apply this one statistical 
thing, do it a little bit better, we would make some 
improvements.''
    But for the first time, we have been starting to see that 
these issues, because they have been recurring again and again, 
that the problems are more deeply rooted in the Census Bureau. 
It was things like the culture of the Census Bureau, the 
management of the Census Bureau, old ways of thinking, 
stovepiped organizations. Now those issues are being addressed, 
and so that is so important.
    Further, developing and using partnerships strategically is 
critical. Working with political leaders and Capitol Hill and 
reaching out to other levels of government, and then everyday 
citizens. And then getting back to internal things that the 
Census Bureau can do, managing people strategically, talent 
management, just making the Census Bureau a desirable place to 
work.
    Senator Carper. Let me ask you a question. I have 
jurisdiction in another Subcommittee that deals with the 
Nuclear Regulatory Commission (NRC), and for a number of years, 
consecutive years, they have been at the top of the charts in 
terms of a place where people like to work. They get very good 
reviews. Any idea what it is like at the Census Bureau in terms 
of those rankings? Any idea?
    Mr. Groves. Well, we have employees--actually, you told me 
this in one of my first hearings, so I followed up.
    Senator Carper. Oh, good.
    Mr. Groves. I tried to decompose why is it that NRC have 
employee surveys that measure attitudes, and it is a very mixed 
bag. There are some parts of the environment that are really 
energized, believe they are doing important things. I have 
analyzed those data myself and there are some bad things in the 
data. We get very high ranks on people believing what they are 
doing is important. That is a good thing for a public servant 
to have that.
    One of the most negative things is having to work next to 
someone who is not working very hard, the inequity of 
productivity and the lack of reward systems that allow one to 
get rid of deadwood and stuff like that. That is a drag on 
employee morale that is in the data. They are telling us that 
quite honestly and that is a hard thing to work on.
    Senator Carper. OK, thank you. I am sorry, Mr. Goldenkoff, 
I interrupted you there.
    Mr. Goldenkoff. Just to build on that, making an 
organization a desirable place to work, and that is exactly 
what I was getting at. You were referring to the Partnership 
for Public Service, the survey that they do on best places to 
work in the Federal Government. And if you look at the Nuclear 
Regulatory Commission and other organizations that have always 
ranked high, and I have to say GAO is among them, that--
    Senator Carper. I like the way you worked that in there. 
[Laughter.]
    Mr. Goldenkoff. I know my boss was up here earlier today.
    Senator Carper. He never used a note. I told someone, I 
said to the other colleagues, I said, for all the years I have 
been here, I have only had two people come in and testify for 
an extensive period of time, and never use a note. He is one of 
them and the other is John Roberts, our Chief Justice of the 
Supreme Court. Very impressive.
    Mr. Goldenkoff. Very. Well, you see me with my binder here 
and papers all over the place.
    Senator Carper. You and me both.
    Mr. Goldenkoff. Comptroller General sets a high bar. But 
one of the things for those organizations that rank highly in 
the best places to work--
    Senator Carper. I have noticed that with the change of the 
Comptroller General, there is a guy who sits right behind him 
over his left shoulder and when Gene Dodaro speaks, the guy 
right behind him, it is like his lips are moving. [Laughter.]
    Senator Coburn. That is absolutely untrue. [Laughter.]
    Mr. Goldenkoff. Well, there is a clear and compelling 
mission of the organization. They have an aggressive recruiting 
program. They have a wonderful on-boarding process. It is one 
thing to recruit people, but what do they do once they get 
there? Is there a buddy system to keep them engaged? Are there 
rotational opportunities so they see all different parts of the 
organization? Is there a good work/life program? Things that 
make people want to stay and feel good about the place that 
they work for.
    And ultimately, what the Partnership for Public Service has 
found out, a lot of what it comes down to is leadership.
    Senator Carper. I like to tell the story about listening on 
my way from my home to the train station last fall, I catch a 
7:15 train many days and I listen to the news. I watch the 7:00 
top of the hour, listen to NPR on my way to the train station.
    A number of months ago, late last year, NPR was reporting 
on an international survey. The question was being asked in the 
survey of people all over the world, thousands of people, is 
what is it about your job that makes you like it? And people 
had a wide range of responses, but a lot of people said, 
``Well, I like getting paid. I like getting a paycheck.''
    Some people said, ``I like having health care benefits,'' 
some people said ``I like receiving a pension,'' some people 
said, ``I like the folks that I work with or the kind of 
surroundings in which I work.''
    What most people said was the thing that was the most 
important to them in terms of liking their work is they felt 
that they were working on something that was important and that 
they were making progress.
    Mr. Goldenkoff. Challenging work and making a difference.
    Senator Carper. Yes, and that was the key for more people 
than not. One of the questions, when we get to the end of this 
session I am going to ask--the last question I will probably 
ask of you and you can just think about it now--is just think 
what each of you have said and I just want you to share with us 
something that you heard from one of the other two witnesses 
that maybe made you think a little differently or that you 
think is especially important that you would just like to 
underline and leave for us as we look at it. Just be thinking 
about that.
    I think Dr. Coburn got into this a little bit, but I am 
going to ask this question anyway and just ask you again, if 
you will. Starting with you, Dr. Groves, but just to reiterate 
again some of those important lessons that we learned from the 
2010 census that the Bureau should really be focused on now 
looking ahead.
    Mr. Groves. Well, I was not able to talk about some of the 
things, so I thank you for this question, first of all. We had, 
in 2010--this was before my time so I cannot claim credit for 
it--the wisdom to design the post-enumeration survey, which is 
our way of evaluating the census, in a way that allows us to 
decompose the quality, the marginal impacts on quality of 
different operations.
    So looking forward on 2020, we have, for the first time, a 
way of asking the question, how valuable was a particular 
operation to the overall quality? Now, we do not have perfect 
cost data on each of those operations, but we could do pretty 
good work, and for the first time, I think this decade will 
have cost/quality trade-off commentary on different components.
    Sometime this decade, if all goes well, we will publicly be 
able to say we could do this operation that costs X hundred 
million dollars and we think it would impact the quality this 
amount. We have never had a discussion in this country about 
how good does the census have to be for a particular cost. We 
will need Congressional help on this.
    Senator Carper. How so?
    Mr. Groves. It is crystal clear, in Article 1, Section 2, 
that the census will be done in a manner that by law Congress 
shall direct. For the first time, I think we will be able to 
give cost/quality trade-off. The mandate in all paths, if you 
study the history of this is, do better, do better, do better. 
Or at a particular point in the decade, make it cheap, make it 
cheap, make it cheap, but we have to bring these together.
    Senator Carper. There you go.
    Mr. Groves. And it is a tough trade-off decision and it 
belongs to Congress, I believe.
    Senator Coburn. Yes, but there is a law of diminishing 
returns.
    Mr. Groves. Absolutely and we know it. We see it in the 
data. But it is that law or that diminishing return that poses 
the issue. How much more money should be spent for what payoff, 
and what payoff is worth getting, which is basically a 
Congressional question.
    Senator Coburn. Well, we have 10 tough years in front of us 
at least.
    Mr. Groves. Yes.
    Senator Coburn. And so those are going to be difficult 
discussions.
    Senator Carper. Mr. Zinser, do you want to respond to my 
question? When you look at some of the most important lessons 
learned from the 2010 census that the Bureau ought to be 
focused on at this time, would you just again restate those?
    Mr. Zinser. We identified seven challenges looking forward 
for 2020, based on what we saw in 2010. One Bureau priority is 
to stay focused on cost issues. A problem we saw in 2010 was 
the inability of the Census Bureau to accurately estimate the 
costs of its field operations and maintain the transparency of 
its money tracking.
    As a result one important lesson is to develop the cost-
tracking capacity to be able to know, going through the 
operations how much they cost and how much progress or lack of 
progress is actually costing.
    Senator Carper. All right, thank you. Mr. Goldenkoff.
    Mr. Goldenkoff. I would say sustaining the reforms. The 
Bureau, past experience has shown that the Bureau has always 
gotten off to a good start. 1990 got off to a good start, 2000 
got off to a good start, 2010 got off to a good start. And then 
things went awry before too long. Part of that was because, I 
think--and this goes back to the transparency issue--the Bureau 
was always late to acknowledge that there was a problem and 
that created more difficulties down the road.
    I think that is the greater challenge. It is one thing to 
come up with good ideas at the very beginning of the census 
planning cycle, but good ideas and getting off to a good start 
are not enough. The Census Bureau needs to focus on priorities, 
risk management, continue to focus on changing the culture, and 
that is going to be difficult in the years ahead given the 
fiscal uncertainties, the leadership uncertainties.
    Senator Carper. OK, good. I am going to ask maybe one more 
set of questions about the ACS here in a minute, but before I 
do that, our Governor Jack Markell, has just become since last 
Sunday the Chairman of the National Governor's Association 
(NGA), which is a great thing for him and for our State. It is 
a post I once held when I was Governor of Delaware.
    I used to describe the States as 50 laboratories of 
democracy. They are not all the same, but we can learn from one 
another. In fact, we did. As Governors, we used to try to steal 
the best ideas from other States, and frankly, I would gladly 
share it with other States what we were doing, how it actually 
made sense.
    When you look around, again, the world, we look around the 
world at countries like us, maybe a little different from us, 
what are some of the lessons that we have learned from them 
that we actually are incorporating? Maybe what are some of the 
lessons we have yet to learn from them that we ought to be 
incorporating with the idea of better results, less money or 
better results for the same amount of money. Dr. Groves.
    Mr. Groves. Well, I was a party to sort of a summit meeting 
of the English-speaking world.
    Senator Carper. Really? When was this?
    Mr. Groves. Statisticians, several months ago.
    Senator Carper. I will bet that was pretty exciting.
    Mr. Groves. Now, now. Statisticians can be exciting.
    Senator Carper. You all did not have it in Las Vegas, did 
you? [Laughter.]
    Mr. Groves. No, not at all. What I learned is that we face 
similar problems. Our societies, modern societies are moving in 
the direction that is common. We are more diverse because of 
immigration. Every one of the countries is encountering more 
new immigrants. We differ in various ways. By the way, we all 
believe there is a consensus that we must move to multiple ways 
of collecting data simultaneously.
    Senator Carper. So everybody pretty much agrees on that?
    Mr. Groves. That is clear. We all agree that administrative 
records must be a component to our future. Where we differ is 
the legal infrastructure. So if you take Australia as an 
example, the chief statistician in Australia has the authority 
to use, for statistical purposes, all record systems in the 
country, government, private sector, for statistical purposes.
    U.K. has a statistics law that spans all agencies and 
defines what rights and responsibilities a statistical agency 
has. Because of our history of a dispersed structure, we do not 
have unifying legislation like that. And that is, at this 
moment in history, a deal-breaker for our progress, because if 
administrative records are part of the statistical future, we 
are decades behind some of these countries. So that is a 
critical issue we need to learn from.
    The other thing, Brazil, the Brazilians did their census 
with handhelds after visiting the Census Bureau in the late 
2000s, noticing that we were doing handhelds, and they 
succeeded in a very interesting way. We sent a team down to 
research.
    So there are lessons to be learned all over, and our 
decennial folks are trying to learn those lessons quite 
actively now because it is a period of enormous change in how 
statistics are collected and presented and we have got to stay 
up with them.
    Senator Carper. Good, thanks. Mr. Zinser, any comments on 
my question?
    Mr. Zinser. We have not done any comparative analysis of 
how different countries carry out their census.
    Senator Carper. OK.
    Mr. Zinser. We face particular challenges in the United 
States with, with the American Community Survey, especially 
tying it back to the constitutional issues that Dr. Coburn 
raised. I am not sure those kinds of issues exist in other 
countries.
    There are countries, for example, where the citizens are 
actually issued I.D. numbers that make it easier to count them. 
That is never going to happen here. In this respect, the United 
States does have a lot of unique challenges that other 
countries may not have.
    Senator Carper. OK, thanks. Mr. Goldenkoff, anything on 
this particular point?
    Mr. Goldenkoff. There could be lessons learned out there. 
Canada is one possibility, for example. They have been able to 
hold their census costs very steady for a number of cycles now. 
I think this country needs to have a better understanding of 
what is behind that. Is it because their census is conducted 
differently? Are they able--do they accept higher error rates 
than would be acceptable in this country?
    Is it because of something about their culture that they 
are more willing to give up personal information? Or are they 
doing something that the Census Bureau can truly adopt here in 
this country to help us control costs, unique and innovative?
    Senator Carper. Let me just ask, Dr. Groves, 60 seconds on 
that. Tell us what you know, a little bit of what you know.
    Mr. Groves. Statistics Canada is really a sister agency. We 
are back and forth all along. They do something that would be a 
radical change for us in the United States and that is, when a 
small area, a certain percentage of people have responded on 
the census, they will cut off efforts, and those are in the 
90s, a pretty high rate.
    But many people do not know that most of our money is spent 
on those last percentages. This would take a different 
discussion in this country. Are we willing to stop at a certain 
point knowing that it is incomplete, but we have gone through a 
certain predefined set of efforts. We have never had that 
discussion politically.
    Senator Carper. All right, thank you. A couple more 
questions and I will be excusing you. Dr. Groves, some of the 
critics of the ACS have suggested that all or much of the data 
from that survey are available from or could be culled, I 
guess, from other sources, maybe even from data that other 
government agencies collect for different purposes. Could you 
address that for us just very briefly?
    Mr. Groves. My professional judgment is that that is 
incorrect and it is incorrect largely because of ACS's attempt 
to have estimates, statistics on small neighborhoods 
universally, and there is nothing like that available.
    Senator Carper. All right. Is that true, to some degree, 
and how would the Census Bureau go about supplementing what it 
collects in the ACS with data already available from other 
sources? I think you have spoken to this, but just a little bit 
more, if you would.
    Mr. Groves. There are some of the attributes that we ask, 
questions that we ask people that do appear on record system 
that the Federal Government has collected. For those variables, 
we could have alternative ways of estimating. If it were done 
jointly with ACS, it would be even more powerful, in my 
professional judgment, and that is the vision we have for the 
future.
    Senator Carper. OK. If I could, for Mr. Goldenkoff and Mr. 
Zinser, if the American Community Survey is de-funded or made 
voluntary, how would these changes impact planning, designing, 
and executing the next decennial census?
    Mr. Zinser. The biggest impact that we forsee relates to 
how we recommended that the Census Bureau use the ACS as a test 
bed for the 2020 decennial, for example with questionnaire 
content--
    Senator Carper. Would you have it be done over the 
Internet?
    Mr. Zinser. That would be one of the tests, yes; the 
Bureau, in fact, is going to use the Internet for the ACS. With 
respect to Internet response options for the 2020 decennial 
having the ACS in place gives the Census Bureau the opportunity 
to perfect using the Internet for 2020 because it is not 
something that can be put in place overnight. It has to be 
developed. Security measures have to be developed.
    If the Bureau does not have an ACS, it will have to come up 
with another way to develop that capacity. That is one of the 
biggest impacts on the decennial that we have identified.
    Senator Carper. Good, thank you. I appreciate that. Mr. 
Goldenkoff.
    Mr. Goldenkoff. It throws a monkey wrench into the future 
plans because so much is riding on having, as Todd said, the 
ACS as a test bed for all these various systems, as well as 
puts the future costs to possibly raise them, because what has 
happened in the past is that the census systems will be 
designed for use in the decennial, one-time use. They would 
have to perform flawlessly the first time and they would never 
benefit any other survey.
    And so, the whole approach here is to start developing 
systems that can be used for the ACS, for the decennial, and 
other surveys, and greatly reduce costs. And that throws that 
whole plan into doubt.
    Senator Carper. OK, thanks. Earlier I said the last 
question would kick it back to you and just say, is there 
anything here today that you would like to just reiterate, 
whether it is from one of your fellow witnesses or a thought 
that came to mind, parting shots? But do you want to go first, 
Mr. Goldenkoff? Anything you want to underline for us?
    Mr. Goldenkoff. Constructive oversight, that leads to the 
transparency that was so important the last couple of years. It 
has been very effective. We have all been able to get together 
here to talk about problems, solve them as a group. No 
surprises. And that is what is going to be so important going 
forward.
    Senator Carper. OK, thank you. Mr. Zinser.
    Mr. Zinser. I was struck by how focused everybody is on 
cost. When I first arrived at the Department, what I was told 
is that, at the end of the day, Congress is going to give the 
Census Bureau the money that it needs.
    I would be concerned that attitude might resurface, but--if 
the Bureau were to run into the same trouble as with 2010 and 
had to ask Congress for an extra $2 billion for 2020--I really 
question whether or not Congress is going to be able to give 
the Bureau an extra $2 billion.
    As a result, this issue of cost containment is what 
everybody is focused on and that is very important.
    Senator Carper. Good, thanks. I do not know how long I will 
be around here; that is really up to the people of Delaware, 
but for however long, I am here. We are going to focus in this 
Subcommittee and our Committee, at least from my perspective, 
on how do we get better results for less money or better 
results for the same amount of money, just about everything 
that we do.
    Dr. Coburn intimated we are going to be focused on that, I 
think, for probably as long as he and I will be here. Dr. 
Groves.
    Mr. Groves. Well, let me take on something Robert said, my 
own version of it. I think this is a decade where we need help 
from Congress and we need an Executive Branch and Congress to 
come together on matters involving the decennial census. And 
the two principal ones we have talked about, but let me 
reiterate.
    Freeing up our ability to use administrative records to 
save money and improve quality is critical for where we want to 
go. And second, that will lead to a discussion this decade of 
trade-offs, of how much money is worth spending for what 
outcome. And we believe we are trying to assemble the kind of 
information to allow Congress to make wise decisions on that.
    But these are hard decisions to make and it is, if I read 
the Constitution right, the role of Congress to make that 
decision. So our job is to give enough information about these 
trade-off decisions and then for Congress to pay attention at 
the right time. One of the great difficulties we have is that 
2020 seems like forever away in today's world, but the 
decisions we need are going to be much closer to mid-decade to 
allow us to reduce risk on where we are going in 2020. So we 
need your help.
    Senator Carper. Well, God willing and the voters of 
Delaware are willing, you will have it, and your successor will 
have it as well. I want to again echo the comments of Mr. 
Zinser and Mr. Goldenkoff about you and your leadership just to 
say what a joy it has been, a real privilege, for us--and I 
know I speak for Dr. Coburn--to have had a chance to work with 
you and your colleagues at the Census Bureau.
    You are not going to be that far away and my hope is that 
if you get a phone call from us somewhere along the line you 
will take it. My other hope is that you would spend some time, 
and it may be months from now, be willing to spend some time 
with your successor when he or she is confirmed.
    We presume there will be an interim director and hopefully 
somebody could--I am not sure who that is going to be, but I 
think I have an idea. We just hope that you will be willing to 
impart some guidance from time to time to whoever does succeed 
you on a more permanent basis. In the Navy, when people are 
ready to ship out--when do you leave?
    Mr. Groves. August 10 is my last day.
    Senator Carper. And what is today, the 18th? So it is like 
23 days. That is 23 days and a wake-up, so we say in the Navy 
when we are thinking about coming home from overseas 23 days 
and a wake-up and it will be your time to ship out and actually 
sail for home.
    I just want to say, as we say in the Navy, fair winds and a 
following sea. And whenever people, in the Navy, do something 
especially good, well done, we say, bravo zulu, and for you and 
your team, bravo zulu. Thank you so much.
    Mr. Groves. Thank you.
    Senator Carper. I am going to call just a brief recess and 
I am going to ask our next panel to go ahead and take their 
place and then we will resume as soon as I get back. Thank you. 
[Recess.]
    All right. Let us go ahead and resume our hearing. Thank 
you for understanding the brief recess. Dr. Jason Providakes. 
Did I get it close? Is that correct?
    Mr. Providakes. Providakes.
    Senator Carper. So I got it right actually. Senior Vice 
President and the General Manager of the Center for Connected 
Government at MITRE. He leads the delivery of MITRE expertise 
to civilian agencies through the federally-funded research and 
development centers. The goal for the Center for Connected 
Government under his leadership is to provide enduring 
technical capabilities supporting the missions of civilian 
agencies accountable for leading critical national challenges.
    Mr. Providakes holds his doctorate from Cornell University, 
along with his master's and bachelor's degrees in electrical 
engineering from--is it Worcester?
    Mr. Providakes. Worcester.
    Senator Carper. Worcester. All right. I should know that. 
Worcester. I apologize. Thanks a lot for joining us today and 
for all of the help that MITRE has provided.
    Dr. Jack Baker, an easier name to pronounce, is Senior 
Research Scientist at the Geospatial and Population Studies 
program at the University of New Mexico and serves as New 
Mexico's representative to the Census Bureau programs on 
population estimates and projections.
    He participated extensively in the preparation for the 2010 
census and chaired the group to redesign the 2010 count review 
program. In the fall of 2009, he was named a member of the 
National Academy of Sciences Panel to Review the 2010 census. 
He holds a B.A. from the University of North Dakota. Are you a 
native of North Dakota?
    Mr. Baker. No, sir.
    Senator Carper. All right. And M.S. and Ph.D. degrees from 
the University of New Mexico, all in anthropology. We are glad 
you are here. We have got three of them, three doctors here.
    Dr. Andrew Reamer, Research Professor at George Washington 
Institute of Public Policy at the George Washington University 
(GWU) where he focuses on Federal statistical policy and 
programs, sources and uses of socioeconomic data, State and 
local economic analysis and strategy, and the geography of 
innovation and other regional studies. That is a lot. That is a 
mouthful, is it not?
    Mr. Reamer holds a bachelor's degree in Economics from the 
University of Pennsylvania, right up the road from where I 
live, and a Ph.D. in economic development and public policy 
from MIT, where our oldest son went to school. He is a lot 
smarter than his dad, I can assure you.
    But we are happy you all are here and welcome your 
testimony. It is an important hearing. I am glad that Dr. 
Coburn suggested it. He will be back if he can and we will just 
say your whole statement will be made part of the record and 
feel free to proceed. I let the first panel go on for a while 
and we will do that a little bit, but I have some time 
constraints at the end here, so we will not be able to go quite 
as long. All right. Dr. Providakes, welcome and thanks so much.
    Mr. Providakes. Thank you and I can actually keep to the 5 
minutes to allow questions. I know that.

TESTIMONY OF JASON PROVIDAKES, PH.D.,\1\ SENIOR VICE PRESIDENT 
           AND GENERAL MANAGER, THE MITRE CORPORATION

    Mr. Providakes. Chairman Carper, thank you for the 
invitation to produce a more complete and effective enumeration 
of the census. Let me talk about that today for you. As you 
know, MITRE is a long-standing partner with the Department of 
Defense and many civilian agencies; the Internal Revenue 
Service (IRS), for example. We also work with the Department of 
Veteran Affairs (VA), the Federal Aviation Administration (FAA) 
and the Department of Homeland Security (DHS).
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Providakes appears in the 
appendix on page 99.
---------------------------------------------------------------------------
    As you mentioned earlier, MITRE is a not-for-profit 
company. Our sole activity is the operation of federally funded 
research development centers (FFRDCs), which operate for the 
benefit of their Federal Government sponsors. As alluded to 
earlier, our expertise is in scientific research and the 
analysis, development, and acquisition of system engineering 
and integration.
    I am here today sharing with you my perspective as a system 
engineer and as a technologist derived from many years of 
experience in large and complex systems, and as a contributor, 
also, to several scientific advisory boards.
    MITRE supported the Census Bureau in preparation for the 
2010. We continue to work with the Census as they prepare for 
the 2020. Our role in helping to mitigate large risks that 
developed during the 2010 program informs my comments today. So 
I am going to try to get to those lessons learned that we 
alluded to earlier.
    The single most important question, in my view the Census 
needs to think about is--and indeed all government departments 
and agencies--how do we effectively and affordably capture 
technology innovation. I heard that most of today. How do you 
capture that innovation in terms of bending that cost curve 
that some of us are concerned with.
    The good news is that today there is no shortage of 
hardware, software, and middleware that enables more efficient 
and accurate census taking that does exist. We are in the midst 
of geometric growth of technologies, in fact, and will continue 
through 2020. In fact, by 2020, we at MITRE join many others in 
predicting that the snapshot of today's available technologies 
will feel as antiquated as, for example, the rotary dial phone 
did in 2010.
    The bad news is that challenges of technology selection and 
implementation in order of magnitude are more complex than they 
were in previous planning and R&D cycles. So how can the Census 
Bureau effectively and affordably capture the value from 
technology innovation? That is really the question we are 
trying to address.
    Our experience suggests there are two areas of operational 
changes in order for the Census to be more effective. I will 
try to address those two quickly. The first is--we heard this 
theme again--engineering for a data rich ecosystem. These 
datasets today across government are inter-dependent and lend 
themselves to some very valuable insight--
    Senator Carper. Let me just interrupt you here. You can 
slow down just a little bit. Do not feel you have to rush. OK?
    Mr. Providakes. All right. Of that data rich ecosystem, and 
that is what we are trying to get to, my hope is that 
technology exists today. That for a paperless census, the 
technology will likely exist through 2020 for the automated 
census. I should also point out, without the physical storage 
space and logistics required for massive amounts of paper, the 
field infrastructure which lends itself so much for cost, can 
be radically reduced and re-envisioned.
    In MITRE'S extensive experience with the IRS, FAA, and 
others in the government, we found it is most effective with 
regard to enterprise modernization when ownership and control 
of this technical baseline exists within the agency. I will say 
more about that. The technical baseline effectively defines the 
capabilities and characteristics needed to deliver specific 
outcomes and guides the priorities for an acquisition program.
    It permits continuous trade space analysis of the optimal 
design. Our observation is that successful programs that use 
engineering, particularly enterprise engineering, and the 
technical baseline as a compass for navigating uncertainty and 
complexity of tomorrow's technological advances.
    The second area is really this alignment of budget cycle 
and spending to realities of technology planning and 
acquisition. The census is a 10-year planning cycle. MITRE 
continues to recommend that agencies be given flexibility to 
adapt funding to react to changes in technology requirements.
    This should include multi-year authority and the authority 
to fund the up-front systems engineering and necessary trends 
of analysis and to evaluate and estimate the scope, cost, and 
schedule of their proposed investments without prior approval.
    This up-front investment is not trivial. In fact, perhaps 
up to 20 percent of the total life cycle costs of an 
acquisition would be expensed. Overall program success is 
highly correlated to early investment in concept development 
and systems engineering. It is critical to successful 
execution, but today not available to the management budget 
process.
    There are two specific benefits really to this budgeting 
flexibility. First, it would permit utilization of contemporary 
risk mitigation techniques like beta testing we heard earlier. 
How do you test? Agile development, which we talked about, 
identification of trade space and negotiation, open source, and 
non-proprietary solutions, to name a few.
    Second, capability development could be undertaken on a 
continuous improvement basis rather than the point in time 
delivery basis we see many times by programs. I think those two 
areas and recommendations should be considered and I welcome 
your questions. Thank you.
    Senator Carper. Thanks so much. Dr. Baker, please.

 TESTIMONY OF JACK BAKER, PH.D.,\1\ SENIOR RESEARCH SCIENTIST, 
GEOSPATIAL AND POPULATION STUDIES, THE UNIVERSITY OF NEW MEXICO

    Mr. Baker. Thank you, Mr. Chairman. As you said in your 
introduction, I am a member of the National Research Council 
panel to review the 2010 census. Last year, the panel's chair, 
Tom Cook, reviewed the findings of the panel's first report, 
``Change and the 2020 census, Not Whether, But How.'' Today I 
would like to continue the conversation that Dr. Cook started 
and offer further suggestions on 2020 planning. 
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Baker appears in the appendix on 
page 105.
---------------------------------------------------------------------------
    From my own background as an experienced demographic 
methodologist, I need to make the disclaimer that my opinions 
in this testimony and this hearing are my own and should not be 
construed as formal word from the panel or the National 
Academies.
    It is fair to say that the panel supports the basic 
conclusion from its first report that the 2020 census can and 
should be conducted in a way that sees large scale reductions 
in costs per housing unit while maintaining quality. In the 
report, we identified four priority areas for 2020 R&D. These 
include applying modern operations engineering to field 
operations and making fuller use of administrative records.
    My comments today mainly cover the other two areas, 
emphasizing multiple response modes to the census, particularly 
the Internet, and continuous improvement of the Bureau's 
geographic resources. I think the Bureau has embraced the 
notion that ongoing testing and experimentation are important 
aspects of 2020 planning.
    The Bureau is working on a more adaptive process in 
operations and field management for its data collection 
programs. This will be tested and implemented first in the 
Bureau's other surveys and then eventually form 2020 census 
systems. Under this process, respondents may move between 
different response modes and interviewer approaches based on 
past contact attempts and contextual information.
    This process of rethinking census taking as a more organic 
process rather than a string of only loosely integrated 
operations is crucial. It fosters the kind of exploratory and 
bold thinking that I think is necessary for the Bureau to meet 
its challenges.
    For decades, the Bureau has tended to layer on more and 
more operations, often in the name of improving overall 
quality, without stepping back to consider costs and benefits 
and cost/quality trade-offs. I think that a management 
framework built on adaptive design can allow a better 
understanding of redundancies of effort and the resulting need 
to prioritize. My hope is that analogs of the same adaptive 
design may be brought to bear in other parts of the census 
besides managing contacts.
    That said, successful retooling of the census is only 
possible if 2020 planning and R&D activities that will directly 
inform it are made an adequately funded priority. I believe the 
short-term up-front costs associated with 2020 planning are 
worthwhile investments with major long-term cost offsets.
    Gains from changes in the census processes will be 
undermined if the Bureau's geographic resources are inadequate 
if individual census responses cannot be accurately linked to 
precise geographic coordinates specifically. Therefore, 
effective means for updating the Bureau's geographic data 
resources, its master address file, and its TIGER geographic 
database are a key aspect of any census design effort.
    To this end, the Bureau faces a major decision on the 
extent to which it will conduct address canvassing prior to 
2020. As you know, the Bureau sent staff to every block in most 
of the country in 2009 to verify or correct address list 
entries, and as others have noted, that is quite expensive to 
do.
    Looking ahead to the decision for 2020, the Census Bureau 
launched its Geographic Support Systems Initiative, which the 
panel generally endorsed in our first report. As this work 
progresses, I suggest that, first, there is a pressing need for 
quality metrics for both MAF and TIGER. To its credit, I think 
the Bureau has been candid in noting shortcomings in 
recognizing the need for improvement. This is just to say that 
there is a danger in slipping into the mind set that the 
Bureau's own files are some sort of unassailable gold standard 
which they should avoid.
    Second, the Bureau's geographic research should focus on 
the coverage properties of MAF and TIGER and those of 
alternative resources as others have mentioned. Those 
alternative sources include commercial databases such as Google 
maps as well as additional data from the U.S. Postal Service 
and other agencies. In a nutshell, my point is that it is 
exceedingly unlikely that there will be one single source that 
is universally best.
    The Bureau's research should consider differences in 
quality and their effect on the resulting accuracy across 
different sources and for specific geographic and demographic 
subgroupings.
    Third, I think that the Census Bureau should consider the 
same kind of adaptive or flexible approach for updating its 
geographic resources as it hopes to implement in field 
operations. In approaching the address canvassing, my hope is 
that the Bureau can avoid the one-size-fits-all approach that 
has driven operations in the past.
    The geography and the housing address stock of downtown 
Chicago is different from that in the pueblos in New Mexico 
where I reside. Likewise, I think that developing models of 
MAF/TIGER coverage and using them to strike a balance between 
field collection and tapping existing resources should lead to 
different approaches for those areas, too.
    Fourth, the Census Bureau can learn a great deal from 
outside its walls, from private and public sector organizations 
faced with similar challenges, as well as from statistical 
agencies in other countries. The Bureau should consider the 
techniques used by commercial MAF vendors in updating their 
products, draw from the experience of firms such as UPS, and 
study the specific operations conducted by agencies such as 
Statistics Canada.
    In closing, I understand that recent developments involving 
the ACS are a secondary topic for today's hearing. As a regular 
user of ACS data, my personal hope is that the Senate will undo 
the appropriations amendment passed in the House. But in 
keeping with the main theme of this hearing, I would like to 
close by emphasizing that a healthy ACS is critical to an 
improved 2020 census. The ACS will be a critical proving ground 
for the adaptive process I spoke about earlier prior to ruling 
it out in 2020.
    I think that one flaw in 2020 planning to date is that the 
role of the ACS has not been fully fleshed out and trust that 
that will change. But, of course, the ACS cannot be a test bed 
for 2020 if it does not exist or exists in a severely hobbled 
form. I thank you again for the opportunity to testify and I 
welcome your questions.
    Senator Carper. Thanks, thanks very much for those words. 
Dr. Reamer, you are recognized, please.

   TESTIMONY OF ANDREW REAMER,\1\ PH.D., RESEARCH PROFESSOR, 
GEORGE WASHINGTON INSTITUTE OF PUBLIC POLICY, GEORGE WASHINGTON 
                           UNIVERSITY

    Mr. Reamer. Chairman Carper, I appreciate the opportunity 
to discuss recent developments concerning the American 
Community Survey. I will indicate why ACS cancellation would be 
destructive; review the negative consequences of a voluntary 
ACS, including exacerbating rather than reducing constituent 
concerns; and then I will offer an alternative approach to 
addressing those concerns.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Reamer appears in the appendix on 
page 110.
---------------------------------------------------------------------------
    For decades, small area census data have been essential to 
the proper functioning of government, the economy, and our 
communities. Today, for example, the Federal Government uses 
ACS data to build statistics such as State personal income and 
annual population change, design, implement, and evaluate 
policies and programs, distribute over $450 billion in 
financial assistance across the States, and enforce the Voting 
Rights Act.
    State and local governments use ACS data to allocate scarce 
resources, calculate caps on taxing and spending, and redraw 
legislative districts. Economic development groups, businesses, 
non-profit organizations, and research organizations use ACS 
data to make informed decisions in their respective domains. 
And the public uses ACS data to understand changes in their 
communities.
    The origins of the ACS begin with Congressman James 
Madison, who on the floor of the House in 1790 said he wished 
that each census would gather information beyond bare 
enumeration so that future Congresses can adapt public measures 
to the particular circumstances of the community and mark the 
progress of society.
    For 222 years, Congress has fulfilled Madison's wish, and 
in 2004 approved the long-sought collection of census data more 
often than once a decade, through the American Community 
Survey. The advent of the ACS has made possible more informed 
public, business, and personal decisions in light of widespread 
long-term reliance on long form-type data. The survey's 
cancellation would cause economic disruption, misapplication of 
scarce community resources, and increase waste of government 
funds.
    In terms of the issue of voluntary versus mandatory 
response, since 1790, the government has had the authority to 
fine anyone refusing to answer census questions or providing a 
false response. While current census law says the fine for not 
responding to the ACS can be up to $100, this cap has been 
superseded by a uniform sentencing act that Congress passed in 
the 1980s that sets the limit of the fine at $5,000.
    Members of Congress are hearing several types of complaints 
from constituents receiving the ACS. Some constituents 
experience the questions as an invasion of privacy; some 
distrust the government's use of the data; some feel terrified 
or coerced by a possible $5,000 fine; and some who do not mail 
back the form feel harassed by Census Bureau staff.
    Several members have responded to these complaints by 
proposing to prohibit the government from imposing a fine for 
ACS non-response, in effect making the survey voluntary. 
However, a voluntary ACS would have the perverse effect of 
aggravating, not eliminating, constituent concerns.
    In memos, the Census Bureau says that to produce 
sufficiently reliable small area estimates, it needs to 
maintain the current number of completed ACS surveys and that 
failure to reach that number would lead to, quote-unquote, 
unacceptable estimates. It also says that based on field tests, 
a voluntary ACS would lead to a 20 percent drop in the mail-
back response rate.
    The Bureau has asked, under a voluntary ACS, what changes 
would need to be made to get the current number of finished 
surveys, and the answers, as I understand them reading through 
the memos is, one, 23 percent more households would get an ACS 
form. So a sample almost a quarter larger.
    Two, 18 percent more households would get a follow-up phone 
call because if the response rate falls, there are more phone 
calls. And then three, 39 percent more households would get a 
visit from Census Bureau staff. So if the goal is to reduce the 
number of touches of the Census Bureau, to households this is 
going in the opposite direction.
    And then finally, the cost of these extra efforts would be 
about between $70 and $100 million added to the current $242 
million base to get the same number of completed surveys. So I 
do not think these are the impacts of a voluntary ACS 
proponents are seeking.
    And in light of these findings, I suggest an alternative 
approach that relies on better information in communications 
and a minor tweak, a legal tweak, while maintaining a mandatory 
response. First, to address constituent concerns about privacy 
and data misuse, I suggest that the Census Bureau offer the 
public access to information on ACS uses by State and place.
    With spidering technology, the Bureau could build a 
searchable online database with links to thousands of national, 
State, and local ACS applications. My belief, and it is a 
testable one, is that when seeing these applications, 
recipients would be more willing to complete the survey.
    Second, I encourage the Bureau to create an ACS partnership 
program modeled on the one it created for the decennial census. 
The Bureau would find and train trusted third-party 
organizations willing to give constituents information and 
reassurance on ACS data uses and confidentiality.
    To address concerns about the $5,000 fine, I suggest that 
Congress exempt the Census Bureau from the uniform sentencing 
statute and allow it to revert to fine caps of $100 for non-
response and $500 for false statements, caps that were set in 
1929, by the way.
    I recommend that the Bureau review and revise staff 
protocols and incentives so that non-respondents do not feel 
harassed. The Bureau might consider creating a hotline or an 
ombudsman for constituents.
    And finally, I encourage the Census Bureau to communicate 
more with Members of Congress about the ACS. The Bureau could 
periodically provide examples of recent ACS uses in a Member's 
State or district, provide updates on efforts to encourage 
constituent response, and with each ACS release, provide the 
new socioeconomic profile of each Member's State or district.
    In these several ways, I think constituents' discomforts 
with the ACS can be addressed while avoiding steps that 
compromise the integrity of the data.
    So to conclude, going back in history once again, in George 
Washington's 1790 State of the Union message, he offers a 
statement rich with relevance for the management of the ACS, to 
paraphrase, In every country, knowledge is the surest base of 
public happiness. Increased knowledge contributes to the 
security of a free Constitution in multiple ways, including 
teaching those in authority the need to gain the enlightened 
confidence of the people and teaching the people to distinguish 
between necessary exercise of lawful authority and oppression.
    I believe that with the Subcommittee's guidance, the Census 
Bureau can find an approach that results in constituents 
experiencing a proper balance between an individual's rights 
and duty to community and Nation. Thank you and I look forward 
to your questions.
    Senator Carper. You bet. Thank you very much--let me just 
ask Dr. Providakes and Dr. Baker just to react briefly to what 
you have heard from Dr. Reamer. Any particular aspect of his 
testimony that you just want to underline, rebut--
    Mr. Providakes. I would not say rebut.
    Senator Carper [continuing.] Hold up for ridicule?
    Mr. Providakes. We do throw technology around a lot as sort 
of the lynchpin to many of these solutions, almost like the 
Savior in terms of how do we deal with some of these issues. 
And it reminds me of the topic we just heard about, the 
Internet, and they want to use the Internet. To do that, to be 
informed for doing some experimentation or testing of that with 
a large population, I think, is critically important.
    I think if you go back and look at what happened during the 
FDCA days, there was a vision and a view of handheld devices, 
but we did not really have full understanding of in terms of 
how people and their behaviors and how they do their jobs, 
actually utilize those. The same thing, I think, with the 
populace, understanding what people will do when you give them 
surveys and tell them to come to the Internet and come to a 
location.
    Getting some experience from that would have a lot to do 
with understanding not just that modality, but we heard about 
other modalities as well in terms of wireless devices. Getting 
some experience with that from a Census construct, I think, 
would add a lot of insight and value in terms of what that 
final architecture would look like going into the 2020.
    Senator Carper. OK, thanks. Dr. Baker, just a brief 
reaction to some of what you heard from Dr. Reamer?
    Mr. Baker. Well, I think my experience is most relevant to 
what Dr. Reamer was mentioned, was in working in the Local 
Updated of Census Addresses Program in 2007-2008, in which 
there was a large interplay between State and local governments 
and the Census Bureau in an interchange of address data.
    I think my experience reinforces Dr. Reamer's comment that 
in some sense, the local governments, and this extends down to 
the constituents of those governments who are hesitant about 
these responses, not see a reward specifically for them in 
filling out the response. In other words, there is a burden, 
but they do not see how there is something within those 
returns.
    I think building the sort of partnerships that Dr. Reamer 
had mentioned becomes essential because people need to 
understand and see that there is something concrete that they 
get out of their participation. So I would just reinforce that 
comment and I do not have any other.
    Senator Carper. OK, good. Thanks. Any response to what you 
just heard there, Dr. Reamer? No? OK, good. Maybe I will go 
back to a couple questions I would like to ask. The first one 
would be for Dr. Providakes, if I could, and we will work you 
in this line up, Dr. Baker, and maybe Dr. Reamer, too.
    In your testimony, you describe how it is important that 
the Federal agencies effectively and affordably capture value 
from technological innovation. You pretty much said that again 
here. Could you again help us understand how the Census Bureau 
could use cloud computing in its 2020 design? And also, what 
specific applications or services do you see the Census 
contracting for?
    Mr. Providakes. Well, I think clearly from a cloud 
computing perspective, that really talks to the fundamental 
infrastructure of how the Census collect data and would share 
data. One of the challenges I think one would find from a 
lesson learned from the 2010, when we did eventually rule out 
these handheld devices, the fundamental infrastructure that the 
people were talking to--there was a fundamental performance 
topics associated with that.
    What cloud computing eventually does, and I think the 
Director, Dr. Groves, mentioned that, is that it allows one to 
scale affordably in time for your needs. As you attempt to 
build out this infrastructure for what you think that 2020 
architecture will look like, cloud computing gives you an 
affordable way of doing that. You do not have to buy it all at 
once tomorrow. You can buy as you need.
    So that provisioning is a very powerful tool in terms of 
agility as the technology evolves. So by 2020, I would suspect 
you would see quite a bit of advancement in public clouds. I 
think you would also see significant advancements in whether we 
call them private or hybrid clouds where there is some 
dimension that is closed, but then again, there is some access 
to that hybrid cloud by other agencies and institutions in 
terms of sharing information.
    So that is just one example of how cloud computing can help 
bend that cost curve. Its agility of provisioning technology, 
when you need it and size and cost.
    Senator Carper. All right. Dr. Baker, Dr. Reamer, do you 
want to respond or react to what we just heard from Dr. 
Providakes?
    Mr. Baker. Well, I would respond just to reinforce and add. 
My special area of expertise is really in geographic mapping 
resources, and from the perspective of sharing information back 
and forth between the Census Bureau, that allows them to 
enumerate effectively, and local governments, who have 
information that would help them to do that, building this sort 
of infrastructure makes a lot of sense in that, because often 
times, it is very difficult to detect, for example, like with 
like a targeted address canvassing sort of program where you 
should be looking, but locals always know where you should be 
looking.
    And so, this sort of infrastructure would allow that to be 
done efficiently and at a very low cost and very effectively.
    Senator Carper. Thank you. Dr. Reamer, anything you would 
like to add or take away?
    Mr. Reamer. I am not a technical person or a statistician 
or an engineer, but I am doing something that is complementary 
to this.
    Senator Carper. OK.
    Mr. Reamer. In May, I held at G.W. a data fair.
    Senator Carper. I am sorry. Say that again.
    Mr. Reamer. A data fair where we had 50 different 
organizations that provide, in an innovative way, data that was 
not possible to provide even 10 years ago. We had Federal 
agencies and we had the private sector. We had Amazon, we had 
Microsoft, we had big data organizations. And the purpose was 
to have people meet each other.
    And so, I think Dr. Groves can attest to this, he was 
there, it was basically a 2-day opportunity for people, through 
serendipity, to find technologies, both from the public side 
and the private side, with synergies, I hope to get funding to 
do the fair again. I think we learned that a number of 
agencies, I believe including the Census Bureau, found some 
positive connections there in relation to cloud computing and 
other activities. Our role at G.W. in facilitating this 
process.
    Senator Carper. OK, thanks. If I could come back to Dr. 
Baker, for the 2010 census, the Bureau dropped plans to use the 
Internet as a method for collecting data. What challenges might 
the Bureau face with an Internet response option for the census 
in 2020?
    Mr. Baker. Well, I think first off, the consideration of 
Internet would be a considerable improvement. It is probably 
the one thing that I can think of that could reduce costs, 
wholesale across the board in taking the census. The main 
challenge that I see with this--actually comes from where the 
sort of work that I do--is in geographic location of 
respondents.
    The Census is geographically grounded because it is used 
for apportionment, but often people are unaware that locally 
within the State it is also used for recalculating voting 
districts. And so that becomes important about where you place 
them on the ground, and there are a lot of open methodological 
questions about how you do that when it is an Internet-based 
response.
    They have given us a lot of thought. It is done in other 
places, so I am not saying it is insurmountable, but it does 
present a challenge that may be specific and unique to a highly 
mobile population such as we have.
    Senator Carper. OK, thank you. Dr. Providakes, if you want 
to take a shot at that question as well?
    Mr. Providakes. I think on the Internet perspective, there 
is this view--that there is this significant cost saving that 
is going to come with going with one of the modalities of using 
the Internet as a replacement to, for example, mailings. I 
think that is a legitimate modality to look at. I think it is 
going to provide significant improvements, not just in cost, 
but, for example, in quality and engagement with the 
population.
    As I think Dr. Groves alluded to, the significant cost that 
I found, particularly in 2010, generally has to do with that 
smaller population, the ones that you have got to go out and 
touch, and the question is, do these modalities like the 
Internet help you really improve on that significant cost 
factor that we struggle with? It is that last 20, 30 percent.
    Today I think I saw a number of something like 70 percent 
people did mail in. I would expect that same sort of number of 
users would use the Internet. Can the Internet or other 
modalities get at those other elements where you try to go and 
touch where you have a higher massive number of people and hit 
the streets and actually count and get information from?
    That is where I think the 2020 time frame technologies of 
different modalities, people with wireless devices getting 
access to the Internet. Absent, then provide opportunities not 
just the Internet, but other modalities to get at those 
remaining 20 or 30 percent non-response.
    Senator Carper. OK, thank you. I want to come back to the 
American Community Survey and pose maybe a couple of questions. 
Before I ask a question, let me just set it up this way, Dr. 
Reamer, if I could. This would be just for you. In recent 
months--and we have talked about this a fair amount today, but 
in recent months, there have been several proposals put forth 
in Congress that call for the elimination of the American 
Community Survey or to prohibit the government from penalizing 
those individuals that decline to respond to the survey.
    I would just ask, you talked about this a little bit 
already, but I just want to ask you more directly. How would 
the loss of the American Community Survey affect the cost and 
the scope of planning for the 2020 census, as well as future 
decennial censuses, and other Census Bureau surveys and 
estimate programs?
    Mr. Reamer. Well, in terms of the relation between the ACS 
and planning for the 2020, I do not feel I have the expertise 
for that question. But if there were no ACS, Congress would 
then have to make a decision, does the Nation go back to a 2020 
long form? And so, that would be an extra expense. I do not 
know if this is actually the case, but when the Census Bureau 
asked Congress to fund the ACS, it said it would make every 
effort to make the change from the long form to the ACS cost 
neutral.
    And so, the cost over a decade is $2.4 billion, so that 
could add that amount of money again, people at Census could 
tell you, but it would be an extra expense. If there were no 
ACS, the impact on the decennial would be the least--it would 
be an issue, it would be the least of the issues, because so 
much of the operation of the public and the private sector rely 
on the ACS for decision making in ways that are obvious and 
ways that are not obvious.
    So, for instance, the Medicaid reimbursement formula that 
Delaware gets is based on a Bureau of Economic Analysis (BEA) 
state per capita income, and the richer the State is, the lower 
its reimbursement rate. So per capita income is total income 
for the State divided by its population, and each of those 
numbers depends on the ACS.
    Total income. There are a lot of people who live in 
Delaware but do not work in Delaware. The ACS allows BEA to 
adjust people's place of work earnings to place of residence 
earnings so that Delaware gets the right amount of total income 
for the State.
    On the population side, the ACS provides the data for 
international migration to the United States. So both those 
numbers, and then the ratio, which is how Medicaid's 
reimbursement is determined, depend on the ACS. The ACS goes--
we just would have to rely on old ACS data until 2022 or 2023 
if there is a long form. And I can go on, there are many 
examples.
    Senator Carper. OK, thank you. A follow-up question if I 
can. We know that many Americans who were asked to complete the 
ACS do not understand the rational for a number of the 
questions and how the data that they are being asked to provide 
is actually going to benefit their communities, their quality 
of life. Any ideas? And this would be a question for you and 
maybe for other panelists as well. But what could the Census 
Bureau do? Any specific ideas that you have to better explain 
to the American people the purpose not only of the survey 
generally, but of each question, as well as the data each 
question yields? Any advice?
    Mr. Reamer. Yes. In the testimony, I mentioned two ideas 
and I probably can think of a few more right here, but one is 
to create a uses database. I gave each Member of the 
Subcommittee a packet. In the packet is a 10-page listing of 
the uses of the ACS that I put together in 3 days.
    The Census Bureau has the ability to find uses on the web 
that are specific to a community. In your packet also are 
examples of uses specific to Delaware. We just did a search on 
Delaware and American Community Survey. So the Census Bureau 
could, in its cover letter, say, If you would like to learn 
more about how the data are used in your community, go our Web 
site, put in your ZIP code, and we will provide you with links 
to examples.
    The Census Bureau could tag the examples by category. If 
people are interested in health or people are interested in 
bicycling or people are interested in disaster management, that 
those uses for Delaware could come up. So that is one.
    The second way, I think, is, as I mentioned, having an ACS 
partnership program so that there are Delaware-based partners, 
whether it is community groups or the State of Delaware or the 
City of Wilmington, or any of the above. I think there were 
over a thousand partners for the decennial census who would be 
trusted, who would be seen as trusted by local constituents, 
who could call them and ask, Is this for real, how is this 
stuff being used, and get a third party opinion about that.
    And then I think the Census Bureau could also, on its Web 
site, take each question and have someone click on that 
question and show nationally just the wide array of ways that 
the data collected from that particular question, whether it is 
flush toilets or journey to work, are used. So, I think in all 
those ways, the Census Bureau can make it easy and accessible 
for people to get information.
    Senator Carper. Gentlemen, anything you want to add to 
that?
    Mr. Providakes. I would like to just echo that. There is 
great opportunity here in 2020. This is where I think the power 
of the Internet really comes into play. If, in fact, you put 
out an ACS on the Internet, you can almost envision where I go 
to fill out my survey, there is not just a form that pops up, 
but there are other links. There is information. There is a way 
to incentivize the user to better understand why Census is 
asking--what is the meaning of this question, what is the value 
of answering those questions.
    It gives more interactiveness as opposed to even a letter 
or a form in the mail. It is an opportunity to really help 
inform the people who are filling out the forms why the Census 
is important. It is at the point of experimenting with some of 
that behavior early to get a sense of what will people react 
to, what are they missing that is inhibiting them to just want 
to fill out or answer certain questions.
    You really gain a lot from that knowledge and you can 
actually see that reaction just like social networking does 
today, to how people are reacting in what districts, how they 
are reacting to certain questions and the like. So I think it 
is an opportunity to just think a little bit out of the box in 
terms of what the Internet could bring from that perspective.
    Senator Carper. Thanks. Dr. Reamer, did you want to add 
something?
    Mr. Reamer. Yes. The ACS is facing a start-up challenge now 
because it is being sent to homes in the absence of any to-do 
about a decennial census. People have received a long form 
since 1960, but it was always preceded by a Presidential 
proclamation, that it is everyone's duty to fill out the form. 
So, when people got it, it was, Oh, geez, I got the long form. 
One out of six households got the long form.
    Now, people are getting an American Community Survey and 
they do not have enough context. So it is easy to be fearful, 
it is easy to lack understanding, it is easy to have concerns 
because there is no infrastructure--informational 
infrastructure around it. One of the challenges of the Census 
Bureau is to build up recognition and trust and information and 
familiarity with the American Community Survey.
    Actually, the Census has faced this issue before, in the 
1790 census. All right? There was an argument in Congress--not 
argument--debate about a concern: Would people fill out the 
census form, because the Nation had never done a national 
census before? Well, it turns out that people in the north, in 
what used to be the northern colonies, Massachusetts and New 
York, were comfortable with the census because as colonies, 
they had censuses before, but people in the south had never 
seen a census and there was more difficulty.
    So it took a couple times for all the States to understand 
the value of the census, and in some ways, the ACS is facing 
the same situation that the original census did in 1790.
    Senator Carper. OK. Before we wrap up, you heard the first 
panel testify and then respond to questions and have some 
discussion among themselves with us, and we have had a little 
bit of that here. Anything you may add? I just like to ask for 
closing statements. Just maybe take a minute, if you would, 
something you would like to underline, emphasize, change, or 
just something you thought of. But I would welcome that. I 
really do not care who goes first. Dr. Baker, would you like to 
go first?
    Mr. Baker. Sure, I would be happy to. Thank you. I think 
the two things that stand out for me are a commitment to early 
planning, which I think is evidenced in the recent actions of 
the Bureau, so I would want to reinforce that. And also 
reinforce the idea of a more flexible, adaptive planning and 
operational set of processes that are enabled by existing 
technology, but that are really driven by the desire for 
accurate information.
    And so, balancing the use of technology and the gaining of 
statistically valid information, I think, really has to go in 
the direction of that process to be successful in the world 
that we live in today, and those two points, I think, are 
worthy of a take-home message.
    Senator Carper. OK, good. Thanks. Dr. Reamer.
    Mr. Reamer. Yeah. Two things. One is, in looking back over 
Congress's discussions about data collection in the 18th and 
19th Centuries, I was struck by the passion that these Members 
of Congress had about data collection, because it was so 
difficult to do. And they just kept trying and trying, both on 
the demographic side and the economic side.
    And finally, by the end of the 19th Century, they had 
figured a lot of things out, the Executive Branch did, and I am 
struck by the difference between that and the lack of passion--
    Senator Carper. The what?
    Mr. Reamer. A lack of passion, to a large degree, about 
some Members of recent Congresses regarding data, because data 
have become like the plumbing. There is, I think, a great 
under-appreciation of the value of data, of the teeny amount of 
money it costs the Federal Government to collect data that have 
orders of--uses valued at orders of magnitude beyond the 
original cost. You could not run the country without the 
decennial census.
    So, for instance, Mr. Webster, who introduced the amendment 
to prohibit carrying out the ACS, oversaw the redistricting 
process in Florida when he was Senate Majority Leader there, 
and set up an open process for people to go online, My 
Redistricting, and you could use software to draw districts in 
Florida.
    Well, guess what those data are based on? The American 
Community Survey. Because you need the ACS for the citizenship 
data to be compliant with the Voting Rights Act. And then the--
so, that is an example. I think Mr. Webster does not really 
understand the extent to which the ACS data are necessary.
    The second point is this. The Census Bureau--we were 
talking culture in the first panel, about the culture of the 
Census Bureau--and for decades, the culture of most Federal 
statistical agencies was one of production shops. Right? It was 
their job to just produce the data. Experts, the Council of 
Economic Advisors, Congress in developing formulas for 
allocation, they would use the data.
    It was not the Census Bureau's role to figure out how the 
data should be used or even whether the data should be 
collected. That has changed. Right? With the Internet, all of a 
sudden all these data are accessible and 20 years ago for me to 
look at census data, I had to go to a Federal depository 
library. Right?
    Now it is instantaneous. It has been instantaneous for 15 
years, but the Census Bureau is delivering data in a way--there 
are parts of the Census Bureau culture--that don't understand 
that the Census Bureau has spun off as an independent 
organization. Parts of the Census Bureau still think they are a 
back-office shop.
    I think one of the great accomplishments of Dr. Groves has 
been to have the Census Bureau recognize that it is in 
business. It has customers. It needs to understand what its 
customers are looking for in terms of data, customers including 
Congress, but the public as well. And to have a back and forth, 
have a relationship with the users. This historically has not 
been the case because historically it was very difficult to do. 
So I think in this day and age, it is necessary for that 
cultural change.
    Senator Carper. Something you just said. A question I often 
ask myself, even in government, who is our customer? And in 
thinking of the Census Bureau, the customers are obviously 
businesses, they are State and local governments that are 
involved in redistricting, and there are variety of others. But 
in a way, the customers are the people who are going to be 
filling out the surveys, because if they do not provide the 
information in a timely and accurate way, then what is being 
provided to the other customers that we have alluded to is not 
going to be as valuable. Dr. Providakes, last word?
    Mr. Providakes. I do want to echo the whole data topic and 
what I said the two key elements are, one of them being 
engineering a data rich ecosystem. I can just imagine by 2020 
such an opportunity here looking at how do you understand and 
interconnect these various data sources which exist today, 
particularly at the administrative data sources, and 
engineering that construct, how you would use it for the 2020, 
I think, is an important imperative for the census and for 
broader government as a whole.
    Second, there has been this theme about how do you 
capitalize on technology? I think from a government 
perspective, which is not unique to Census, is how does an 
organization or a department become a smart buyer of 
technology. That is where the real challenge is. So getting a 
handle on what does that mean? It is getting smart.
    I mentioned earlier it is about getting a handle on what is 
your technical baseline, what are you trying to do to find 
those requirements to some degree so industry can best respond 
and even partner with you as you are thinking through 
conceptually what that technical baseline will look like going 
into the future?
    And then finally a third, which is the organization. In my 
experience with Census through the 2010, I have to agree with 
the previous speakers that there has been a sea state change in 
terms of an organization being very transparent. So when I 
alluded earlier of empowering organizations to have more 
accountability of the life cycle and making decisions, I think 
that goes hand in hand with greater transparency.
    If Congress decides, OK, I am going to give you guys more 
leeway, that means you have got to provide greater 
transparency. So there is an opportunity here. It is a two-way 
street between accountability and transparency. When you talk 
about organizations and government which I have seen are 
exceptional, and I see Census moving down that way in terms of 
leadership, particularly with the current leadership in place, 
I think what I see that they fundamentally have instilled is 
what I call a learning organization.
    I see a substantial shift in Census being a learning 
organization. And when organizations are in that State of 
learning, I think that is where you find the greatest 
innovation. I think you find people having a better sense of 
purpose. They are willing to take risks. They are willing to be 
uncomfortable, because you learn the most when you are most 
uncomfortable, and I think the leadership is pushing the 
organization to be uncomfortable, to think outside the box, and 
I think that is very positive, particularly for the challenges 
for the 2020.
    Senator Carper. Thank you. Velvet handed me another 
question. I am not going to ask for you to respond right now, 
but I just want to--Dr. Reamer, you talked a little bit about 
partnerships, I think in one of your earlier comments, but I 
will ask you all to respond for the record to this question and 
that is, can you describe the value of the partnership program 
and assess its overall effectiveness in ensuring full 
participation of hard-to-count groups? And what should Census 
do in 2020 to keep stakeholders better informed? And we will 
provide that in writing.
    Appreciate you to kind of amplify, build on what you have 
said earlier. And for other witnesses, if you would like to 
respond, that would be great. It would help me. In terms of how 
long our colleagues have to submit any questions that they 
have, 2 weeks. All right. Two weeks. I can think of a lot by 
then.
    We would just ask as you receive those questions from Dr. 
Coburn and myself or Senator Brown, or from others, that you 
respond in a timely and prompt way. It has been a most 
informative hearing. Not one where people are actually beating 
down the doors to get in, Members or the press, but I think it 
is just really a very important, very important subject and an 
important point, timely point to ask, have this kind of 
hearing, and we appreciate you making it a real beneficial one.
    I think it was largely Dr. Coburn's idea, but one I readily 
joined in, in saying let us do that. So with that having been 
said, we thank you all for your testimony. Sitting here, I can 
envision like the clouds darkening outside. If it does not rain 
tonight, we have missed a great opportunity here. So hopefully 
it will because we could sure use all that. Obviously it is not 
just God playing a trick on us in getting our hopes up and then 
saying, Ah. Like in a Charlie Brown cartoon where Lucy pulls 
the football away. So hopefully that is not going to be the 
case here.
    Thank you all. It was a great hearing. We appreciate very 
much your participation. With that, we are 
adjourned.[Whereupon, at 5:11 p.m., the hearing was adjourned.]

                            A P P E N D I X

                              ----------                              



[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]



                                 
