[House Hearing, 113 Congress]
[From the U.S. Government Publishing Office]
U.S. CENSUS BUREAU: ADDRESSING DATA COLLECTION VULNERABILITIES
=======================================================================
HEARING
before the
SUBCOMMITTEE ON FEDERAL WORKFORCE,
US POSTAL SERVICE AND THE CENSUS
of the
COMMITTEE ON OVERSIGHT
AND GOVERNMENT REFORM
HOUSE OF REPRESENTATIVES
ONE HUNDRED THIRTEENTH CONGRESS
SECOND SESSION
__________
SEPTEMBER 18, 2014
__________
Serial No. 113-146
__________
Printed for the use of the Committee on Oversight and Government Reform
Available via the World Wide Web: http://www.fdsys.gov
http://www.house.gov/reform
______
U.S. GOVERNMENT PRINTING OFFICE
91-127 PDF WASHINGTON : 2014
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800;
DC area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001
COMMITTEE ON OVERSIGHT AND GOVERNMENT REFORM
DARRELL E. ISSA, California, Chairman
JOHN L. MICA, Florida ELIJAH E. CUMMINGS, Maryland,
MICHAEL R. TURNER, Ohio Ranking Minority Member
JOHN J. DUNCAN, JR., Tennessee CAROLYN B. MALONEY, New York
PATRICK T. McHENRY, North Carolina ELEANOR HOLMES NORTON, District of
JIM JORDAN, Ohio Columbia
JASON CHAFFETZ, Utah JOHN F. TIERNEY, Massachusetts
TIM WALBERG, Michigan WM. LACY CLAY, Missouri
JAMES LANKFORD, Oklahoma STEPHEN F. LYNCH, Massachusetts
JUSTIN AMASH, Michigan JIM COOPER, Tennessee
PAUL A. GOSAR, Arizona GERALD E. CONNOLLY, Virginia
PATRICK MEEHAN, Pennsylvania JACKIE SPEIER, California
SCOTT DesJARLAIS, Tennessee MATTHEW A. CARTWRIGHT,
TREY GOWDY, South Carolina Pennsylvania
BLAKE FARENTHOLD, Texas TAMMY DUCKWORTH, Illinois
DOC HASTINGS, Washington ROBIN L. KELLY, Illinois
CYNTHIA M. LUMMIS, Wyoming DANNY K. DAVIS, Illinois
ROB WOODALL, Georgia TONY CARDENAS, California
THOMAS MASSIE, Kentucky STEVEN A. HORSFORD, Nevada
DOUG COLLINS, Georgia MICHELLE LUJAN GRISHAM, New Mexico
MARK MEADOWS, North Carolina Vacancy
KERRY L. BENTIVOLIO, Michigan
RON DeSANTIS, Florida
Lawrence J. Brady, Staff Director
John D. Cuaderes, Deputy Staff Director
Stephen Castor, General Counsel
Linda A. Good, Chief Clerk
David Rapallo, Minority Staff Director
Subcommittee on Federal Workforce, U.S. Postal Service and the Census
BLAKE FARENTHOLD, Texas, Chairman
TIM WALBERG, Michigan STEPHEN F. LYNCH, Massachusetts,
TREY GOWDY, South Carolina Ranking Minority Member
DOUG COLLINS, Georgia ELEANOR HOLMES NORTON, District of
RON DeSANTIS, Florida Columbia
WM. LACY CLAY, Missouri
C O N T E N T S
----------
Page
Hearing held on September 18, 2014............................... 1
WITNESSES
The Hon. Erik Paulsen, a Member of Congress, from the state of
Minnesota
Written Statement............................................ 6
The Hon. John H. Thompson, Director, U.S. Census Bureau
Oral Statement............................................... 11
Written Statement............................................ 13
The Hon. Todd Zinser, Inspector General, U.S. Department of
Commerce
Oral Statement............................................... 18
Written Statement............................................ 20
APPENDIX
The Hon. Elijah Cummings, Member of Congress from the State of
Maryland, Opening Statement.................................... 44
House Committee on Oversight and Government Reform, U.S. Congress
Joint Economic Committee Staff Report.......................... 49
U.S. CENSUS BUREAU: ADDRESSING DATA COLLECTION VULNERABILITIES
----------
Thursday, September 18, 2014,
House of Representatives,
Subcommittee on Federal Workforce, U.S. Postal
Service and The Census,
Committee on Oversight and Government Reform,
Washington, D.C.
The subcommittee met, pursuant to call, at 9:35 a.m., in
Room 2154, Rayburn House Office Building, Hon. Blake Farenthold
[chairman of the subcommittee] presiding.
Present: Representatives Farenthold, Lynch, and Clay.
Also Present: Representative Paulsen.
Staff Present: Alexa Armstrong, Majority Legislative
Assistant; John Cuaderes, Majority Deputy Staff Director;
Jessica L. Donlon, Majority Senior Counsel; Jeffrey Post,
Majority Senior Professional Staff Member; Jonathan J.
Skladany, Majority Deputy General Counsel; Sarah Vance,
Majority Assistant Clerk; Jaron Bourke, Minority Administrative
Director; Marianna Boyd, Minority Counsel; Aryele Bradford,
Minority Press Secretary; Juan McCullum, Minority Clerk; and
Mark Stephenson, Minority Director of Legislation.
Mr. Farenthold. The Subcommittee on the Federal Workforce,
U.S. Postal Service and Census will come to order.
I would like to begin this hearing by stating the Oversight
Committee mission statement. That is how we start out all of
our meetings.
We exist to secure two fundamental principles: first,
Americans have a right to know that the money Washington takes
from them is well spent and, second, Americans deserve an
efficient, effective Government that works for them. Our duty
on the Oversight and Government Reform Committee is to protect
these rights. Our solemn responsibility is to hold Government
accountable to taxpayers, because taxpayers have a right to
know what they get from their Government. We will work
tirelessly in partnership with citizen watchdogs to deliver the
facts to the American people and bring genuine reform to the
Federal bureaucracy. This is the mission of the Oversight and
Government Reform Committee.
At this point I will recognize myself for an opening
statement.
In addition to its namesake activity, the Census Bureau
conducts a number of monthly surveys for the Federal
Government. Perhaps the most important among them is the
Consumer Population Survey used to calculate the national
unemployment rate, one of our most important economic
indicators. The collection of this data is critical to both
Main Street and Wall Street, and ensuring its timeliness and
accuracy is the responsibility of the Census Bureau.
Unfortunately, in November of last year, we heard
allegations that senior officials in the Bureau's Philadelphia
Regional Office were instructing field workers to falsify
survey responses. After hearing these allegations, the
Oversight and Government Reform Committee and the Joint
Economic Committee launched a joint investigation. Our
investigation found that data falsification did occur, but,
fortunately, there was no evidence that this falsification was
pervasive or systemic.
I want to be clear on this. The committee found no evidence
that the falsification was on an order of magnitude close
enough to affect the national unemployment rate numbers. Nor,
thank goodness, was there any evidence of a systemic plot in
support of some grand political scheme.
However, we did find substantial problems in the Bureau's
data collection and review process, problems that unfortunately
create a very real incentive for managers to overlook or even
possibly encourage data falsification by field staff. The
Bureau needs to take swift corrective action and measures to
address these vulnerabilities.
Data integrity is mission-critical to the Bureau, the
Government, and America as a whole. Even the appearance of
impropriety or inappropriate activities or lingering questions
about the trustworthiness of Census Bureau data is
unacceptable. As the Nation's, if not the world's, preeminent
statistical agency, the Census Bureau's methods must be above
reproach.
The committee staff report released earlier today, in
partnership with the Joint Economic Committee's majority staff,
identified a number of flaws and provided recommendations that
would address these concerns. Chief among the findings was the
data review process does little to discourage data
falsification. Namely, supervisors who assign work are also
responsible for reviewing data quality, and they are paid on
the basis of completion, speed, and response rate of the
surveys that are worked on. This creates unacceptable incentive
for unscrupulous managers to ask field staff to falsify data
with the promise a manager will cover for the falsification. If
successful, this scheme would improve alleged response rates
and allow workers to complete surveys faster, though with
little basis in reality.
Under the current system, the people responsible for
maximizing response rates are the same folks responsible for
maintaining data quality. They are expected to do a job with
two conflicting objectives. That is not good policy and not
good management. Obviously, most Census workers and managers
are doing the right thing; they are fulfilling their commitment
to data quality. But we cannot expect every one of them to do
the right thing when policies and procedures incentivize doing
something else, or at least turning a blind eye to it.
Quality control needs to be outside the chain of command.
As our friends of the Inspector General's Office pointed out in
their report, the Bureau does things differently for the
Decennial Census. They have a separate unit responsible for
quality control. That is logical. That is good policy and that
is good management. It makes sense and that should be the
policy for all surveys.
Census Bureau practices also bring into question just how
committee the Bureau is to committing data falsification. Its
employee training barely makes mention of it. They still rely
on carbon paper-based forms to track suspected falsification. I
hadn't seen one of these for a while. Apparently they still
exist. They also have some issues with disparate computer
operating systems that don't interchange data well. The Census
Bureau needs to send a very clear message that it is serious
about the quality of its data and the integrity of the census.
It is my sincere hope that today Director Thompson will be
able to talk about the ways the Census Bureau is actively
taking steps to address data integrity and other concerns
raised by this subcommittee and the inspector general.
At this point, I yield to Mr. Lynch, the ranking member.
Mr. Farenthold. Actually, before we start, Mr. Lynch, if
you will indulge me.
Mr. Lynch. Sure.
Mr. Farenthold. I would like to ask unanimous consent that
our colleague from Minnesota, Mr. Paulsen, be allowed to
participate in this hearing.
Without objection, so ordered.
Mr. Lynch?
Mr. Lynch. Thank you. Thank you, Mr. Chairman. I would like
to thank you for holding this hearing to examine the Census
Bureau's data collection and quality assurance procedures for
the Current Population Survey.
I would also like to thank our witnesses for being here
today to discuss the findings and recommendations issued by the
Department of Commerce Office of the Inspector General and the
progress of the Census Bureau in implementing these
recommendations.
Over the past year, as you have noted, Mr. Chairman, there
have been a number of allegations made against employees within
the Census Bureau. A November 2013 New York Post article made
several of these allegations publicly. And as the chairman has
noted, these allegations were thoroughly investigated by the
Department of Commerce Office of the Inspector General and were
found to be without merit. I would note that the inspector
general, Todd Zinser, was appointed by President George W. Bush
in December of 2007 and is with us here today.
In particular, after reviewing over 3,000 pages of
documents and conducting more than 100 interviews of former and
current Census Bureau employees, the Inspector General's Office
issued its final investigative report in May 2014, which was
appropriately entitled ``Unsubstantiated Allegations that the
Philadelphia Regional Office Manipulated the Unemployment
Survey Leading Up to the 2012 Presidential Election to Cause a
Decrease in the National Unemployment Rate.''
This report, ``found no evidence that the management in the
Philadelphia Regional Office instructed staff to falsify data
at any time or for any reason.'' The inspector general also
found, ``no evidence of systemic data falsification in the
Philadelphia Regional Office.'' Again, another finding, they
found no evidence that the national unemployment rate was
manipulated by staff in the Philadelphia Regional Office in the
months leading up to the 2012 presidential election.
The inspector general's report further determined that in
order to manipulate the unemployment rate, ``it would have
taken at least 78 Census Bureau field representatives working
together in a coordinated way to report each and every
unemployed person included in that sample as employed or not in
the labor force during September 2012.'' The report adds that
this effort likely would have been detected by the Census
Bureau's quality assurance procedures.
The inspector general also found that the decline in the
unemployment rate was consistent with other indicators such as
the payroll estimates by Automatic Data Processing and Moody's
Analytics.
So, in any case, if anyone is keeping score, the New York
Post allegations were determined by the inspector general to be
100 percent false.
Moreover, the inspector general has also reviewed the
Bureau's data collection and quality assurance procedures, and
issued several recommendations to the Bureau to protect against
data falsification. I think it is important to recognize that
the Bureau agrees with these recommendations based on Director
Thompson's testimony. The Bureau has already adopted two of the
inspector general's recommendations and is currently working to
implement the other four recommendations, and I look forward to
hearing the details surrounding each of those recommendations.
Thank you, Mr. Chairman, and I yield back my time.
Mr. Farenthold. Thank you very much.
I believe Mr. Paulsen has an opening statement as well. You
are recognized for five minutes, sir.
Mr. Paulsen. Thank you, Chairman Farenthold. My comments
come in relation to the Joint Economic Committee's interest on
this issue, so I will address the comments to Chairman Issa,
Subcommittee Chairman Farenthold, Ranking Member Cummings, and
Subcommittee Ranking Member Lynch.
Thank you for inviting me to participate in today's hearing
on behalf for the Joint Economic Committee. The Joint Economic
Committee has a vital interest in the accuracy, relevancy, and
timeliness of U.S. economic statistics. The Census Bureau
produces many economic statistics, including monthly statistics
on U.S. international trade in goods and services,
manufacturing sales and inventory, and construction. The Census
Bureau also collects the data used to calculate other economic
statistics such as the gross domestic product by the Bureau of
Economic Analysis and the unemployment and labor force
participation rates by Bureau of Labor Statistics.
Congress uses these statistics to make decisions about
Federal spending and taxes. Both corporate executives and small
entrepreneurs use these statistics to decide whether to open a
new store, build a new factory, or hire more workers. Families
uses these statistics to decide where to buy a new house or how
to invest their retirement funds.
Since so much of economic decision-making relies on
statistics produced by the Census Bureau or based on data
collected by the Census Bureau, allegations that the underlying
data may have been falsified, and thus the economic statistics
based on such data may be inaccurate, are deeply troubling and
deserve congressional scrutiny. That is why the Joint Economic
Committee joined with the Committee on Oversight and Government
Reform to investigate the allegations published in The New York
Post on November 18th, 2013, that a Census Bureau employee may
have falsified responses to the Current Population Survey used
to calculate the unemployment rate, the labor force
participation rates, and other employment statistics prior to
the 2012 presidential election.
While this investigation did find serious structural
problems and systemic deficiencies with the Census Bureau's
data collection processes, especially with respect to the
Bureau's ability to detect data falsification, we found no
evidence that falsification occurred with the intent to
manipulate the U.S. economic statistics, nor was falsification
sufficiently widespread to question the validity of the
statistics generated from the Census Population Survey.
The staff report makes five important recommendations to
assure the integrity of data collection. Checks must be put
into place to prevent any conflict of interest between
achieving a high survey response rate and reporting incidences
of data falsification. Specifically, clear procedures should be
established for staff at any level to report potential
falsification, and the re-interview process, which helps to
identify incidences of potential falsification, should be
independent of the chain of command. While achieving a high
response rate is important, it should not trump the integrity
of the data collected. Implementation of these recommendations
would encourage Census employees to report any suspected
falsification and would help Census managers to detect and
correct any fraud.
Two other recommendations specifically address the accuracy
and effectiveness of the Bureau's record-keeping. First, though
some improvements have been made to case tracking systems, the
Bureau could do more to ensure that all notes and files
associated with a case are tracked. Second, the current paper
form used for reporting suspected falsification must be made
electronic and tracked such that employees suspected of
falsification receive a timely and just investigation into the
matter. These particular recommendations would serve to
increase transparency and accountability within the collection
process.
Then, finally, Census Bureau and the Department of Commerce
should indeed improve their responsiveness to congressional
oversight. The Census Bureau has subsequently taken some steps
to ensure data quality, but real deficiencies remain. I urge
the Census Bureau to implement all of them promptly, and I am
interested in learning of any current or future plans from the
Census to improve data collection and quality control
procedures. America does rely on economic statistics produced
from Census data. The accuracy of these statistics must be
beyond reproach.
I yield back, Mr. Chairman.
[Prepared statement of Mr. Paulsen follows:]
[GRAPHIC] [TIFF OMITTED]
Mr. Farenthold. Thank you very much.
Other members will have seven days to submit their opening
statements for the record.
We will now recognize our panel.
The Honorable John Thompson is Director of the United
States Census Bureau and the Honorable Todd Zinser is Inspector
General for the U.S. Department of Commerce.
Welcome, gentlemen.
Pursuant to committee rules, all witnesses are to be sworn
before they testify. Would you all please rise and raise your
right hand?
Do you solemnly swear or affirm that the testimony you are
about to give will be the truth, the whole truth, and nothing
but the truth?
[Witnesses respond in the affirmative.]
Mr. Farenthold. Let the record reflect both witnesses
answered in the affirmative.
Thank you all. You may be seated.
In order to allow time for discussion and questions, we ask
that you hold your testimony to as close to five minutes as
possible. Your entire written statements will be made part of
the record. In order to abide by the five second rule, you all
have timers in front of you. The green light means you are good
to go; the yellow light means about a minute left; the red
light means stop as soon as you can get to a reasonable place
to do so.
So we will start with you, Mr. Thompson. You are recognized
for five minutes.
Mr. Farenthold.
WITNESS STATEMENTS
STATEMENT OF THE HONORABLE JOHN H. THOMPSON
Mr. Thompson. Good afternoon. On behalf of the U.S. Census
Bureau, thank you, Chairman Farenthold, Ranking Member Lynch,
and the subcommittee for inviting me today. I appreciate your
ongoing support of the Census Bureau and I am pleased to be
able to tell you about our data collection and data integrity
efforts.
In my 27 years as a civil servant at the Census Bureau and
in the year since I was appointed Director, I have come to be
extremely proud of our agency and its employees. They are
dedicated, qualified, and mission-focused public servants who
provide an invaluable service to the American people. Thanks to
them, the Census Bureau is one of the Federal Government's
foremost sources of quality statistics, and I am extremely
proud to lead it.
The Census Bureau emphasizes integrity in every one of our
data collection efforts. This is true of perhaps our best known
activity, the Decennial Census; it is true of the Economic
Census, which we conduct every five years; and of the American
Community Survey, the Nation's premier source of community and
neighborhood level data. It is also true of the Current
Population Survey, which I will discuss today.
The Current Population Survey, or CPS, is administered by
Census Bureau field representatives. The typical field
representative works part-time and earns about $15 an hour.
Their average age is 57, and many of them work for the Census
Bureau to supplement their retirement or other income. They are
members of their local communities and they are the face of the
Census Bureau to every home they visit, and, as such, we hold
them to high standards of performance, professionalism, and
courtesy.
Recently, the Department of Commerce's Office of the
Inspector General, or OIG, investigated an alleged data
falsification in the CPS. This matter was immediately referred
to the OIG. After months of investigation and more than 100
interviews with current and former staff, and as you will hear
shortly, the OIG concluded that these claims were
unsubstantiated. They found no evidence that the Census Bureau
management had instructed field representatives to falsify or
manipulate data.
As part of their review, the OIG also evaluated the
procedures that we use to detect falsification and made six
recommendations for improvement. The Census Bureau
wholeheartedly agrees that these suggestions will further
strengthen our data integrity. We have already put three of the
recommendations into practice and are implementing the other
three.
For example, we now immediately suspend work assignments to
field representatives who are suspected of falsifying data
unless and until they are cleared by an internal investigation.
We have implemented the Unified Tracking System, or UTS. It
provides a view of near real-time cost, progress, and response
data, consolidating data from other production systems and over
time and across surveys. This is a powerful tool for monitoring
employee performance and detecting anomalies that may indicate
falsification.
In addition to the UTS, we maintain detailed keystroke data
which can also be used to investigate falsification. We
generate UTS reports that field managers use to monitor the
quality of our interviewers' work. We have updated and will
continue to update our training materials, and we are
automating the process for reporting falsification, known as
the Form 11-163.
We have improved how we monitor and limit field
representatives' workloads to avoid circumstances that might
encourage shortcuts such as data falsification. We are
improving our quality control process so that data collection
re-interviews are now independent from the chain of command,
and we are on our way to centralizing these efforts. Finally,
we are sending regular reports to all of our survey sponsors,
such as the Bureau of Labor Statistics in the case of the CPS.
I also want to assure the members of the subcommittee that
all field representatives receive guidance on reporting
improprieties, including data falsification. New employees
complete a training that includes how to report fraud, waste,
and abuse to the Office of the Inspector General and also
receive an administrative manual that includes a specific
section with specific instructions about reporting fraud,
waste, and abuse to the Office of the Inspector General, and
this manual also includes the hotline number and a link to the
Web site.
At the Census Bureau, we welcome scrutiny and oversight of
our work. I sincerely appreciate the OIG's suggestions. They
are a clear, practical roadmap for improvement and they will
help the Census Bureau build on its already excellent work.
Thank you for this opportunity to address you. I look
forward to your questions.
[Prepared statement of Mr. Thompson follows:]
[GRAPHIC] [TIFF OMITTED]
Mr. Farenthold. Thank you very much, Mr. Thompson.
Inspector General Zinser, you are up.
STATEMENT OF THE HONORABLE TODD ZINSER
Mr. Zinser. Thank you, Mr. Farenthold, Ranking Member
Lynch, Mr. Clay, and members of the subcommittee. We appreciate
the opportunity to testify today on vulnerabilities on Census
data collection and quality assurance processes.
As you and Director Thompson have noted, one of the many
surveys the Census Bureau conducts throughout the year is the
monthly Current Population Survey, also referred to as CPS.
Earlier this year, on May 1st, my office publicly issued a
report of investigation concerning allegations that the Census
Bureau's Philadelphia Regional Office engaged in widespread
falsification of CPS surveys, including the manipulation of the
CPS unemployment rate in the months leading up to the 2012
presidential election. My written statement summarizes the
allegations concerning the Philadelphia Regional Office and the
results of our investigation, observations we made related to
Census Bureau falsification policies, and recommendations
stemming from our investigation.
The key allegations we investigated concerning the
Philadelphia Office included: number one, were Census field
representatives instructed by their supervisors to falsify
data; two, did members of management alter completed surveys to
manipulate data; and, three, did any alleged data falsification
of the CPS in August and September of 2012 have a measurable
impact on the unemployment rate leading up to the 2012
presidential election.
We conducted over 100 interviews of current and former
Census Bureau employees in the Philadelphia Regional Office,
Headquarters, and other regional offices. We reviewed training
materials, interviewing procedures, quality control processes,
and performance assessments. We also conducted extensive
analysis of Census CPS data and BLS employment statistics data,
as well as other data relevant to our investigation.
While our investigation did not substantiate these
allegations, we did identify several vulnerabilities with
respect to the Census Bureau policies and processes and made a
series of recommendations. We made the following four
observations: one, survey supervisors do not consistently use
the tools available to them for detecting and preventing data
falsification; two, field representatives suspected of
falsifying data are sometimes allowed to continue conducting
surveys while the matter is under investigation, there is not a
consistent practice across offices; three, the regional office
quality assurance process creates the potential for a conflict
of interest because the same supervisors who manage staff and
could direct the falsification of survey data are responsible
for reporting instances when their staff falsifies data; and,
four, CPS procedural manuals and training materials are
outdated, inconsistent, and do not discuss the serious
consequences of falsification.
We made the following six recommendations: one, implement a
formal policy that prohibits employees suspected of
falsification from collecting survey data during the
investigative process; two, update procedural manuals and
training materials to inform field representatives about the
consequences of committing falsification; three, implement an
independent quality assurance process for all survey operations
similar to the one used during Decennial Census operations;
four, ensure that all survey supervisors are properly using all
available tools to safeguard against falsification of survey
data; five, implement internal controls to effectively monitor
and limit field representative workloads in order to reduce the
risk of falsification; and, six, implement a reporting
mechanism for confirmed data falsifications to those
organizations that sponsor Census Bureau surveys.
As noted by Director Thompson, the Census Bureau agreed
with our findings and is in the process of implementing our
recommendations.
Thank you again for the opportunity to testify, and I would
be happy to answer any questions you or other members of the
subcommittee may have.
[Prepared statement of Mr. Zinser follows:]
[GRAPHIC] [TIFF OMITTED]
Mr. Farenthold. Thank you very much. You mentioned your
investigative report. I have it here. I would like to enter it
into the record. Without objection, so ordered.
This report can be found at: http://www.cig.doc.gov/
OIGPublications/14-0073.pdf
Mr. Farenthold. It is good to actually be holding a hearing
in this committee where we find out the situation isn't as bad
as we initially thought. Unfortunately, in some of our other
hearings it turns out the situation is worse than we initially
thought. So this is actually a happy occasion for me to be
chairing this subcommittee hearing where, yes, there are
problems, but it is not as bad as we thought.
Mr. Thompson, I would like to start with you for a second.
I think a lot of the issues that we saw in the VA and I think
we see in some areas of the Government is what the culture is
within the organization. You all testified that you do have
training materials early on for your new recruits to talk about
the issues with data falsification and how to report it.
Typically, how long is a new hire trained?
Mr. Thompson. Congressman, I will have to get back to you
on that.
Mr. Farenthold. My concern is, is the data falsification
five minutes in a two-day training or is it something that is
pervasive throughout the training? That is what I am trying to
get at.
Mr. Thompson. Okay, so we have revised our training to
emphasize data integrity. In fact, we are implementing a form
that each interviewer will sign, and they will re-sign it every
year, that explains to them the importance of data integrity
and the dangers in falsifying data.
Mr. Farenthold. And what are the consequences to an
employee caught falsifying data?
Mr. Thompson. It is a fireable offense.
Mr. Farenthold. Okay. That is good to hear.
Now, you talk about your UTS system, your Unified Tracking
System. It is my understanding that that tracks who assigns
what and who completes what cases. But there are actually three
other systems, the Blaze File, which is the actual survey
response, and the Trace File, which is a time-stamped keystroke
log. These are three separate systems, right? How do they talk
to each other, or do they?
Mr. Thompson. That is one of the next steps in our process,
is to put together a method for making these systems talk to
each other. However, I have to say that we are continually
revising the way in which we analyze the data from our Unified
Tracking System to provide more and more useful information to
our field managers.
Mr. Farenthold. Great. It is also my understanding that
regional offices are responsible for both data collection and
quality control, and they often have conflicting objectives.
Most Census Bureau employees are evaluated, at least in part,
on survey response rates. The Census Bureau can minimize a
potential for conflict of interest by separating the interview
from the regional chain of command, thus allowing quality
control to function independently. It is my understanding that
you all have started, in July, a pilot program where there is a
re-interview process that comes out of your Jeffersonville
connect center, rather than the regional offices. Do you think
this initiative has ben successful so far? Can you give us an
idea how it is working?
Mr. Thompson. Certainly. We have already moved two of our
regional offices to the Jeffersonville facility where a totally
independent staff will do the telephone interviewing. And by
July, I am sorry, by April 15, by April, this coming April, we
will have moved all the regional offices to our Jeffersonville
facility.
Mr. Farenthold. And have you run into any problems or is it
working out pretty well?
Mr. Thompson. It has been working very well.
Mr. Farenthold. Great. Great. One of the committee's
witnesses reports that they have faced retaliation after she
cooperated with the committee's investigation. Can we have your
commitment that you will work with our staff to ensure that
Bureau employees are not retaliated for whistleblowing and
cooperating with Congress?
Mr. Thompson. Certainly, Congressman.
Mr. Farenthold. All right. We had a hearing the other day
in the full committee on the difficulties that some
whistleblowers face, and this committee is almost unanimous in
believing that whistleblowers need the highest level of
protection, that they are our partners in being watchdogs.
With that, I will go on and let Mr. Lynch have his
questions, and I am going to check to make sure we have gotten
everything I need to get covered as well. We will go to Mr.
Lynch now.
Mr. Lynch. All right, thank you, Mr. Chairman. I just want
to go back. This was basically the headline just prior to the
election, where it says the Census faked the 2012 election jobs
report, the implication being that Census employees falsified
the records, and their goal was to manipulate the unemployment
numbers so that President Obama would get re-elected. That was
basically the conspiracy that they alleged. So I want to just
drill down on that a little bit. I think that is very
unfortunate and gives freedom of the press a bad name.
Mr. Zinser, let's just go back over your report a little
bit. I just want to straighten the record out. So after
reviewing over 3,000 pages of documents and conducting more
than 100 interviews of former and current Census Bureau
employees, your office found, and this is a quote from your
report, ``no evidence that management in the Philadelphia
Regional Office instructed staff to falsify data at any time
for any reason.`` Is that correct?
Mr. Zinser. That is correct, sir.
Mr. Lynch. Okay. And can you explain to me briefly what led
you to this conclusion?
Mr. Zinser. Well, we asked employees about two main
allegations: we asked them whether they were ever instructed to
falsify data, and none of the employees that we interviewed
said that they had been so instructed.
Mr. Lynch. Now, most of the employees, are they newly hired
since President Obama took office, or have they been there a
while?
Mr. Zinser. I don't know the exact answer to that, but they
were a cross-section of employees picked at random throughout
the region.
Mr. Lynch. Okay.
Mr. Zinser. And there was no other selective factor there
in terms of how long they had been on staff.
Mr. Lynch. Okay, good enough.
Mr. Zinser. The other issue that we asked them is whether
they had any concerns about their completed surveys being
altered by their supervisors, and none indicated that they had
such concerns.
Mr. Lynch. Okay. And then you checked with the managers to
see if there was any changes implemented by them?
Mr. Zinser. Yes, sir. There were three managers who were
essentially subjects of the investigation. We did interview
each of them; they cooperated with our investigation. They
denied any efforts to instruct employees to change surveys.
They denied that they had ever changed or altered surveys
themselves.
Mr. Lynch. Okay. You also indicated in your report,
contrary to The New York Post allegations, that there was no
evidence that any of this data was used to manipulate the
unemployment rate leading up to the election.
Mr. Zinser. Yes, sir. We went into the investigation
understanding that theoretically it could be possible to
manipulate the unemployment rate.
Mr. Lynch. Let's talk about that a little bit. What would
have to happen in order for that number to change? I understand
it is broad-based data and the indicia of unemployment is sort
of a blended number. What would be required for manipulation to
actually happen?
Mr. Zinser. Well, actually, the Census Bureau did an
analysis and then our staff evaluated that, and basically the
average workload of a CPS field representative or field
representative that works on the CPS, the average workload is
30 cases. So if the unemployment rate is 8.1 or 8 percent, that
means, of those 30 cases, 2.4 of those cases you are going to
find somebody who is unemployed, 2.4 out of the 30. So in order
to change the unemployment rate by just .1 percent, it would
take 26 field representatives changing all of their unemployed
cases to employed. And then in order to get a .3 percent
decline in the unemployment rate, you would multiply the 26
times 3 and you would come up with 78 field representatives.
Mr. Lynch. Okay. And that is what we had there, right? At
that time all the indicators were the same, right?
Mr. Zinser. Yes, sir.
Mr. Lynch. Well, they were all consistent. I won't say they
were the same, but they were all consistent that there was a
drop in unemployment.
Mr. Zinser. Yes. There was a drop from 8.1 percent to 7.8
percent.
Mr. Lynch. Okay. All right. I see my time has expired and I
will yield back.
Thank you.
Mr. Farenthold. Thank you very much.
I have a couple more questions I will do, then we will go
back to Mr. Clay when he gets back.
Mr. Thompson, is there any computer management or
electronic falsification screening? For example, in the
Consumer Population Survey, households are in for four months,
out for eight, and back in for another four. Does any screening
occur automatically to identify inconsistencies over the period
of the households in the survey? Do you want to talk about any
other automated data integrity systems you have in place?
Mr. Thompson. The systematic processes that we use right
now are preparing reports from our Unified Tracking System that
are designed to identify anomalies that we can look at. We are
working aggressively to put together more of a data analytics
team that will then allow for really a more systematic way of
doing this.
Mr. Farenthold. Great.
Mr. Zinser, we have heard testimony in other hearings about
inspectors general having problems getting their job done and
being blocked. Your testimony tended to indicate that you had
pretty good cooperation. Did you run into any roadblocks in
your investigation or feel like there were folks not
cooperating with your investigation?
Mr. Zinser. No, sir, we didn't have problems with
cooperation. Early on, and I think the staff report kind of
points out that the General Counsel's Office of the department
did get involved in the investigation and we did make sure that
our investigation was completely independent of anything that
office was doing with respect to the committee's oversight.
Mr. Farenthold. Mr. Thompson, we didn't have quite as good
of an experience, or at least the committee staff reported to
me that despite the Census Bureau having its own legislative
affairs staff and lawyers that ordinarily handle the Bureau's
engagements with Congress and this committee, in this case the
Commerce Department stepped in and assigned its own staff to
manage the congressional investigation. Their posture toward
the committee's investigation, according to committee staff,
was confrontational from the onset. The pace of the committee's
investigation was slowed because Commerce Department officials
slow-rolled document productions and interfered with witness
interviews.
Did you or your predecessor request the Department of
Commerce to manage the response to the committee's
investigation?
Mr. Thompson. Congressman, we always work closely with our
colleagues at the Department of Commerce. In fact, our
attorneys administratively do report to the Office of General
counsel to ensure consistency. So we basically followed the
procedures that were in place for responding to the
subcommittee.
Mr. Farenthold. So you are saying it is standard practice
for the Department of Commerce staff to take over management of
these types of inquiries? Was it a takeover or was it just
cooperation?
Mr. Thompson. It was cooperation, it wasn't a takeover. We
have a very good relationship with the Department of Commerce.
Mr. Farenthold. And did the Department of Commerce tell you
why they jumped in on this one more than they had in the past,
at least in dealing with us?
Mr. Thompson. I didn't have conversations with them about
anything like that.
Mr. Farenthold. All right. And the Department didn't
express any concerns with your legislative affairs folks and
them handling it, did they?
Mr. Thompson. No.
Mr. Farenthold. All right. Well, I was just kind of curious
why they jumped in on this one.
All right, I see Mr. Clay is available now. We will give
him his first round of questions and then give Mr. Lynch his
second, if he has some more.
Mr. Clay. Thank you, Mr. Chairman.
I want to thank Director Thompson and Mr. Zinser for being
here today.
However, Mr. Chairman, I am just kind of bewildered on how
and why we are at this point. As the ranking member stated
earlier, I guess this inquiry started from a November 18, 2013
New York Post article, and it seems to me, it appears as if
quality control may be needed at The New York Post. Clearly,
the information contained in the article was not reliable, and
I see that some of the history of how we got to this point was
in response to that Post article.
On November the 19th, 2013, Chairman Issa, you, Mr.
Chairman, and Kevin Brady sent a letter to Director Thompson
seeking documents related to possible employee data
falsification in the CPS, and in response to that request the
Bureau has produced over 4800 pages of documents and the
committee has conducted six transcribed interviews with current
and former Census Bureau employees.
I did hear you say earlier in this hearing that this is
good news, so I am just curious as to how we got to this point
off of an unreliable New York Post article.
Mr. Farenthold. If the gentleman would yield.
Mr. Clay. I yield.
Mr. Farenthold. As I did point out, we found that the
problem was not nearly as bad as The New York Post suggested,
but I think we are having this hearing to ensure that some of
the reforms from the inspector general's report and the result
of this investigation are going to be implemented. It is our
duty to conduct rigorous oversight, and, as I said at the
onset, I am glad that it didn't turn out to be as bad as we all
thought it did. It is actually pleasurable for me to have a
good news hearing, and I do think, for the most part, this is a
good news hearing.
Mr. Clay. And that is a good thing, because I think the
people at the Census Bureau work awfully hard to get it right,
to supply this Government and the American people with the data
necessary to make good decisions.
Over the past year we have heard a number of allegations
against the Census Bureau as it relates to a potential data
falsification scandal in the Philadelphia Regional Office.
However, let's set the record straight. There was no scandal
here. The IG conducted what I think we would all agree was a
very thorough investigation and the allegations were found to
be without merit. And after reviewing over 3,000 pages of
documents and conducting more than 100 interviews of former and
current Census Bureau employees, the IG found no evidence that
management in the Philadelphia Regional Office instructed staff
to falsify data at any time, for any reason.
Is that correct, Mr. Zinser?
Mr. Zinser. Yes, sir, it is.
Mr. Clay. And can you please explain what led you to that
conclusion?
Mr. Zinser. Yes, sir. We followed our investigative plan,
and that included interviewing current and former employees, it
included interviewing the managers who the complainants had
pointed to as instructing them to do this. The managers
actually submitted to polygraph examinations, as well. We also
looked at the actual computer files with respect to an analysis
of whether any changes had been made to any of the files, and
we did not find any evidence of any alteration of the survey
results.
Mr. Clay. And in your investigation did you find any
evidence that management changed survey responses or covered up
falsification?
Mr. Zinser. No, sir, we didn't. As I mentioned, we did go
into the computer systems and looked at the trace files, which
are actually audit trails, to see the activity of the
supervisors, and we did not find any occasions where results
had been altered.
Mr. Clay. I thank you for your response.
I see my time has expired. Thank you, Mr. Chairman.
Mr. Farenthold. Thank you, Mr. Clay.
Mr. Lynch?
Mr. Lynch. Thank you, Mr. Chairman.
It is unfortunate, as my friend from Missouri has pointed
out, that a news report would lead us to this point, but
separate and apart, I think, as the chairman has noted, this is
a good news hearing, and I think it was important to lay this
all out because there has been a steady drumbeat of criticism
for Government employees and agencies. Government employees, in
the last several years, have endured furloughs; they had a
three-year pay freeze; the pension contributions for all these
Federal employees have been increased, so they are paying more
for their pensions and getting less at the end of it. But it is
very important to make sure that people understand that these
reports were baseless. So the good reputation of the Census
employees involved in these cases were sullied because of the
allegations made.
We have done a thorough investigation, congressionally and
through our inspectors general, and the claims against these
employees were completely baseless. So I apologize to those
Census employees that had to endure this and I would just
caution that sometimes the environment in which Federal
employees have to work is hostile to their morale and also to
the performance of their duties.
So I guess my confidence in the employees of the Census is
reaffirmed. I think they have performed exceedingly well their
responsibilities, even during the time of these allegations.
They have done a good job. Our inspectors general have done a
commendable job, as well, to be thorough in their analysis, and
I think that the recommendations are the best thing to come out
of this, other than absolving the Census Bureau of any blame
here. I think the recommendations are sound, I think it will
end up allowing us to really have greater confidence than we
had before and eliminate any possibility that there might be
manipulation in the future.
So I appreciate it, Mr. Chairman. I thank you for your
indulgence and thank you for holding this hearing.
Mr. Farenthold. And thank you, Mr. Lynch.
Mr. Clay, did you have any more questions?
Mr. Clay. I do not, Mr. Chairman.
Mr. Farenthold. All right.
While I agree almost completely with Mr. Lynch, I do want
to point out that this investigation was not a waste of time;
we got some good results from it. We learned that there are
some potential problems that need to be addressed in training,
in implementing management issues. I think we made a good step
by having a completely different division doing the call-backs
for data security. We are well on the way to improving our
computer system, so I certainly don't think it is a waste of
time.
Mr. Lynch. Mr. Chairman, briefly. I neglected to offer the
opening statement of Mr. Issa, our full committee ranking
member, and I would like to offer that for the record.
Mr. Farenthold. Absolutely.
Mr. Lynch. Okay, thank you.
Mr. Farenthold. The ranking member.
Mr. Lynch. Mr. Cummings, right. What did I say?
[Laughter.]
Mr. Lynch. I wish he was the ranking member.
Mr. Farenthold. Wishful thinking on your side, I guess.
We are happy to have Mr. Cummings as the ranking member.
So, without objection, this will be entered into the record.
Mr. Farenthold. I would like to thank our witnesses for
taking time to be with us and for your hard work for the
hardworking American taxpayers.
We are adjourned.
[Whereupon, at 3:03 p.m., the subcommittee was adjourned.]
APPENDIX
----------
Material Submitted for the Hearing Record
[GRAPHIC] [TIFF OMITTED]