[Senate Hearing 109-922]
[From the U.S. Government Publishing Office]



                                                       S. Hrg. 109-922
 
                 CENSUS 2010, OFF-LINE AND OFF-BUDGET: 
                   THE HIGH COST OF LOW-TECH COUNTING 
=======================================================================



                                HEARING

                               before the

                FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT
                     INFORMATION, AND INTERNATIONAL
                         SECURITY SUBCOMMITTEE

                                 of the

                              COMMITTEE ON
                         HOMELAND SECURITY AND
                          GOVERNMENTAL AFFAIRS
                          UNITED STATES SENATE


                       ONE HUNDRED NINTH CONGRESS

                             SECOND SESSION

                               __________

                              JUNE 6, 2006

                               __________

        Available via http://www.access.gpo.gov/congress/senate

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




                      U.S. GOVERNMENT PRINTING OFFICE
29-501 PDF                    WASHINGTON  :  2007
---------------------------------------------------------------------
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)512-2250 Mail: Stop SSOP,
Washington, DC 20402-0001 



        COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS

                   SUSAN M. COLLINS, Maine, Chairman
TED STEVENS, Alaska                  JOSEPH I. LIEBERMAN, Connecticut
GEORGE V. VOINOVICH, Ohio            CARL LEVIN, Michigan
NORM COLEMAN, Minnesota              DANIEL K. AKAKA, Hawaii
TOM COBURN, Oklahoma                 THOMAS R. CARPER, Delaware
LINCOLN D. CHAFEE, Rhode Island      MARK DAYTON, Minnesota
ROBERT F. BENNETT, Utah              FRANK LAUTENBERG, New Jersey
PETE V. DOMENICI, New Mexico         MARK PRYOR, Arkansas
JOHN W. WARNER, Virginia

           Michael D. Bopp, Staff Director and Chief Counsel
             Michael L. Alexander, Minority Staff Director
                  Trina Driessnack Tyrer, Chief Clerk


FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT INFORMATION, AND INTERNATIONAL 
                         SECURITY SUBCOMMITTEE

                     TOM COBURN, Oklahoma, Chairman
TED STEVENS, Alaska                  THOMAS CARPER, Delaware
GEORGE V. VOINOVICH, Ohio            CARL LEVIN, Michigan
LINCOLN D. CHAFEE, Rhode Island      DANIEL K. AKAKA, Hawaii
ROBERT F. BENNETT, Utah              MARK DAYTON, Minnesota
PETE V. DOMENICI, New Mexico         FRANK LAUTENBERG, New Jersey
JOHN W. WARNER, Virginia             MARK PRYOR, Arkansas

                      Katy French, Staff Director
                 Sheila Murphy, Minority Staff Director
            John Kilvington, Minority Deputy Staff Director
                       Liz Scranton, Chief Clerk


                            C O N T E N T S

                                 ------                                
Opening statements:
                                                                   Page
    Senator Coburn...............................................     1
    Senator Carper...............................................    14

                               WITNESSES
                         Tuesday, June 6, 2006

Hon. Louis Kincannon, Director, U.S. Census Bureau...............     7
Brenda S. Farrell, Acting Director, Strategic Issues, U.S. 
  Government Accountability Office...............................     9

                     Alphabetical List of Witnesses

Farrell, Brenda S.:
    Testimony....................................................     9
    Prepared statement...........................................    72
Kincannon, Hon. Louis:
    Testimony....................................................     7
    Prepared statement with attachments..........................    34

                                APPENDIX

Charts submitted by Senator Coburn:
    ``After Inflation Cost of the Census: 1970-2010''............    31
    ``Cost of the Census: 1940-2010''............................    32
    ``Percentage of U.S. Adults Online''.........................    33
Questions and responses for the Record from Senators Coburn and 
  Carper for:
    Mr. Kincannon................................................    94
    Ms. Farrell..................................................   101


    CENSUS 2010, OFF-LINE AND OFF-BUDGET: THE HIGH COST OF LOW-TECH 
                                COUNTING

                              ----------                              


                         TUESDAY, JUNE 6, 2006

                                     U.S. Senate,  
          Subcommittee on Federal Financial Management,    
       Government Information, and International Security  
                      of the Committee on Homeland Security and    
                                            Governmental Affairs,  
                                                    Washington, DC.
    The Subcommittee met, pursuant to notice, at 2:32 p.m., in 
room SD-342, Dirksen Senate Office Building, Hon. Tom Coburn, 
Chairman of the Subcommittee, presiding.
    Present: Senators Coburn and Carper.
    Senator Coburn. Good afternoon. The Federal Financial 
Management Subcommittee of the Homeland Security and 
Governmental Affairs Committee will come to order.
    I want to welcome each of our guests.
    I have an opening statement and Senator Carper will be 
arriving shortly. We will go on with our hearing and, dependent 
on when he arrives, we will allow him a chance to give an 
opening statement.
    I want to thank you for the preparation for this hearing 
and working with our staffs. It has been great to work with 
you.
    Usually, when we think about the census, we think about 
statistics. The Census Bureau has become the largest 
statistical agency in the country, if not the world. But behind 
its data collection is a steadily increasing price tag for the 
decennial census which, until recently, has managed to stay 
under the radar of Congress. As we approach the 2010 census, 
though, it is becoming increasingly apparent to me that costs 
are spiraling upward at a startling rate.
    The 2010 census is projected at the present time to cost 
nearly $12 billion. That is $5 billion more, an 80 percent 
increase, over the 2000 census. And that is the estimate which 
we are going to hear about today, the numbers behind that.
    The 2000 census, in turn, cost $4 billion more than the 
1990 census, at the time a more than 100 percent increase. This 
is all part of a disturbing trend in recent decades which 
witnessed dramatic cost increases from one census to the next.
    Adding to our cost problem is a culture problem. The census 
seems to be operating under an early 20th-Century mentality 
when pen and paper were the only tools available. The Internet 
is now available. For the next census in 2010, the Bureau has 
decided not to offer an online option, choosing rather to stick 
with the system that is in place as of today.
    In an age when people do everything online, from shopping 
to banking to filing their tax returns, a record 70 million tax 
returns this last year were filed online, the Census Bureau is 
lagging behind, needlessly adding to its already high cost and 
also adding to its time delay.
    I think this is also a mission problem. Census is tasked 
with counting the population and it needs the help of all 
citizens to pull it off. Participation in the census would be 
easier to obtain for more people with fewer census personnel if 
an online option were available.
    The purpose of this hearing today is to examine what is 
behind the skyrocketing cost at the census and what can be done 
about it. I hope to get answers to questions as how well has 
the census been planning for the 2010 count? What assurances 
can we have that the cost overruns in the billions will not 
take place next time like they have so many times before? And 
why was an online option suddenly rejected? And what will it 
take to get that back into the plans for 2010?
    The best cost estimate being provided by the Census Bureau 
for 2010 is $11.3 billion. Unfortunately though, if history is 
any guide, that estimate will bear little resemblance to 
reality in 2010. As you can see from the chart,\1\ from 1940 to 
2010, the 1970 to 2000 cost for the census increased sharply. 
Some of that is related to Congress' requests for increased 
data.
---------------------------------------------------------------------------
    \1\ The chart referred to appears in the Appendix on page 31.
---------------------------------------------------------------------------
    Costs jumped most significantly between 1970 and 1980. 
Since 1980 the cost of the census has doubled every decade. In 
2010 it is shaping up to be the same story once again with a 
cost increase over the 2000 census of at least $5 billion and 
most probably $7 billion. No one seems to be willing to apply 
the brakes.
    Some, including the Census Bureau itself, have blamed 
inflation and population growth, but what we need to do is look 
at the facts. After inflation cost of the Census, if you look 
at the next chart from 1970 to 2010,\2\ in the decade between 
1990 and 2000, when inflation was amazingly low, 27 percent, 
the cost of the census increased 154 percent. Between 2000 and 
2010 with 10-year inflation numbers again expected to be low, 
the cost of the census is expected to be increased between 70 
and 90 percent.
---------------------------------------------------------------------------
    \2\ The chart referred to appears in the Appendix on page 32.
---------------------------------------------------------------------------
    The situation is the same when the population growth 
numbers are compared with census costs. In 1990, the census 
cost $10 a person and in 2000 it cost $23 per person. But in 
2010, the census will cost a staggering, at a minimum, $36 per 
man, woman, and child in this country.
    That is much more than it cost to file your taxes 
electronically with the IRS, and yet the Constitution requires 
us to count the heads.
    The bottom line is that the census costs are shooting 
upward at an unacceptable rate. Many of the problems are with 
Congress and what we have asked for. But some of the excuses 
that we have been given are without merit.
    What then are the real causes of the large cost increase 
between 2000 and 2010? The Census Bureau, through their 
testimony, attributes it to factors such as increased 
difficulty of finding non-English speakers and people living in 
non-traditional housing. The Bureau also claims that as the 
population grows, counters will have to knock on more doors to 
make up for people that do not mail back their forms and that 
costs money. That is true. But when all these factors are 
accounted for, it still remains unclear how we get to a number 
approaching $12 billion to $13 billion.
    The Government Accountability Office, Congress' watchdog 
agency, has analyzed the Bureau's cost projection and was 
equally mystified. As a result, they recommended more than 2 
years ago that the Census Bureau compile all its planning 
information into one master document to help Congress 
understand its long-term budget. Census agreed to do so but two 
appropriation cycles have now come and gone, and still there is 
no document.
    How is Congress supposed to fulfill our oversight duty 
effectively without understanding this basic information? Or is 
that the point? I assure you, we will not let this issue drop. 
There will be a planning document that itemizes the cost 
projections down to the dollar.
    Knowing projected costs is only the first step. You will 
see, on this chart,\1\ transparency is only the first step to 
accountability. Frankly, all of these issues should have been 
worked out after the 2000 budget busting debacle. As late as 
1998, the Bureau projected cost of $4 billion to $5 billion. 
When all was said and done, the final cost was more than $6.5 
billion, a cost overrun over estimates of greater than 30 
percent. If the 2010 census faces a cost overruns similar to 
that in 2000, it will put the final price tag at $15 billion. 
It is not simply a matter of possibility, it is an 
inevitability unless something is done right now to reassess 
the cost structure associated with the census.
---------------------------------------------------------------------------
    \1\ The chart referred to appears in the Appendix on page 32.
---------------------------------------------------------------------------
    One of the most obvious solutions to long-term cost 
containment is for the Agency to join the rest of the world in 
cyberspace and offer the census online. As you can see from 
this chart,\2\ the percentage of American adults online now 
exceeds 72 percent. It is estimated that will be above 85 
percent in the year 2010. An online census would allow the 
Census Bureau to virtually eliminate its paper intensive 
systems, to cut back dramatically on the need for house calls, 
and to allow faster data integration.
---------------------------------------------------------------------------
    \2\ The chart referred to appears in the Appendix on page 33.
---------------------------------------------------------------------------
    In just the last 5 years, the Federal Government has made 
extraordinary strides with its e-government initiatives to the 
point that every citizen can now file their taxes online. 
Certainly, if citizens can file their taxes online, they can be 
counted on line. And so it is puzzling to me why the census has 
taken the online option off the table for 2010.
    To say an online option is not practical or cannot be done 
simply defies the plain fact that 73 percent of all Americans 
are already online and the Federal Government e-government 
sites are the No. 1 place that they visit.
    Canada just last month showed us that it can be done and 
conducted its national census and offered it online to all of 
its citizens. This is not just something that we can do. It is 
something that must be done.
    In the medical world, we have a word for it when the number 
of cells in the body increases at a rate faster than the 
underlying conditions that usually govern cell division would 
predict. It is called cancer. The underlying factors governing 
the cost of counting Americans do not justify the staggering 
cost increases in the census. Americans get it. They get that 
it is easy to do things online. They get that it is not 
complicated for a Federal agency to know who they are and some 
basic information about them.
    The government, for the most part, already knows 
practically everything there is to know about us, from what is 
in our bank accounts to our health status in retirement. I 
simply cannot sell the cost increases that I am seeing to my 
constituents in Oklahoma, and I will not defend them to the 
constituents in this country. Americans are not buying it.
    There is still time to make mid-course adjustments for 
2010. Our children and grandchildren cannot afford for us to 
punt these problems until 2020 or 2030. My hope is that this 
hearing will help get us back on the right track right away.
    [Prepared statement of Chairman Coburn follows:]
                  OPENING STATEMENT OF CHAIRMAN COBURN
    Usually, when we think about the Census, we think statistics. The 
Census Bureau has become the largest statistical agency in the country, 
if not the world. But, behind its data collection is a steadily 
increasing price tag for the decennial census, which until recently has 
managed to stay under the radar of Congress. As we approach the 2010 
Census, though, it is becoming increasingly apparent that costs are 
spiraling upward at a startling rate.
    The 2010 Census is projected to cost nearly $12 billion--that's $5 
billion more--a startling 80 percent increase--over the 2000 Census. 
The 2000 Census in turn cost $4 billion more than the 1990 Census--at 
the time, a more than 100 percent increase. This is all part of a 
disturbing trend in recent decades, which witnessed dramatic cost 
increases from one census to the next.
    Adding to our cost problem is a culture problem. The Census seems 
to be operating under an early 20th Century mentality, when pen and 
paper were the only tools available. The Internet is now available. For 
the next census in 2010, the Bureau has decided not to offer an online 
option, choosing rather to stick with a paper system. In an age when 
people do everything online from shopping to banking to filing their 
taxes, the Census Bureau is lagging behind, needlessly adding to its 
already high costs.
    This is also a mission problem. Census is tasked with counting the 
population and it needs the help of all citizens to pull it off. 
Participation in the census would be easier to obtain from more people 
with fewer Census personnel if an online option were available.
    The purpose of this hearing today is to examine what is behind the 
skyrocketing costs at the census, and what can be done about it. I hope 
to get answers to questions such as: How well has the Census been 
planning for the 2010 count? What assurances can we have that cost 
overruns in the billions won't take place next time like they have so 
many times before? Why was an online option suddenly rejected and what 
will it take to get it back into the plans for 2010?
    The best cost estimate being provided by the Census Bureau for the 
2010 Census is $11.3 billion. Unfortunately, though, if history is any 
guide, that estimate will bear little resemblance to the reality in 
2010. As you can see from this chart, between 1970-2000 costs for the 
census increased sharply. Costs jumped most significantly between 1970 
and 1980, going from $250 million to well over $1 billion in 1980. 
Since 1980, the cost of census has doubled every decade. In 2010, it is 
shaping up to be the same story once again with a cost increase over 
the 2000 Census of at least $5 billion, and possibly more. No one seems 
willing or able to apply the brakes.
    Some, including the Census Bureau itself, have blamed inflation or 
population growth, but let's take a look at the facts. [refer to 
posters] As you can see from this poster, even after inflation is 
accounted for, costs still climb from one census to the next. In the 
decade between 1990 and 2000, when inflation was an amazingly low 27 
percent, the cost of the census increased by 154 percent. Between 2000 
and 2010, with ten-year inflation numbers again expected to be low, the 
cost of the census is expected to increase by between 70-90 percent.
    The situation is the same when population growth numbers are 
compared with census costs. In 1990, the census cost $10 a person, and 
in 2000 the Census cost $23 per person; but in 2010, the census will 
cost a staggering $36 for every man, woman and child living in this 
country. That's much more than it costs to file your taxes 
electronically with the IRS and yet all the Constitution requires us to 
do is count heads here. The bottom line is that census costs are 
shooting upward at an unacceptable rate, and the excuses given are 
without merit.
    What, then, are the real causes of the large cost increase between 
2000 and 2010? The Census Bureau attributes it to factors such as the 
increased difficulty of finding non-English speakers and people living 
in non-traditional housing. The Bureau also claims that as the 
population grows, counters will have to knock on more doors to make up 
for people that don't mail back their forms--and that costs money. But, 
when all of these factors are accounted for, it still remains unclear 
how we get to a number approaching $12 billion.
    GAO--Congress' watchdog agency--has analyzed the Bureau's cost 
projections and was equally mystified. As a result, they recommended--
more than two years ago--that the Bureau compile its planning 
information into one master document to help Congress understand its 
long-term budget. Census agreed to do so, but two appropriations cycles 
have now come and gone and there still is no document. How are we 
supposed to fulfill our oversight duty effectively without 
understanding this basic information? Or is that the point? I assure 
you, we will not let this drop. There will be a planning document that 
itemizes cost projections down to the dollar.
    Knowing projected costs is only the first step--you'll see on this 
chart--transparency is only the first step to accountability. Then we 
have to figure out how to contain those costs. Frankly, all these 
issues should have been worked out after the 2000 budget-busting 
debacle. As late as 1998, the Bureau projected costs of $4-5 billion. 
When all was said and done, the final cost was more than $6.5 billion--
a cost overrun of more than 30 percent. If the 2010 Census faces a cost 
overrun similar to that in 2000, it will put the final price tag at $15 
billion. This is not simply a matter of possibility, it is an 
inevitability unless something is done right now to curb the 
skyrocketing costs.
    One of the most obvious solutions to long-term cost containment is 
for the agency to join the rest of the world in cyberspace and offer 
the census online. An online census would allow the Census Bureau to 
virtually eliminate its paper-intensive systems, to cut back 
dramatically on the need for house calls and to allow faster data 
integration. In just the last five years, the federal government has 
made extraordinary strides with its e-Government initiatives to the 
point that every citizen can now file their taxes online--certainly if 
citizens can file their taxes online, they can be counted online. And 
so it is puzzling to me why Census has taken the online option off the 
table for 2010.
    I am deeply concerned that the Census is mired in a bureaucratic, 
``pen and paper'' mentality that refuses to change the way things have 
always been done. To say an online option is not practical or cannot be 
done simply defies the plain fact that 73 percent of all American 
adults are online already. Canada just last month showed us that it can 
be done and conducted its national census and offered it online to all 
of its citizens. This is not just something that we can do, it 
something that must be done. I assure you that this subcommittee will 
not drop this issue. The 2010 Census will be online.
    In the medical world, we have a word for when the number of cells 
in the body increases at a rate faster than the underlying conditions 
that usually govern cell division would predict--cancer. The underlying 
factors governing the cost of counting Americans do not justify the 
staggering cost increases at the Census. Americans get this. They get 
that it's easy to do things online. They get that it's not that 
complicated for a Federal agency to know who they are, and some basic 
information about them. The government for the most part, already knows 
practically everything there is to know about us--from what's in our 
bank accounts to our health status in retirement. I simply can't sell 
these cost increases to my constituents back in Oklahoma. Americans 
aren't buying it.
    There's still time to make mid-course adjustments for 2010. Our 
children and grandchildren can't afford for us to punt these problems 
to the 2020 or 2030 count. I hope that this hearing will help us get 
back on track right away. I want to thank our witnesses for being here 
today and for their time and preparation.

    Senator Coburn. And I want to thank you again for your 
efforts to be here and our thank you for our witnesses to be 
here.
    Let me introduce to you, if I can, our panel of witnesses. 
First is Brenda Farrell. She is Acting Director of Strategic 
Issues, U.S. Government Accountability Office.
    In November 2005, Ms. Farrell was appointed Acting Director 
for Strategic Issues, where she is responsible for overseeing 
three major bodies of work related to census, strategic human 
capital, and government regulation issues.
    Prior to joining Strategic Issues teams, Ms. Farrell was 
Assistant Director for Defense Capabilities in Management and 
led military personnel engagements encompassing bodies of work 
in military pay and benefits, Reserve and National Guard 
mobilization issues, and military officer requirements and 
career development.
    She began her career at GAO in 1981 and has served in a 
number of areas. In 2001, she was selected to enter the 
National Defense University Industrial College of the Armed 
Forces and earned a master's degree in national resources 
strategy.
    She has also completed other specialized training in 
subject matter expertise such as defense manpower and force 
management. She completed the Leadership Development Program at 
Eckerd College in 2004. And in 2005, she completed the Senior 
Executive Fellow Program at Harvard University.
    Her numerous awards include Results Through Teamwork Awards 
in 2004 and 2003, an award for high quality products and client 
relations in 2003, and a GAO honor award for sustained 
extraordinary performance leading multiple highly complex 
defensive reviews in 2002.
    Charles Louis Kincannon is the Director of the U.S. Census 
Bureau. He was appointed by President Bush and was unanimous 
confirmed by the Senate on March 13, 2002.
    He began his career as a statistician at the U.S. Census 
Bureau in 1963 after graduating from the University of Texas at 
Austin. Congratulations on that wonderful national 
championship.
    He held positions of leadership at the Census Bureau and 
also with the Office of Management and Budget. He served as 
Deputy Director of the Census Bureau during the 1980s and as 
Acting Director during the crucial final phase of preparation 
for the 1990 census.
    Throughout his career with the Federal Government, Mr. 
Kincannon sought to strengthen relationships between 
statistical agencies as well as data users in order to produce 
timely, relevant data that informs public policy and 
decisionmaking.
    In October 1992, Mr. Kincannon was appointed as the first 
Chief Statistician in the Organization for Economic Cooperation 
and Development, (OECD), in Paris to coordinate the 
organization's statistical programs, as well as advise the OECD 
Secretary on general statistical policy. During that time, he 
encouraged cooperation and understanding amongst statistical 
agencies and underscoring the large relationships between the 
nations.
    I again want to thank each of you for your cooperation. Mr. 
Kincannon, we are going to recognize you first and give you an 
opportunity to speak. Take the time that you need. And then we 
will recognize Ms. Farrell. You are recognized.

  TESTIMONY OF HON. LOUIS KINCANNON,\1\ DIRECTOR, U.S. CENSUS 
                             BUREAU

    Mr. Kincannon. Thank you, Mr. Chairman. Let me move this a 
little closer.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Kincannon with attachments 
appears in the Appendix on page 34.
---------------------------------------------------------------------------
    On behalf of the Census Bureau, I want to thank the 
Chairman and presently Senator Carper for the opportunity to 
update the Senate on the re-engineered 2010 census program. The 
decennial census program is the Bureau's largest activity and 
its highest budget priority. In fact, it is one of this 
Nation's largest peacetime mobilizations and is mandated by the 
Constitution.
    In the past, the census provided comprehensive detailed 
information once every decade. Yet there is an increasing need 
for such data more frequently at the local level. The American 
Community Survey, one of the components of the re-engineered 
2010 census program, will address this need.
    The American Community Survey (ACS), replaces the long form 
of the census, a crucial step in realizing a short form only 
census. In the past, we collected long form data as part of the 
decennial census. As such, it was costly and it complicated our 
effort to conduct a basic enumeration.
    The American Community Survey collects information on 
education, income, and other social and economic 
characteristics. Every question on the ACS is mandated by 
Federal law or fulfills Federal requirements.
    The ACS will provide timely, accurate information for every 
county, city, and neighborhood each year, not just once a 
decade. These data will help city and community leaders in 
every State and allow the Census Bureau to focus its efforts in 
2010 on the core constitutional count used as the basis for 
apportionment and redistricting.
    The success of the 2010 re-engineered census program will 
also depend on the MAF/TIGER or geographic tools enhancement 
program, an extensive nationwide operation to modernize and 
consolidate the census address list and map. This is a 
multifaceted effort taking advantage of well-established 
technologies, such as GPS capabilities, to improve outdated 
error prone map systems currently in place.
    Much of this work is being done through a major contract 
with the Harris Corporation, estimated at $200 million in cost 
at the time of its award in June 2002. This activity is within 
budget and on schedule for completion in 2008. This geographic 
improvement program is important because ensuring the accuracy 
of the location of each address is the guarantee that political 
representation and resources can be distributed fairly to 
States, cities, towns, census tracts, and blocks as they are 
demanded.
    Our overriding goal for the 2010 census is to improve the 
coverage and accuracy of the census and to contain costs. In 
response to numerous GAO recommendations, we have developed a 
rigorous planning and testing program that includes many long 
sought census improvements such as bilingual questionnaires, a 
second mailing of the questionnaire and targeted census 
coverage improvement programs.
    Another significant improvement is the expanded use of 
technology. Our efforts have centered on two major systems, the 
2010 Decennial Response Integration System (DRIS), and the 
Field Data Collection Automation System, or FDCA as we rather 
uneuphoniously refer to it.
    Both of these are IT contracts together totalling over $1 
billion. The purpose of the DRIS contract, which was awarded 
last year to Lockheed Martin Corporation, is to ensure the 
accurate and protected collection and storage of American's 
data, whether by paper form, handheld computer, or telephone.
    The FDCA contract was awarded this spring to the Harris 
Company. The purpose of FDCA is to capture directly the 
information collected by mobile computer devices during the 
personal interviews and non-response follow-up. This eliminates 
the need for paper forms, address lists, and maps for the major 
field data collection operations. The use of this technology is 
a revolutionary improvement in the way we conduct the largest 
and most expensive activity of the decennial census.
    All of this underscores the importance of Congressional 
support for all aspects of the 2010 decennial census. Thousands 
of individual operations and procedures must be successfully 
implemented in less than 4 years to ensure the success of the 
2010 census.
    The President's 2007 budget request for the Census Bureau 
is over $800 million. $512 million of that is for the decennial 
programs. In the course of the decade, we expect the re-
engineered census will cost more than $11 billion, as the 
Chairman said.
    To understand the cost, consider the scope of the task. It 
is our responsibility to count every person in every community 
on every street and in every household. For the Census 2000, we 
sent questionnaires to more than 117 million households, 80 
million of those households responded by mail. For the rest, we 
sent census takers to collect the census information. We opened 
520 local census offices and hired more than 860,000 temporary 
workers.
    For 2010, we are projecting there will be more than 310 
million persons living in America and that we will have to 
count them in more than 130 million households.
    Our increasingly diverse population is more difficult to 
count. As we plan and test new data collection efforts, we try 
to estimate the effect they will have on the overall response 
rate, since the high non-response follow up is truly the cost 
driver for the census.
    We have successfully tested and plan to implement bilingual 
questionnaires in selected communities, guided by the results 
from the ACS. A second mailing will be sent to non-responding 
households and automated field data collection is a device 
that, along with these others, are steps that will reduce cost 
and improve quality.
    We have also considered other data collection and methods, 
including Internet data collection. Based on our research, 
testing, experience, and the knowledge of experience in other 
countries like Canada and Australia, Internet data collection 
would not significantly improve the overall response rate to 
the census or reduce field data collection costs.
    In 2003 and 2005, census tests offered an Internet response 
option. And in both cases the Internet response was low and did 
not increase the overall response rate. It merely diverted some 
small percentage, about 7 percent, from the paper medium to the 
Internet, not enough to substantially change our paper 
collection or field data collection costs.
    We are also concerned that utilizing the Internet could 
jeopardize other planned improvements that we know will save 
money. At this point in the decade, efforts to develop an 
Internet response would divert attention and resources from 
these tested and planned improvements that we know will 
increase the overall response rate by several percentage points 
and save money.
    A successful census is more than a technical achievement. 
It is the creation of a national resource that empowers 
decisionmaking. I hope, Mr. Chairman, you will agree it is a 
success worth supporting.
    I thank you for this opportunity to provide an update to 
the census and look forward to your questions.
    Senator Coburn. Ms. Farrell.

 TESTIMONY OF BRENDA S. FARRELL,\1\ ACTING DIRECTOR, STRATEGIC 
         ISSUES, U.S. GOVERNMENT ACCOUNTABILITY OFFICE

    Ms. Farrell. Thank you, Mr. Chairman.
---------------------------------------------------------------------------
    \1\ The prepared statement of Ms. Farrell appears in the Appendix 
on page 72.
---------------------------------------------------------------------------
    Senator Coburn. It is hard to remember.
    Ms. Farrell. I know, and I was warned beforehand, too.
    Mr. Chairman, thank you for the opportunity to be here 
today to discuss the mushrooming costs of the decennial census, 
now estimated to be over $11 billion, as well as the actions 
that the Census Bureau is taking to contain those costs.
    Let me briefly summarize my written statement that is based 
on findings from our issued reports, as well as preliminary 
results from ongoing work that we plan to issue within the next 
few weeks on the Bureau's efforts to build a complete and 
accurate address list, the foundation for a successful census.
    A cost effective decennial census is a monumental 
management challenge. It is long-term. The 2010 Census 
protected life cycle costs spans 13 fiscal years.
    It is large-scale. For example, if recruitment goals are 
similar to the 2000 Census, 2.4 million applicants could be 
recruited to carry out census operations.
    It is costly. As already noted, according to the Bureau, 
the next census will cost over $11 billion.
    It is a high risk, in that the Census Bureau has one 
opportunity to get it right on April 1, 2010.
    Further, we are closely monitoring the 2010 Census to 
determine if we should put it on GAO's high-risk list.
    The sheer size of the census means that small problems can 
magnify quickly and bit problems could be overwhelming. For 
example, 60 seconds might seem like an inconsequential amount 
of time. But in 2000, if enumerators had spent just one minute 
more at each household during non-response follow-up, it could 
have added almost $10 million to the cost of the census.
    My statement today is presented in three parts. The first 
addresses the extent to which the Bureau has developed timely 
and detailed cost data for effective oversight and cost 
control. Despite a history of cost increases, the Bureau's most 
recent life cycle cost estimate does not reflect the most 
current information from testing and evaluation, nor provide 
complete information on how changing assumptions may affect 
costs.
    Given the cost of the census in an era of serious national 
fiscal challenges, it is crucial for the Bureau to provide 
Congress with more complete information such as sensitivity 
analyses about the likelihood--high, medium, or low--that 
certain assumptions would drive costs.
    For example, for the 2000 Census, the Bureau's supplemental 
funding request for $1.7 billion in fiscal year 2000 primarily 
involved changes in assumptions related to increased workload, 
reduced employee productivity and increased advertising.
    The second part of my testimony addresses the progress the 
Bureau has made to reduce non-response follow-up costs. Since 
2000, the Bureau has re-engineered the decennial census and has 
begun new initiatives to reduce non-response follow-up costs
    These initiatives include: One, using only a short form 
census questionnaire. Two, automating field operations. Three, 
using a targeted second mailing to households that fail to 
respond to the initial census questionnaire instead of sending 
an enumerator to visit houses that have not responded.
    These initiatives could reduce the workload and cost of 
non-response follow-up. While these initiatives show promise, 
the Bureau will be to address technological challenges with the 
handheld mobile computing devices that will be used to collect 
the data for non-response follow-up.
    Third and finally, Mr. Chairman, not withstanding the 
significant progress the Bureau has made to address lessons 
learned from the 2000 Census, I wish to note several challenges 
of, if not properly managed, could increase the cost of the 
census. These challenges include overseeing contractors 
responsible for conducting key census-taking operations 
totaling almost $2 billion in contracts; successfully updating 
address and map files; and assessing the resources that will be 
needed to update the address and maps for areas affected by 
Hurricanes Katrina and Rita.
    We have made recommendations in our reports for each of 
these three areas and the Bureau has said that it is taking 
action on many of them. We will continue to assist Congress in 
monitoring the Bureau's progress.
    Mr. Chairman, that concludes my opening statement and I 
will be happy to take questions at this time.
    Senator Coburn. Thank you.
    Let me go to Mr. Kincannon. And I want you to feel free to 
take time, if you heard something that you do not think is 
right, Mr. Kincannon, to address it. If you think there are 
assumptions that were made in her testimony or something I have 
said, please feel free to address those issues as we go 
through. This is about to get the information out so that we 
all know.
    Right now, we are talking about the 2010 Census costing $5 
billion more than the 2000. What are the two or three biggest 
cost drivers in that that would account, other than inflation 
which is going to be about 25 percent. What are the two or 
three biggest cost drivers that are accounting for why this 
thing would increase by $5 billion?
    Mr. Kincannon. Before we go to the second part of your 
question, our figures indicate that so-called Federal 
inflation, that is the inflation rate used by OMB to estimate 
out-year budgets, accounts for about two-thirds of the total 
cost increase between censuses. So it is not a quarter but two-
thirds, unless we have a different set of figures in mind.
    Senator Coburn. The last census cost what?
    Mr. Kincannon. The last census cost $7.6 billion in 
constant 2010 dollars.
    Senator Coburn. No, what did it cost in dollars then? You 
cannot use both sides of the inflation number. If you are going 
to give me inflation-adjusted, it was $6 billion, $6.4 billion 
or $6.5 billion.
    Mr. Kincannon. $6.4 billion, if you add together the 
dollars spent at each year in the 13-year cycle.
    Senator Coburn. We are talking 10-year periods. We are 
talking about the same thing. You are talking about, at a 
minimum $11 billion, and probably more likely much greater than 
that.
    So we are talking $5 billion.
    The American public, if we are going to use cost-adjusted, 
then we need to use cost-adjusted all the way. And so we are 
talking real dollars.
    The fact is in 2000 dollars, it is a 50 percent increase in 
2000 dollars. If you are talking 2000 dollars.
    Mr. Kincannon. If you are talking nominal dollars in 2000 
and nominal dollars in 2010, then yes, it would be $5 billion. 
It is $6.4 billion in 2000.
    Senator Coburn. We have had an inflation rate of under 3 
percent each year. So at the most, we are going to have 30 
percent, or 1.3 times, so you are going to have 33 or 35 
percent. Why is it going to cost $2.5 billion more?
    Mr. Kincannon. It costs more because of increase in 
population, increase in the number of housing units, an 
increase in the number of people per housing unit, which means 
that a housing unit is really the unit of work in the census. 
So those things go together.
    The increased difficulty in getting people to respond to 
Federal surveys or inquiries of any kind.
    Senator Coburn. So we know that as a fact, that there is a 
harder factor to get anybody to respond today?
    Mr. Kincannon. Yes, there are plenty of indicators that it 
is harder to get people to respond to surveys.
    Senator Coburn. And there is no economies of scale? If we 
have 600 million people, we should keep rising, in terms of the 
cost per person to count them?
    Mr. Kincannon. It will more than rise, in terms of the cost 
to count each person, if there are smaller housing units, 
smaller families living in houses or more elderly living in 
housing units alone. The smaller the housing unit, the less the 
productivity of getting data from each housing unit.
    Senator Coburn. The cost per person in 1970 was $1.22. At 
best, we are talking $36.57 per person, and probably more 
likely over $40 per person. In 2000 the cost per person was 
$23.45, which was 130 percent more than in 1990.
    I do not think the American people are going to buy the 
fact that if we doubled the population we would get no 
economies of scale out of the census organization in terms of 
the numbers responding. If you are going to mail out a survey, 
what was your percentage in the 2000 Census, and terms of 
response to the mailing?
    Mr. Kincannon. It was 67 percent, I believe, housing units 
mailed back returns.
    Senator Coburn. So you would not assume that you would get 
60-some percent out of 600 million, as you would out of 300 
million?
    Mr. Kincannon. I think we will get a higher percentage out 
of the mail response in 2010, because we will have only a short 
form census. And I think people will be more cooperative.
    Senator Coburn. So there is cost savings associated with 
that?
    Mr. Kincannon. It is not relevant to speak of the cost of 
counting a person because we do not count the person one by 
one. We count in housing units. So you go to the door with a 
questionnaire by mail, or in-person if necessary. So that is 
the relevant unit of cost.
    Senator Coburn. So if that is the relevant unit of cost, it 
costs $56 to do that in 2000 and it is going to cost $88 in 
2010, based on your best estimates right now.
    And you are going to be using the short form on everybody. 
So explain to me why that is going to shoot up 50 percent, more 
than 50 percent, on the cost per household, based on your own 
estimates of the numbers that you gave the Subcommittee?
    Mr. Kincannon. I thought the numbers that we gave the 
Subcommittee, put in constant dollars, showed an increase of 35 
percent.
    Senator Coburn. Let us just talk about dollars. You gave 
the Subcommittee $56 per household to $88 per household. So 
that is from $32 to $56. That is a significant increase. I will 
not quibble with the numbers.
    The question is you are going to the small form, the short 
form. You are going to have more numbers that are going to be 
returned because it is going to be a short form. How do you 
explain to the American people that the cost is going up $32 
per household over 10 years on a short form now, when a third 
or 10 percent of them used to be the long form. How do we 
explain them? How do we justify that?
    Mr. Kincannon. Well, the cost per housing unit is a 
function of many things. But you have to get to the housing 
unit, you have to have the mailing list, the address list, the 
mapping all done. That is a big component of cost. And that is 
probably the single most important basic phase, as Ms. Farrell 
pointed out, to making the key, the foundation for an accurate 
census.
    Senator Coburn. I guess probably the reason I am asking 
these questions is because the planning documents have not ever 
been brought forward on how you are assessing these costs? How 
you are doing it? How do you measure it? How do we get a look 
at it so that we have a confidence level?
    I will tell you that I will be your best friend or your 
worst enemy when it comes to getting extra money for the 
census. Because if it is not efficient--every year between now 
and 2010, we are going to be looking to make sure that the 
planning and the efficiency that can be gotten is going to be 
gotten there.
    The itemization of costs as a part of the planning document 
that has been asked for two appropriation cycles, that still is 
not there, let us just go to that question.
    Where is that document? When is it coming?
    Mr. Kincannon. I thought we had provided that information 
to the Congress in terms of the life cycle cost document, and 
quite a lot of dialogue about how we put that together and how 
we updated it. If we have not satisfied on that, then we need 
to get more specific.
    Senator Coburn. I will have staff follow up with you on 
that.
    Mr. Kincannon. Thank you.
    Senator Coburn. Let me make one other point. Welcome, 
Senator Carper. Glad you are here.
    According to our calculations from what we have gotten from 
you all, the non-response follow-up in 2000, from 2000 to 2010, 
by your own submission, will cost $1 billion more. But the 
overall costs are increasing by $5 billion. So if those numbers 
are right, 20 percent of the increase in costs is for the non-
response. What is the other 80 percent?
    I know you have $2 billion set inside for all of your 
mapping and the other programs. What is the other $2 billion?
    Mr. Kincannon. The other $2 billion is composed of changes 
in the number of people per housing unit, the cost of hiring 
and paying people, and does not yet even factor in the probable 
increased cost in security that we will be dealing with in 
hiring the number of people that we need.
    If we have not given you the linkage between how we 
composed the cost for 2010, then we can do that and we will do 
that.
    Senator Coburn. That will be very helpful to us. I am not 
sure that we have got that.
    I am not going to hold you to this. I just want you to 
guess. What do you think the highest possible total cost for 
the 2010 census is going to be?
    Mr. Kincannon. I would not expect a variance in real terms 
of more than say 5 to 7 percent. And I hope there will not be 
that much. That is a guess. That is not an administration 
statement.
    Senator Coburn. I understand that and you are on the record 
as a guess.
    Mr. Kincannon. I think it is important to look back at 2000 
and realize that we did not have the kind of careful planning, 
testing, revision of plans, and systematic moving forward that 
we have had so far for this census. Among other things, with a 
year left before the census, the Supreme Court handed down a 
decision that meant the Administration at that time and the 
Census Bureau had to completely revise plans on the ground for 
taking the census. If you do that kind of change late in the 
cycle, without speaking to the wisdom of the change or anything 
else, then you do have sharp increases at the very end. I hope 
we are not going to have that kind of change again.
    Senator Coburn. I hope so, too. The itemization of costs is 
a part of the planning document, that is one of the things that 
we want to see is the itemization of how you got there.
    Part of our problem, as Members of Congress, is trying to 
get our hands around an agency that you have your hands around 
and you are somewhat familiar with. We have to try to become 
familiar with that. And so more information is better, rather 
than less.
    I think I will stop now and welcome my co-chair, Senator 
Carper, for a short statement and any questions.

              OPENING STATEMENT OF SENATOR CARPER

    Senator Carper. Thanks, Mr. Chairman.
    I have no statement that I will give, but I do have one for 
the record, if I could offer that.
    Senator Coburn. It will be made part of the record.
    [The prepared statement of Senator Carper follows:]
              PREPARED OPENING STATEMENT OF SENATOR CARPER
    I want to thank Chairman Coburn for holding today's hearing to 
examine the costs and the information technology components of the 2010 
Census. Although Census costs double every decade, we must remember 
that the Census Bureau is tasked with an enormous undertaking--to count 
everyone in the United States. With the ever changing dynamics of the 
U.S. population, I believe the Cenus Bureau is doing the right thing by 
using information technology to help stream line tasks in the field.
    The questions that we will ask here today are whether or not those 
initiatives are being implemented in the most appropriate manner and 
through the most efficient means. Because information technology 
accounts for nearly 17 percent of the 2010 Census' total costs, poor 
oversight of various information technology components could have a 
disastrous affect on the success and cost of the Census. The Census 
Bureau has the responsbility to immediately address any risks before 
Census Day 2010.
    The Census Bureau has also decided not to offer the 2010 Census 
online. This is surprising, since e-government is a leading priority 
for our Federal Government. Various agencies have implemented Internet 
initiatives to help invididuals better communicate and do business with 
the Federal Government. This year at the IRS, online tax filing reached 
record levels.
    I look forward to hearing, in detail, the Census Bureau's reasoning 
for not offering the 2010 Census online and their decision to back-away 
from the e-government trend. Census Day 2010 is rapidly approaching. We 
each have a responsibility to ensure that the Census is conducted in 
the most efficient manner. I want to thank each of our witnesses for 
your service, and I look forward to your testimony.
    Thank you.

    Senator Carper. I suspect the Chairman has already delved 
into this, but I am going to come back and revisit it anyway.
    In the last couple of months, we have witnessed in this 
country an effort to sign up literally tens of millions of 
senior citizens for Medicare Part D prescription drug program. 
A lot of that has been done on the telephone, people call, wait 
to get somebody on the line and call back and finally maybe get 
somebody. They call my office, and they probably call Senator 
Coburn's office, as well, and we try to help, too.
    A lot of people, though, signed on to the benefit online. 
For those who did not have the computer skills were able to 
find people in their senior center or their family to help them 
to sign up online.
    We have tens of millions of people who file their taxes in 
the month of April or other times during the year. A lot of 
those folks did that online, as well.
    When I was governor of Delaware, we began filing State 
taxes, accepting State tax filings, online as well.
    I understand that the Census Bureau has considered whether 
or not there is a business case that justifies doing the census 
or part of the census online. And I understand that you have 
concluded that there is not.
    I would just ask for you, Mr. Kincannon, to talk about 
that, particularly in light of the work we have done in other 
areas involving the Federal Government, Medicare, and IRS.
    And then I would ask, Ms. Farrell, if you would comment on 
it, as well. But Mr. Kincannon, if you would take it first.
    Mr. Kincannon. The Internet is an enticing option and we 
use electronic reporting extensively in the business data that 
we collect. Businesses, particularly larger scale businesses, 
seem to find that a very efficient way of reporting for 
multiple establishments. So it is not as though we do not use 
the Internet and other electronic means of reporting when it 
seems to be received well by respondents.
    Almost all of the export data that we collect is collected 
in an automated form. And both the exporters and the Census 
Bureau like that very much because it is faster and more 
accurate, lower in cost for us and for them.
    We have tested Internet response to the short form only 
census because it is short, and it would seem like it would be 
an easier thing to handle online than an application for 
Medicare. And certainly--you send in your completed taxes based 
on commercial software that you file. You do not actually do 
your taxes online in most cases, although I guess in some cases 
they may do it with somebody's online system.
    What we found is that when we offered respondents, in a 
test, a controlled test, the chance to fill out the short form 
online, a few people did. My recollection it was less than 10 
percent, 7 to 10 percent. The total response rate of the people 
responding by Internet and by mail on paper was no greater than 
the control group. So we did not gain any net response. We did 
not do any more to reduce the costly non-response follow-up. 
That is the biggest cost driver in the census, and it is our 
target for trying to reduce that.
    When we conducted a test where we emphasized the importance 
of responding on the Internet, we sent people a letter or a 
card, I do not remember which, where we said we want you to 
complete this form. Go to this site, use this control number so 
we know who you are and where you are, what your address is. 
And if you do not have access to a computer or do not wish to 
use the Internet, call this number and we will mail you a 
questionnaire.
    The overall response plummeted. About 30 percent of people 
did file on the Internet, but the total response was less than 
half the universe that we expected.
    So looking at our experience there, we do not see that we 
gain any business advantage of reduced cost or being able to 
predictably reduce substantially our infrastructure for 
handling the paper questionnaires. I do not know why that is, 
but it is a fact that we have tested that, and that is the 
indication.
    It may be that the paper questionnaire, being only about 8 
questions, tests takes about 10 minutes for a family of four to 
fill it out. The easiest thing to do is just to fill it out and 
mail it in. Or maybe people decide they are going to do it on 
the Internet and then do not get around to doing it.
    I do not know the explanation.
    Senator Carper. Let me just interrupt you for a moment.
    Roughly how much does it cost per household to get them to 
complete and submit their questionnaire for the census? Can you 
attribute a cost of that? Is it $50, $60, or $70 for 
responders?
    Mr. Kincannon. For people returning their questionnaires, 
for responders? I cannot. Do we know the cost?
    I do not know, but let us say it is $10. I do not know what 
the cost is. You print the questionnaire. You mail it. You pay 
the postage coming back, and you scan it in. It is very modest.
    Senator Carper. Does that include all of the costs? Is 
there something missing there?
    Mr. Kincannon. I do not know whether it includes all of the 
costs. It includes the operational cost of sending out and 
receiving.
    Senator Carper. I think the Chairman said those are the 
variable costs.
    Chairman Coburn. Those are the variable costs.
    Senator Carper. And are there fixed costs that you are able 
to----
    Mr. Kincannon. Sure. You have to have the maps, you have to 
have the tabulating software and all kinds of things to deal 
with that. And you have to have all of the receiving, scanning, 
and other kinds of equipment there to do.
    If you take responses also on the Internet, you have to 
have a means of converting those to the same compatible format 
with this other information.
    So that is all fixed costs. You have to do that if you get 
one back by Internet or two.
    Senator Carper. Let me just continue on where I am going. 
Could you conceive of a situation where we could significantly 
increase the percentage of folks who would respond online by 
offering them, rather than just to say thank you but offering 
them some kind of financial remuneration for those who 
responded online?
    Mr. Kincannon. There is a good deal of evidence in survey 
research literature that offering cash incentives or other 
kinds of incentives can have an effect on response. But it also 
costs something.
    Senator Carper. Have you all ever looked at whether or not 
the amount of remuneration that might be called for would more 
than pay for itself?
    Mr. Kincannon. I am not aware that we have looked at that 
on the census. We have examined it and do use incentives on 
household surveys. And we may have looked at it, but I am not 
aware of that.
    Senator Carper. Thank you for your responses. Let me turn 
to Ms. Farrell if I could, and your comments on these issues, 
please.
    Ms. Farrell. The Bureau raises some important 
considerations regarding the security of the Internet and the 
cost savings. As technology has advanced, we know that Federal 
agencies have found the benefits of using the Internet and 
other collections. And it should be noted that GAO did put 
information security on our high-risk list back in 1997. But 
that does not mean that those obstacles cannot be overcome, and 
that they should not be explored to be overcome.
    We have not seen what the business case is behind the 
Bureau's decision to drop the Internet. We have asked. We were 
told that there was not a business case made for that 
determination. But the decision was a sound business decision.
    Senator Carper. Would you say that again, please? Just 
repeat what you said.
    Ms. Farrell. In terms of the business case, we were under 
the impression that the Bureau had developed a sound business 
case to base that decision to drop the Internet from their 
contract that was let last October. But when we asked for such 
information, we were informed that we had misunderstood and 
that there was no business case that they had actually 
developed.
    I think it is important to note that the Bureau did explore 
and offered the Internet as an option for the 2000 census, and 
they had a low response rate. it perhaps could have been 
because of low advertising. We do not know. We have not seen 
what the Bureau has done to explore the use of the Internet 
from 2000.
    It has been puzzling to us, as to when the Bureau did 
mention its use of the Internet in its 2000 life cycle cost 
estimate, which is a very top level cost estimate without the 
itemized cost that you are referring to, Mr. Chairman, they 
referred to it as a possible cost savings. By the time they did 
a revision 2 years later, they noted that the response rate was 
not as high as they had anticipated it would be.
    But following that June 2003 referral to the response rate 
not being as high, was included--our understanding, in the 
contract that was let in 2005 to offer it.
    Thus, we just feel that the decision to drop the Internet 
has raised more questions about what the decision was based on 
and what the true facts are behind the response rate and how it 
was offered.
    Senator Carper. We are not the only country that does a 
census. I presume most of the major countries in the world do a 
census. I do not know if they do it every 10 years. Can you 
just give us some idea, Mr. Kincannon, if that is the case?
    Mr. Kincannon. Most countries throughout the world do 
conduct censuses, some at irregular intervals, some every 10 
years, a few every 5 years. Increasingly, countries, 
particularly in Latin America and some European countries, are 
moving to activities somewhat like the ACS where a part of the 
census is taken on a continuing basis and if an enumeration is 
legally needed, they take that.
    A number of European countries no longer take a census. 
Either they rely on a population register or other kinds of 
administrative records as a basis for an estimate of 
population. And they may use that as a basis for surveys. You 
know we use our census as a basis sampling frame for surveys.
    We do not have a population register. We do not have any 
consistent or coherent set of administrative records that form 
the equivalent of a census.
    Most European countries, frankly, do not have very 
dramatically changing populations, either because of natural 
increase or immigration. So we face a different situation.
    Senator Carper. Let me just ask, if I can, Mr. Chairman, 
just one follow-up question. Are we aware of some practices 
that other countries are following what we might want to 
consider emulating? Are there some best practices out there, 
that either of you are aware of, that we have borrowed from or 
maybe we ought to? Particularly with respect to the use of the 
Internet.
    Mr. Kincannon. We have examined use of the Internet in some 
other countries, in Australia, New Zealand, and Canada. In some 
cases they have a slightly higher return rate on the Internet 
than we have had in our tests. But in no cases, in their view, 
has it managed to save them money as an offset by increasing 
total response. Again, I do not know how to explain that, but 
it does seem to be a similar experience.
    Canada offered it to everybody because under Canadian law 
government communications must be available to people in 
Internet form, as well as other forms, and in two languages. 
But they found it cost them more and did not, as my 
understanding at this stage of things, that it has not 
increased overall response.
    We do look at what other countries do. There are systematic 
examinations, particularly done through U.N. bodies, where 
methods are looked at cross-country and shared. And we have, 
over time, incorporated some of the ways that other countries 
have improved their censuses and vice versa.
    Senator Carper. Thank you.
    Ms. Farrell, anything that you want to add to that, 
quickly?
    Ms. Farrell. No, we have not looked at other countries. We 
are aware of the Canadians, but we have not actually studied 
them.
    Senator Carper. Thank you both.
    Senator Coburn. I am a little bit aware. The Canadians just 
completed their first one. They had a 22 percent participation 
rate. That is three times what you testified that your test 
was. And the number that you all tested was, I think you will 
agree, an extremely small number in your test batch; correct?
    Mr. Kincannon. 250,000 households.
    Senator Coburn. 250,000 households. And that was done 2 
years ago; is that right?
    Mr. Kincannon. In 2003 and 2005.
    Senator Coburn. So it was done in 2003 and 2005.
    Mr. Kincannon. That encompasses the control group and the 
test groups.
    Senator Coburn. So if, in fact, you just had a 22 percent 
response rate in the United States, you would save $300 million 
online. You said it is $10 variable cost to mail it out, to 
have them fill it out, pay the postage and bring it back and 
then code it in. To do that online, you would save $300 million 
if you only had 22 percent.
    And then you divide that by $80 rather than $88 for a non-
responder, and what you get is you can contact another 20,000 
homes by the money that you could save, or 25,000 homes--no, 
35,000 homes, with the money you could save just if you had a 
22 percent response rate.
    Mr. Kincannon. Mr. Chairman, if we got a 70 percent 
response rate, we could pay off part of the national debt, I 
suppose. But we do not have that.
    Senator Coburn. No, we cannot.
    The point is that you are looking at the box as it is 
today, and I am wanting you to look at the box at what it can 
be on the Internet. Things have changed between now and 2000, 
in terms of the response rate. The Internet changes so fast.
    And the fact is that most people, if given the opportunity 
and the inducement, or at least the awareness through 
advertisement, I would guarantee if you just polled them. Would 
you rather fill out something online or fill a piece of paper 
out and put it in the mail, they would much rather--90 percent 
of the people who are computer literate in this country would 
rather send it the other way.
    So if, in fact, there are savings to be made by a small 
number, if you only got a quarter of the people doing it, you 
would tremendously save money both in terms of the variable 
costs, but also in terms of the non-responder cost.
    And so I do not understand why you take at a point in time 
now and say because we had this one test, that we are going to 
make an assumption that in 2010 we are not going to use the 
most modern communication methods that we have, that have all 
of the potential, and then try to promote them. Rather than to 
say work we are going to throw this out and we are not going to 
utilize this system that everybody already has, 74 percent of 
the households in this country already have this tool.
    If you had 74 percent of them, that is 100 million. That is 
$1 billion that you would save if you could just get them 
online. That $1 billion would come close to paying for a lot of 
the cost of the non-responders.
    Mr. Kincannon. The Canadian response rate, calculated in 
the same terms that we did, would be 14 percent, not 22 
percent. If you take it as percent of the universe invited to 
respond, as opposed to the 22 percent, which is a percent of 
the actual responders. But still, the point remains.
    I would like to know what form your guarantee would take? 
You said you would guarantee that.
    Senator Coburn. A figure of speech.
    The fact is, where is the large test to see what you would 
do? You have done 250,000 people in 2003 and 2005 on a cost 
project that is $25 million. I mean, $25 million, you can put 
this package in. And you could utilize--$25 million compared to 
the cost that you all are going to spend to have a package that 
would allow people to do this, to me, seems a small price to 
try that experiment.
    And then if you promote it, what about just the $10 per 
household that you would save on the people that might file? 
That is not worth it?
    Mr. Kincannon. The Canadians did not save any money either. 
Did they tell you they saved money?
    Senator Coburn. No, we have not finished with the 
Canadians.
    Mr. Kincannon. We asked them about that.
    Senator Coburn. But the point is that this is the first 
year. Under the leadership that I see now, we are never going 
to get to the Internet on this because we are never going to be 
able to say in advance that we can get there.
    I would just tell you, step back for a minute and look at 
everything. People did not used to bank online. You could not 
trust to pay your bills online. You could not use a credit card 
online. You could not do any of those things.
    If the people would have had the same attitude, we would 
not be doing any of the stuff online now.
    What I am asking you to do is reconsider and relook at 
this. And I am interested in how is it that we cannot figure 
out some way to utilize this technology to save us money? And 
what you all have said is we cannot. You have not said maybe 
there is another possibility. You have not said maybe our data 
was wrong. Maybe we ought to take another look at it. You have 
said to heck with it for 2010. And the next shot we get at it 
is 2020.
    And with the costs rising the way they are, this government 
cannot afford one penny overspending anywhere because we are 
stealing it from our grandchildren.
    And so for us to totally 180 degrees say no Internet, not 
going to do it on the 2010 census, says well then, when we get 
some visionary leadership in 2010, we are going to be 10 years 
behind.
    And what I am saying is there has got to be some minds out 
there that can figure out how do we utilize this technology in 
your area of expertise to save this country money?
    I cannot believe that we cannot create a way to do it. 
Whether it is incentivizing, as Senator Carper said. We will 
give you a $5 Baskin-Robbins ice cream cone credit or 
something.
    Senator Carper. I was thinking of pizza for four.
    Senator Coburn. I do not know. But the point is people 
respond. And to totally reject that, I am having trouble 
understanding why that has just been totally taken off the 
table when everything else we are trying to do is to move to 
that direction. So to me, it is not computing.
    What I hear, even the data that you give us, it is kind of 
like this: You have responded, in terms of the Census estimated 
life cycle cost. But there is no detail. You have got total 
cost, $1,707,000,000. No detail on American Community Service. 
MAF/TIGER, $534 million. There is no detail where those costs 
are. All you are doing is listing out what the costs are.
    What we are asking for is where are the details of the 
costs? It is kind of what the GAO has said. What makes it up? 
Why is that not transparent? Why is it not online for all of us 
to be able to see what those costs are?
    That is where this government is going to move. The 
American people are going to be able to see every penny you 
spend at the Census department and why. And the same thing for 
where the GAO spends their money and why, and where we spend 
our money and why. It is going to become available.
    And so to not utilize this technology sets us back not just 
for the 2010 census, it sets us back for the 2020 census and 
the 2030 census. And we cannot afford these cost increases.
    And I, quite frankly, do not buy that there is nothing to 
be gained. I think your testimony is 130 million households 
that you think we are going to have this time? Is that right?
    Mr. Kincannon. Yes.
    Senator Coburn. Three hundred ten million people?
    Mr. Kincannon. Yes.
    Senator Coburn. And that there is no efficiency of scale. 
That there is nothing to be gained by a larger population. It 
is all totally offset because the mix and the complexity, and 
there is a rising number of seniors, that cost--and what is the 
one tool that we know that will not cost much to use, which is 
the Internet, and we are throwing it out.
    Senator Carper. Mr. Kincannon, before you respond, Mr. 
Chairman let me just throw something out, listening to this 
exchange.
    I do not know if there is something that they could do, the 
Census Bureau could do, in conjunction with the Census in 2010 
that would enable us to test a number of different approaches 
to figure out when the next census rolls around in 2020, we 
will have had an opportunity to find out what works and what 
does not work, in terms of getting people to migrate to the 
Internet.
    That is just something I would throw out there for your 
consideration.
    Mr. Kincannon. We can certainly test, and you do not wait 
until 2010 to decide what you are doing about 2010. You do not 
wait till 2020 to see if you examine the question of the 
Internet again.
    I think that we should continue testing that in the coming 
decade and see if we can find ways that either incentivize or 
people become more accustomed to it.
    There are a number of things that I would like to say about 
what you said. First, in the course of every decade, there is a 
period of time when you plan, when you test, and then you have 
to lock everything in. The time when we lock everything in 
always seems unreasonably early to people who sit up here in 
this neighborhood.
    Senator Coburn. I understand that.
    Mr. Kincannon. But we have, as Ms. Farrell said, we have a 
high risk situation. We have one chance to succeed. And we have 
to make sure everything is tested and will work right in 2010.
    Even at that, it is a risky proposition because you do not 
know what may happen, what mood may strike the public and 
inflame their concerns on some particular aspect of it and make 
it difficult for you.
    We will have natural disasters during censuses, a big 
hurricane, a volcano exploding, all of these things have 
happened in Census times. And we have to cope with it. But they 
never affect the entire country.
    So we have tested, and these were extensive, significant 
tests. They do not show us how they are going to reduce 
significantly the cost of the census.
    And they do increase costs. You talk about we all do online 
banking. Me, too. I do online banking probably every week, 3 
weeks out of 4, at any rate. And I do that with a well 
established set of software and high security that is developed 
because the clients of that bank use that every week, and any 
of them every day, I am sure.
    We are talking about something that will be used once a 
decade. That means the investment in security costs 
particularly are going to be very substantial and not spread 
over long periods of time.
    Senator Coburn. You already have that investment in 
security on your American Community Survey that you are doing 
now. That is not secure?
    Mr. Kincannon. We do not accept reports. We tested but it 
did not work out to use the Internet as reporting.
    Senator Coburn. But the point is was there not security 
associated with that?
    Mr. Kincannon. There is security in the way that we 
collect----
    Senator Coburn. Was there security associated with the 
other data that you collect?
    Mr. Kincannon. Yes, sir, but that is not the same thing as 
security on an Internet site. That is a separate set of issues.
    Senator Coburn. I am talking about the people who respond 
to you on the Internet now, like your testimony was earlier, 
that you collect two different sets of information now that are 
filed online. Is that not secure information?
    Mr. Kincannon. It is. And those reports come to us on 
monthly and quarterly and even daily basis. So it is a system 
that is in constant use.
    Senator Coburn. I do not understand if somebody uses 
something once how that changes the complexity of the security 
of a system that would make it unusable for people in this 
country.
    Mr. Kincannon. It is a different system because you are 
getting different inputs from different kinds of respondents. 
You would have to build something different for the 2010 
census.
    Senator Coburn. The number of questions on a census survey 
is how many?
    Mr. Kincannon. On the short form? It is about eight 
questions.
    Senator Coburn. All right, eight questions. And I want all 
the Internet designers out there in the world that are doing 
right now eight questions on 130 million homes, what does it 
cost, and what is the technology that has already been 
developed a number of times in this country, what is it banking 
on, or the IRS or everybody else that has already developed the 
security.
    That is not a satisfactory answer. That data, that 
technology is already out there. That is a $25 million cost at 
the most. We have already talked with all the vendors around 
the country. We spent the time doing it. That is not a 
satisfactory answer. That is not a reason not to do it.
    Again, I just go back, if it is a $10 cost, and it may not 
be $10. It may be $7. That may be why the numbers do not add 
up. But if your variable costs in mailing out a censuses is $10 
per household, all you have to do is get eight households to 
file online to totally pay for one that is a non-responder.
    If it were me, I would be sitting there looking at how in 
the world do we get 80 million people in this country, 80 
million households, to respond online? In other words, ask the 
question the other way?
    The technology is not a problem. You would agree with that. 
The technology can be gotten.
    Mr. Kincannon. It can be gotten, but it is not cost-free.
    Senator Coburn. No, it is not cost-free but what was the 
contract cost that you had on the contract that you all 
terminated?
    What was the cost of the contract that you terminated for 
online Internet census?
    Mr. Kincannon. I am not sure that we terminated a contract.
    Senator Coburn. A $7 million contract with Lockheed.
    Mr. Kincannon. We spent $7 million for the first 2 years of 
work on this with Lockheed.
    Senator Coburn. What was the total contract price?
    Mr. Kincannon. The total contract would have been an 
additional $30 million. But the price to pay for that also 
meant that they would not be able to provide the DRIS for the 
dress rehearsal.
    Senator Coburn. Do you mean, they could not do both? 
Lockheed could not do it? Or we just did not negotiate a 
contract for it?
    Mr. Kincannon. They could not do both in that time schedule 
within the budget that was appropriated to us, of course.
    Senator Coburn. People who file their income tax returns, 
individuals, do it once a year. Once a year with the IRS, that 
is all they file. And you know, 70 million of them did that 
this last April. How do you explain that, when you say people 
cannot file once a year or every 10 years? They cannot 
negotiate the Internet to file a census return?
    Mr. Kincannon. Ninety percent of those who filed had a 
considerable incentive because they were getting a refund. And 
in addition, they paid $30, $40 or $50 for the software 
provided by a private-sector firm to fill out. And then they 
reported to the software vendor, which then relayed it to the 
IRS.
    Senator Coburn. Right, and that is a 30-page form, and we 
are talking about a single page form with eight questions on 
it.
    Mr. Kincannon. Yes.
    Senator Coburn. So the cost difference is not there. I am 
still astounded.
    Mr. Kincannon. The cost for that kind of filing is paid for 
by the filer of the taxes.
    Senator Coburn. Let me go back and ask a question. What is 
wrong with this question? How is it that we, at the Census 
Bureau, figure out a way to reduce the cost by incentivizing 
online filing or online participation with the census, so that 
we have a greater participation, less mail out, and less non-
compliance? Where is the answers to that?
    Mr. Kincannon. The answer to that is in 2011 and 2012, not 
in 2010. We do not have time to test and prove and rehearse 
with a significantly changed method of taking in the data.
    Senator Coburn. When did we start looking at online?
    Mr. Kincannon. Before, in 2001, I suppose because we----
    Senator Coburn. You had a sample on it in 2000.
    Mr. Kincannon. Yes, then we looked at it before that. We 
had that evidence. I thought you meant for this decade.
    We started probably in 2001 getting ready for the test in 
2003, which was the first of the quarter million size test of 
Internet.
    Senator Coburn. There is some question about your handheld 
devices for your enumerators and the accuracy and efficiency of 
those. Could you address those for me and tell me where we are?
    Mr. Kincannon. Yes, Chairman. We have awarded a contract 
this spring to the Harris Company to develop the handheld 
devices that will meet our requirements and will be tested in 
the dress rehearsal and used in 2010. We used devices that we 
made ourselves for testing leading up to that, so that we could 
test the different aspects of using it.
    The devices we built were far less efficient than those 
that can be provided by the private sector, but we learned from 
those tests: A, that someone else could do that task for us 
better than we could do it; but B, that the functionality could 
be handled on handheld devices both for address listing and 
update, for payrolling, for sending maps to enumerators, for 
revising their day's assignment for non-response follow-up 
based on late receipts.
    In the test in Austin, we saved useless calls on people, 
17,000 cases, where households had sent their questionnaires 
back late. And so that saved more than--proportionately more 
than the Internet would save, if you are looking at that.
    Senator Coburn. So do you have a functioning model that 
works today?
    Mr. Kincannon. We had a functioning model that was used in 
the test census in----
    Senator Coburn. It was made by Harris?
    Mr. Kincannon. No, we made that. I do not know who made it.
    Harris made it but it was not a production model, not the 
model that we want for the census.
    Senator Coleman. That is all going to be automatically 
downloaded; right? You are not going to hand-download that? 
That is going to go to a computer and be downloaded; right?
    Mr. Kincannon. It will go to the computer and be 
downloaded, at the end of every workday, either wirelessly or 
overland line, depending on the circumstance and working 
conditions of that enumerator.
    Senator Coburn. What happens if they do not work? What is 
your plan B?
    Mr. Kincannon. They will work. They have worked. You might 
as well ask me what happens if the Postal Service refuses to 
deliver the census forms.
    Senator Coburn. I am not asking it facetiously. I am asking 
you what happens if there is a computer glitch and these 
handheld devices do not work? What is the plan B?
    Mr. Kincannon. The computer devices have been tested and 
proven to work.
    Senator Coburn. All I want you to do is answer my question. 
What if they do not work?
    Mr. Kincannon. We have a big problem then.
    Senator Coburn. So are you going to have to hire more 
people to do the non-response?
    Mr. Kincannon. I do not believe that condition will obtain, 
so I do not----
    Senator Coburn. So there is no planning. So, as we have 
talked about this planning of what-ifs and----
    Mr. Kincannon. We could hire more people. Yes, we could 
hire more people, sir.
    Senator Coburn. Is it not true that GAO has said that this 
handheld device is a huge risk in their testimony?
    Mr. Kincannon. I do not know the precise formulation of 
words, but they say there is a risk associated with using 
handhelds.
    Senator Coburn. So your testimony is to me that there is no 
alternative plan if that does not work?
    Mr. Kincannon. We have no reason to believe that there is 
any systematic risk in all the handhelds. That system will 
work.
    Senator Coburn. Your testimony today is if that does not 
work, if GAO's concerns happen to be borne out, there is no 
alternative plan if it does not work?
    Mr. Kincannon. We would have to hire more people to conduct 
traditional pencil and paper non-response follow-up?
    Senator Coburn. As we did in 2000?
    Mr. Kincannon. Yes, and 1940.
    Senator Coburn. GAO has raised some concerns about the 
level of transparency within your budgeting process. Do you 
believe that your budget estimates are adequately transparent 
for long-term planning for you, but also for us to watch you 
and look at you?
    Mr. Kincannon. I think we can always have improvements in 
transparency internally for planning, and we endeavor to 
improve the collection of cost data, the documentation of cost 
data. But I am sure we still have room for improvement. I do 
not know, apparently we have not provided to this Subcommittee 
the degree of transparency that they want.
    We have provided a lot of information to the Appropriations 
committees on both sides, and maybe that same information could 
be useful to this Subcommittee.
    Senator Coburn. Let me raise just a couple of other 
questions and then I want Ms. Farrell to comment on it.
    You all have a PART evaluation, as every agency within the 
Executive Branch has. The PART assessment had some concerns 
that Census Bureau managers are not held accountable for cost 
containment. Is that a legitimate criticism? And if so, have 
there been steps made to adjust to that?
    Mr. Kincannon. I do not recall that particular finding but 
I believe that managers in the Census Bureau are held 
accountable for cost containment. But that is a principle that 
we try to follow. We do not give money to people in plain brown 
wrappers for them to spend without accountability for doing 
that.
    Senator Coburn. I do not think that is what they are 
talking about. They are talking about systems. The PART 
assessment is do you have the systems and control to be able to 
effectively manage and measure and to have performance 
measurements to know whether or not you have cost containment 
and whether or not somebody is managing something effectively.
    Mr. Kincannon. I think that we do for large programs and 
for continuing programs, in general.
    Senator Coburn. Thank you.
    Are there going to be any consequences--and again, not 
holding you to your 7 percent, let us say 10 or 15 percent. Are 
there any consequences if you run to $15 billion? Should there 
be any consequences to the management inside the Census Bureau 
if it cost $15 billion instead of $11.3 billion?
    Mr. Kincannon. I would think so, yes. That seems 
reasonable.
    Senator Coburn. OK, that is a great answer.
    Mr. Kincannon. I mean, what do you want me to say? Detail 
the punishment or retribution or the guidance or what?
    Senator Coburn. What I am looking for is you have got a 
PART analysis that says you do not have great management 
systems in place to measure cost containment. And if you do 
not, and that is the assessment by the CFOs that look at the 
PART of each agency. They have this wonderful color-coded 
network and they are measuring performance on how everybody is 
improving every year to try to get to the point is if there is 
no consequences--in other words, should somebody be promoted? 
Should somebody not be there anymore if, in fact, we do not 
have good management. That is the question I am asking you.
    The philosophy is yes or no. I am just asking is there the 
management tools in there to say--you know it is the 
expectation of being held accountable. Just like you guys are 
going to be back here in 8 months to answer some of these 
questions and see where we are. Because we are not going to 
spend $4 billion more to do this. We are not going to do it. 
The next two generations are not going to pay for inefficiency 
in the Federal Government.
    So the question is should there be accountability? Is there 
line management? Is there structure? Are there management tools 
there to measure? To know before costs get out of control that 
you know ahead of time that we are getting ready to lose 
control of costs?
    That is what the PART assessment is. It is not about 
personalities, it is about systems.
    Mr. Kincannon. I do not think it is about personalities. I 
did not say it was about personalities.
    I will look at that particular PART finding. I am not aware 
that is there, but I will take a look at that and try to 
understand it better.
    Yes, I do think there should be----
    Senator Coburn. When was the last time you looked at the 
PART system on your agency?
    Mr. Kincannon. About 2 months ago.
    Senator Coburn. And you did not notice that was there?
    Mr. Kincannon. I looked at summary level PART reporting, 
yes.
    Senator Coburn. Ms. Farrell, if you were to look at the 
Census Bureau right now, from what you all have looked at, and 
looking at costs for 2010, is there any one particular thing 
that you would recommend be done to control costs that are not 
being done today?
    Ms. Farrell. It is back to what we have been discussing 
with transparency. It is difficult for us or for you to know 
where the Bureau is in their planning without more information 
behind how that $11 billion was comprised.
    At the same time, I do think it is important to note that 
the Bureau has designed this census earlier in this decade 
compared to where they were at the same point with the last 
2000 census. But the question is, we do not know if that $11 
billion, if it is over. It could be under. We really do not 
know because we have not seen what is behind it.
    Half of the costs are in the field data collection 
mechanisms. And what Dr. Kincannon said about the non-response 
is true, that non-response is probably one of the biggest 
drivers of the cost.
    So if you can get hold of that and find out why people are 
not participating or why it is so difficult to find them and 
make those corrections, you stand a better chance of increasing 
your response rate.
    Senator Coburn. It would make sense though, with the short 
form being the form used this time, that the response rate 
should climb significantly.
    Ms. Farrell. The figures that the Bureau shared with us 
showed that the short form would probably increase the response 
rate, I believe, by 1 percent.
    The bigger bang for the buck is going to be with the 
targeted second mailing, which I think could be 7 or perhaps 
greater percentage in increasing that non-response rate.
    Senator Coburn. One concern I had, in reading your 
testimony and looking at this, is let us say we are about to 
get started planning. You are a year away from the 2010 census. 
And let us say we have the same unemployment rate that we have 
today. Where are you going to get 500,000 people to work on the 
non-responders? And what are you going to have to pay for them? 
That is a real problem that you are going to be faced with.
    Mr. Kincannon. Well, we are still 4 years away and I am not 
aware that anybody is predicting the unemployment rate in 4 
years. If the labor market is very tight, it will cost us more 
to hire people. It cost us more in 2000 to hire people.
    But we live in a market economy. And if labor is tight, 
then we will need to pay to get that. We do not have any other 
source of labor than paying people a reasonably close to market 
rate.
    Senator Coburn. Typically, the people that you hire, are 
they underemployed somewhere else, unemployed or retired? What 
is the mix of the people that you utilize in this non-responder 
army that you have?
    Mr. Kincannon. I do not have any statistical information at 
my fingertips and I am not sure how thorough that is anyway. We 
do attract people into the labor force who are not in it, 
people who are retired, in some cases. People have rather long 
retirements in this country now, and they like to do something 
that is interesting and constructive for a period of time.
    There are still not 100 percent of working-age women 
engaged, and some like to come back to work for a while. Some 
use it as a reentry point after childbearing years. There are 
still women who stay at home and take care of their children 
and they want a reentry and they find that useful.
    There are young people who may not have a very good job and 
they want to add something to their resume.
    This does not necessarily apply as much to the people 
working for a short period on non-response follow-up, but we 
still have tens of thousands of jobs that last a year or more 
in office work. So there are a variety of sources there.
    A lot of this work, the large number of people that do non-
response follow-up, basically have to work in late afternoon, 
evenings, and weekends. So it is a second job.
    Senator Coburn. So they can catch people at home.
    Mr. Kincannon. Yes, that is right.
    Senator Coburn. I want to thank each of you. I want to give 
you, especially you, Mr. Kincannon, an opportunity to say 
anything that you want to say, and offer for the record 
anything where we have had a disagreement or anything, to make 
sure that you can put in what you want to have in the record to 
balance out anything where I might not have seemed fair or been 
fair with you.
    Mr. Kincannon. I think you are a hard salesman in your 
point of view. I would not call you unfair, at least not on 
this day.
    Senator Coburn. A lot of people do, so it is fine.
    Mr. Kincannon. You are coming from a certain point of view 
and you push at it very hard. That is all right.
    I think that we have tested fairly the Internet possibility 
for response at the time that we had to make a decision for 
what we were going to do with that. That does not mean we 
foreclose that possibility in the future. And it may be that it 
will work better and we will learn better ways of incentivizing 
it in the future.
    I do not know whether the Congress as a whole would agree 
to incentivize something that is already a mandatory 
requirement in the law, but that will be your job maybe to sell 
that.
    So I disagree with your point of view that we have out of 
hand rejected something. We have tested it and not found it 
produced results that justified our going down that path.
    I believe that we have constrained cost in the census. And 
looking at the table of figures put into 2010 constant dollars, 
the housing unit cost increase in the decade of the 1980s 
leading up to the 1990 census was half that of the increase in 
the previous decade. I was Deputy Director in that period. I 
did not do that alone, but a lot of people working in the 
Census Bureau were conscious of the need to constrain growth in 
costs. And we were successful.
    This may not meet your standard, but still it is cutting in 
half the rate of increase. And the projected rate of change 
for----
    Senator Coburn. It is. Our chart shows that, as well.
    Mr. Kincannon. So I think we have shown that we can be 
effective in constraining costs. It does not look like we or 
the Congress or whatever, the government, was as successful in 
doing that in the lead up to the 2000 Census. So it shows we 
can do that and we should continue to be as effective as we 
can. And avoid late changes in the way that we are going to 
process the census.
    Senator Coburn. Which have big impact on your costs.
    Mr. Kincannon. Yes.
    Senator Coburn. Let me clarify something, just so those 
that work with you and your agency. I do not doubt the desires 
at all or the work ethic of the people who are there. We have a 
big problem in our country and we have got 9 years to fix it, a 
big asteroid, a financial asteroid is going to hit this country 
at 2016. And we cannot just look at the census. We have to look 
everywhere.
    You are not the only agency. This is our 36th hearing on 
oversight on waste, fraud, and abuse. How do we do it better? 
How do we get accountability, transparency, results? So it is 
not about the Census Bureau or their employees. it is about how 
do we get and create the same opportunities for our children 
and our grandchildren?
    I appreciate the fact that you have spent a lifetime of 
service to our country. And my questioning you does not demean 
that at all and it is not meant to do that, nor any of your 
employees.
    And I know a lot of the volunteers that worked in Oklahoma 
in the last census, and they put in a lot of time. They were 
happy to do it and felt a great part of our country.
    Nevertheless, every penny, every day that we can save is a 
standard of living change for our children and our 
grandchildren. And so we are not going to let up. We are going 
to keep working it. We are going to keep coming back. We are 
going to be still hounding you, asking questions.
    And we do want details. Sometimes inside the forest you 
cannot see the trees. And so different perspectives. My staff 
changes mine all the time when I am asking questions and they 
are asking questions of me.
    But this idea of transparency. Where do you get your budget 
numbers? What makes them? What are the assumptions that make 
those up? What are the components? Why cannot GAO see that? Is 
there a reason they cannot have that? Is there a reason we 
cannot have that? What is wrong with that? We have to create 
that kind of transparency.
    So your service is appreciated and the fact that what you 
are doing is very important. We understand that. We are anxious 
that it be done right but also efficiently.
    Ms. Farrell, any comments?
    Ms. Farrell. Sir, I just would like to thank the Bureau for 
the cooperation we have received from them as we continue to 
monitor their activities, and to emphasize that we do agree 
with the Bureau that at this time any significant change to the 
design could increase costs. But it does not mean that we 
cannot still be looking for ways that there could be a greater 
payoff down the road.
    Senator Coburn. Thank you all, very much. The hearing is 
adjourned.
    [Whereupon, at 3:57 p.m., the Subcommittee was adjourned.]
      

                            A P P E N D I X

                              ----------                              

[GRAPHIC] [TIFF OMITTED] 29501.002

[GRAPHIC] [TIFF OMITTED] 29501.003

[GRAPHIC] [TIFF OMITTED] 29501.001

[GRAPHIC] [TIFF OMITTED] 29501.004

[GRAPHIC] [TIFF OMITTED] 29501.005

[GRAPHIC] [TIFF OMITTED] 29501.006

[GRAPHIC] [TIFF OMITTED] 29501.007

[GRAPHIC] [TIFF OMITTED] 29501.008

[GRAPHIC] [TIFF OMITTED] 29501.009

[GRAPHIC] [TIFF OMITTED] 29501.010

[GRAPHIC] [TIFF OMITTED] 29501.011

[GRAPHIC] [TIFF OMITTED] 29501.012

[GRAPHIC] [TIFF OMITTED] 29501.013

[GRAPHIC] [TIFF OMITTED] 29501.014

[GRAPHIC] [TIFF OMITTED] 29501.015

[GRAPHIC] [TIFF OMITTED] 29501.016

[GRAPHIC] [TIFF OMITTED] 29501.017

[GRAPHIC] [TIFF OMITTED] 29501.018

[GRAPHIC] [TIFF OMITTED] 29501.019

[GRAPHIC] [TIFF OMITTED] 29501.020

[GRAPHIC] [TIFF OMITTED] 29501.021

[GRAPHIC] [TIFF OMITTED] 29501.022

[GRAPHIC] [TIFF OMITTED] 29501.023

[GRAPHIC] [TIFF OMITTED] 29501.024

[GRAPHIC] [TIFF OMITTED] 29501.025

[GRAPHIC] [TIFF OMITTED] 29501.026

[GRAPHIC] [TIFF OMITTED] 29501.027

[GRAPHIC] [TIFF OMITTED] 29501.028

[GRAPHIC] [TIFF OMITTED] 29501.029

[GRAPHIC] [TIFF OMITTED] 29501.030

[GRAPHIC] [TIFF OMITTED] 29501.031

[GRAPHIC] [TIFF OMITTED] 29501.032

[GRAPHIC] [TIFF OMITTED] 29501.033

[GRAPHIC] [TIFF OMITTED] 29501.034

[GRAPHIC] [TIFF OMITTED] 29501.035

[GRAPHIC] [TIFF OMITTED] 29501.036

[GRAPHIC] [TIFF OMITTED] 29501.037

[GRAPHIC] [TIFF OMITTED] 29501.038

[GRAPHIC] [TIFF OMITTED] 29501.039

[GRAPHIC] [TIFF OMITTED] 29501.040

[GRAPHIC] [TIFF OMITTED] 29501.041

[GRAPHIC] [TIFF OMITTED] 29501.042

[GRAPHIC] [TIFF OMITTED] 29501.043

[GRAPHIC] [TIFF OMITTED] 29501.044

[GRAPHIC] [TIFF OMITTED] 29501.045

[GRAPHIC] [TIFF OMITTED] 29501.046

[GRAPHIC] [TIFF OMITTED] 29501.047

[GRAPHIC] [TIFF OMITTED] 29501.048

[GRAPHIC] [TIFF OMITTED] 29501.049

[GRAPHIC] [TIFF OMITTED] 29501.050

[GRAPHIC] [TIFF OMITTED] 29501.051

[GRAPHIC] [TIFF OMITTED] 29501.052

[GRAPHIC] [TIFF OMITTED] 29501.053

[GRAPHIC] [TIFF OMITTED] 29501.054

[GRAPHIC] [TIFF OMITTED] 29501.055

[GRAPHIC] [TIFF OMITTED] 29501.056

[GRAPHIC] [TIFF OMITTED] 29501.057

[GRAPHIC] [TIFF OMITTED] 29501.058

[GRAPHIC] [TIFF OMITTED] 29501.059

[GRAPHIC] [TIFF OMITTED] 29501.060

[GRAPHIC] [TIFF OMITTED] 29501.061

[GRAPHIC] [TIFF OMITTED] 29501.062

[GRAPHIC] [TIFF OMITTED] 29501.063

[GRAPHIC] [TIFF OMITTED] 29501.064

[GRAPHIC] [TIFF OMITTED] 29501.065

[GRAPHIC] [TIFF OMITTED] 29501.066

[GRAPHIC] [TIFF OMITTED] 29501.067

[GRAPHIC] [TIFF OMITTED] 29501.068

[GRAPHIC] [TIFF OMITTED] 29501.069

[GRAPHIC] [TIFF OMITTED] 29501.070

[GRAPHIC] [TIFF OMITTED] 29501.071

[GRAPHIC] [TIFF OMITTED] 29501.072

[GRAPHIC] [TIFF OMITTED] 29501.073

[GRAPHIC] [TIFF OMITTED] 29501.074

                                 
