[House Hearing, 118 Congress]
[From the U.S. Government Publishing Office]
VA INFORMATION TECHNOLOGY
CONTRACTING: CHALLENGES
IN CONSOLIDATION OF COMPETITION
AND CONFLICT OF INTEREST
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
OF THE
COMMITTEE ON VETERANS' AFFAIRS
U.S. HOUSE OF REPRESENTATIVES
ONE HUNDRED EIGHTEENTH CONGRESS
FIRST SESSION
__________
WEDNESDAY, MAY 24, 2023
__________
Serial No. 118-17
__________
Printed for the use of the Committee on Veterans' Affairs
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available via http://govinfo.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
52-732 WASHINGTON : 2024
COMMITTEE ON VETERANS' AFFAIRS
MIKE BOST, Illinois, Chairman
AUMUA AMATA COLEMAN RADEWAGEN, MARK TAKANO, California, Ranking
American Samoa, Vice-Chairwoman Member
JACK BERGMAN, Michigan JULIA BROWNLEY, California
NANCY MACE, South Carolina MIKE LEVIN, California
MATTHEW M. ROSENDALE, SR., Montana CHRIS PAPPAS, New Hampshire
MARIANNETTE MILLER-MEEKS, Iowa FRANK J. MRVAN, Indiana
GREGORY F. MURPHY, North Carolina SHEILA CHERFILUS-MCCORMICK,
C. SCOTT FRANKLIN, Florida Florida
DERRICK VAN ORDEN, Wisconsin CHRISTOPHER R. DELUZIO,
MORGAN LUTTRELL, Texas Pennsylvania
JUAN CISCOMANI, Arizona MORGAN MCGARVEY, Kentucky
ELIJAH CRANE, Arizona DELIA C. RAMIREZ, Illinois
KEITH SELF, Texas GREG LANDSMAN, Ohio
JENNIFER A. KIGGANS, Virginia NIKKI BUDZINSKI, Illinois
Jon Clark, Staff Director
Matt Reel, Democratic Staff Director
SUBCOMMITTEE ON TECHNOLOGY MODERNIZATION
MATTHEW M. ROSENDALE, SR., Montana, Chairman
NANCY MACE, South Carolina SHEILA CHERFILUS-MCCORMICK,
KEITH SELF, Texas Florida, Ranking Member
GREG LANDSMAN, Ohio
Pursuant to clause 2(e)(4) of Rule XI of the Rules of the House, public
hearing records of the Committee on Veterans' Affairs are also
published in electronic form. The printed hearing record remains the
official version. Because electronic submissions are used to prepare
both printed and electronic versions of the hearing record, the process
of converting between various electronic formats may introduce
unintentional errors or omissions. Such occurrences are inherent in the
current publication process and should diminish as the process is
further refined.
C O N T E N T S
----------
WEDNESDAY, MAY 24, 2023
Page
OPENING STATEMENTS
The Honorable Matthew M. Rosendale, Sr., Chairman................ 1
The Honorable Sheila Cherfilus-McCormick, Ranking Member......... 2
WITNESSES
Ms. Shelby Oakley, Director, Contracting and National Security
Acquisitions, U.S. Government Accountability Office............ 3
Ms. Hana Schank, Senior Advisor, New America..................... 4
APPENDIX
Prepared Statements Of Witnesses
Ms. Shelby Oakley Prepared Statement............................. 29
Ms. Hana America Prepared Statement.............................. 43
VA INFORMATION TECHNOLOGY
CONTRACTING: CHALLENGES
IN CONSOLIDATION OF COMPETITION
AND CONFLICT OF INTEREST
----------
WEDNESDAY, MAY 24, 2023
U.S. House of Representatives,
Subcommittee on Technology Modernization,
Committee on Veterans' Affairs,
Washington, D.C.
The subcommittee met, pursuant to notice, at 8:05 a.m., in
room 360, Cannon House Office Building, Hon. Matthew M.
Rosendale, Sr.(chairman of the subcommittee) presiding.
Present: Representatives Rosendale, Mace, Self, and
Cherfilus-McCormick.
OPENING STATEMENT OF MATTHEW M. ROSENDALE, CHAIRMAN
Mr. Rosendale. Good morning. The subcommittee will come to
order.
We are here today to discuss the underlying trends that
make the Department of Veterans Affairs' information technology
contracting so fraught with problems. To put it simply, the
market is concentrating, spending is going up, and the number
of companies receiving the contracts is going down.
The VA is not unique in this respect, but as one of the
largest IT buyers in the Federal Government, it is a striking
example of how the system operates. The system is marked by
bureaucracy, regulatory capture, the revolving door, and
impunity.
What you will not find in this system is much
accountability. This is why IT projects meant to modernize how
services are delivered to our veterans stumble again and again.
They are rewarded with the change orders that enrich vendors,
while squandering valuable and limited resources, because
contracts are not specific enough to hold vendors accountable.
The same companies cycle through the agency year after
year. Not only that, they merge and consolidate. That is
becoming increasingly common. If they fail at one project, they
simply reappear in another office with a different project.
They compete with each other, but they tend to operate within a
closed loop. They rarely face competition from outside the
government sector, which is where the innovation is found that
most people associate with technology.
To be sure, once in a while, a disrupter manages to find
its way through the jungle of paperwork and achieve impressive
results, many times only to be acquired by a larger firm, which
eliminates competition and perpetuates the problems.
We should look around at the barriers that discourage new
entrants or participants. One is the sheer complexity of
government contracting. Another is the inability or
unwillingness of agency officials to understand what the
contractors are doing and to hold them to account. A third is
the widespread practice of putting enormous multiyear contract
vehicles in place that function as an approved bidders list.
This enriches a well-placed view and shuts everyone else out.
The surefire consequence of this concentrating marketplace
is a widespread organizational conflicts of interest. When the
same few companies work in and even make decisions for all of
the major organizations within the VA, there is no way to
maintain fair dealing. Either the supplier base has to expand
or some of these companies have to be barred from holding
certain future contracts.
That is not my opinion. That is what the procurement laws
say. It is the VA's responsibility to police the system, and it
is this committee's oversight responsibility to make sure that
that happens.
American veterans and taxpayers are rightfully outraged
when they read about billions of dollars being paid out year
after year with nothing to show for it, to an industry that
always seems to be recession proof.
I appreciate our witnesses joining us today to help us
better understand the situation.
With that, I will yield to the ranking member,
Representative Cherfilus-McCormick, for her opening statement.
OPENING STATEMENT OF SHEILA CHERFILUS-MCCORMICK, RANKING MEMBER
Ms. Cherfilus-McCormick. Thank you so much, Chairman
Rosendale.
The Technology Modernization Subcommittee conducted
extensive oversight of VA modernization and IT contracting last
Congress. A common thread identified was a fundamental lack of
planning, budgeting, and adherence to contracting best
practices by VA and its contracting centers.
The Government Accountability Office (GAO) has done an
excellent job cataloging what has gone wrong in the past. VA
acquisition management has been on the high-risk list since
2019.
I will be interested to hear from our witness today her
opinion on how much progress has been made since their
inclusion on that list.
GAO has also cataloged issues with competition for IT
contracts, while VA's annual IT obligations have increased from
4.2 billion in 2017 to 6.5 billion in 2021 [sic]. The number of
companies receiving those awards has decreased.
A current question last Congress but continues into this
Congress is: Where are the checks and balances in VA's
oversight of major IT acquisition? I still have not heard a
good answer to this question. Whether it is benefits, financial
management, supply chain, or healthcare record modernization,
there is a lot of talk about following a framework and
accountability, but very little evidence.
As a result, I have cosponsored Ranking Member Takano's IT
Modernization Improvement Act. This will require the VA to
contract for independent verification and validation for these
major IT programs. Veterans and VA employees should not have to
continue to wait any longer for modern IT solutions.
Veterans and employees have been directly impacted by
previous acquisition failures and poor contract management. It
is my hope this Congress that we can work across the aisles to
permanently fix some of these issues and start getting what we
pay for.
Thank you again, Mr. Chairman. I look forward to the
hearing and listening to our witnesses.
Mr. Rosendale. Thank you, Ranking Member Cherfilus-
McCormick.
I will now introduce the witnesses on our first and only
panel.
First, from the Government Accountability Office, we have
Ms. Shelby Oakley, the director of Contracting and National
Security Acquisitions. We also have Ms. Hana Schank, a senior
adviser at New America.
I ask the witnesses to please stand and raise your right
hands.
[Witnesses sworn.]
Mr. Rosendale. Let the record reflect that all witnesses
have answered in the affirmative.
Ms. Oakley, you are now recognized for 5 minutes to deliver
your opening statement.
STATEMENT OF SHELBY OAKLEY
Ms. Oakley. Chairman Rosendale, Ranking Member Cherfilus-
McCormick, and members of the subcommittee, thank you for
having me here this morning to discuss our work on VA IT
contracting.
VA relies on information technology systems to provide
critical benefits and services to millions of veterans and
their families. As a result, VA has one of the highest
obligations for IT procurement in the Federal Government, over
$25 billion from 2017 to 2021.
Our work over the years has found that VA has faced
challenges in its IT efforts and, more broadly, in its ability
to successfully execute acquisitions. We have made about 100
recommendations in these areas. While VA has implemented many
of them, our work continues to show room for improvement.
Today, I will briefly discuss two recent reports, one on
trends in VA IT contracting, and another on VA's efforts to
oversee its IT procurements.
Last year, we reported that from 2017 to 2021, VA's total
IT contract obligations increased by more than half. Large
technology modernization efforts, like the electronic health
records program and the COVID-19 pandemic response, drove much
of this growth. At the same time, the number of contractors
receiving awards dropped by more than 25 percent. This decrease
is likely due to VA awarding 50 percent fewer new contracts
during this same time period.
VA officials told us that the Department has made a
concerted effort to consolidate customer IT requirements. This
kind of approach, called category management, resulted in
fewer, though larger, IT procurements. Our prior work showed
governmentwide contracting for common products and services
through initiatives like category management mirrored this
trend.
The overall trends were also reflected in VA's IT
contracting with service-disabled, veteran-owned small
businesses. While IT obligations to these businesses increased
by almost 30 percent, the number of such contractors fell by 10
percent from 2017 to 2021.
We also found that VA's IT obligations have been
increasingly concentrated with a small group of contractors.
Specifically, over half of VA's IT obligations in 2021 went to
only 10 contractors, up from 45 percent in 2017. More broadly,
30 contractors received about 75 percent of VA's IT obligations
over this same time period.
VA is relying more on two particular contracts--its
Transformation Twenty-One Total Technology Next Generation
contract, known as T4NG, and a governmentwide contract vehicle
managed by National Aeronautics and Space Administration (NASA)
called Solutions for Enterprise-Wide Procurement (SEWP). By
2021, these two contracts accounted for over half of VA's IT
contract obligations, with VA competing orders for 90 percent
of obligations amongst the vendors on these contracts.
With the increase in VA's IT contract obligations in recent
years, oversight is essential. However, in March, we reported
that many of VA's IT contracts were not being reviewed or
approved as required by the agency's chief information officer,
or CIO.
This review is a key aspect of Federal statute aimed at
improving IT investments by ensuring better oversight of IT
contracting. Specifically, we found that VA awarded almost
12,000 new IT contract actions between March 2018 and September
2021. VA did not provide evidence of CIO approval for over
4,500, or 35 percent of these contract actions. In particular,
we found that the IT contract actions executed by non-IT-
focused contracting shops represented the majority of
unapproved actions we sampled.
We and the inspector general have previously reported on
these same challenges, but they have persisted. As such, we
recommended that VA implement an automated reminder for
contracting officers to obtain this CIO approval for IT
procurements. This should ensure that VA's CIO has the
opportunity to provide input on current and planned IT
acquisitions and help avoid awarding contracts that are poorly
conceived or duplicative.
Mr. Chairman, Ranking Member Cherfilus-McCormick, and
members of the subcommittee, this concludes my oral statement.
Thank you for having me here, and I am happy to answer any
questions you have.
[The Prepared Statement Of Shelby Oakley Appears In The
Appendix]
Mr. Rosendale. Thank you, Ms. Oakley.
The written statement of Ms. Oakley will be entered into
the hearing record.
Ms. Schank, you are now recognized for 5 minutes to deliver
your opening statement.
STATEMENT OF HANA SCHANK
Ms. Schank. Thank you for holding this hearing today and
for inviting me to offer my expertise.
I have been a technologist for my entire career, starting
at Andersen Consulting in the nineties. I then started a small
IT consultancy, which I ran for 15 years. It was exactly the
kind of small business that government should want to work
with: efficient, user focused, and low cost.
While I landed some small government contracts, I never
even considered bidding on a major piece of work, because
responding to huge government Request for Proposals (RFPs)
required an equally huge lift. More concerning, the government
RFPs that came across my desk seemed to be written by people
who did not know what they were asking for.
After healthcare.gov failed to launch, I joined the United
States Digital Service, USDS, hoping to bring my technical
skills to government. Like many USDSers, I immediately grasped
that the root of all government system failure was not a lack
of hands-on technical talent, but the procurement process.
When I was at Andersen, we were one of a number of
companies competing for these contracts. Over the years, the
industry has consolidated. Andersen Consulting has become the
mega company Accenture, and government's requirements have
become increasingly onerous, requiring specialized teams who do
nothing but respond to government RFPs. Only the very largest
firms have the capability to compete. According to the GAO, as
of 2021, only 30 contractors accounted for 75 percent of all of
VA's IT obligations.
These big firms do not play fairly, gaming the system to
elbow smaller companies out of the way, often by acquiring
them. They also account for every single one of the large-scale
Federal IT disasters.
At USDS, I worked on one of those systems, Electronic
Immigration System (ELIS), the Nation's immigration case
management system. The United States Citizenship and
Immigration Service (USCIS) began building ELIS during the
George W. Bush administration and scrapped the entire system 7
years later.
By 2015, when USDS came onto the project, they found that
the contractor, the Electronic Immigration System (IBM), had
designed ELIS so that it relied heavily on IBM products, even
when those products did not benefit the system. As a result,
ELIS took five times longer than paper to move the applicants
through.
Today, ELIS is stable and functional. The way USCIS got
there was threefold. First, they extricated themselves from
their monolithic contract with IBM. Next, they carved out
smaller, discrete projects with clearly articulated outcomes
goals, mostly--which mostly went to smaller vendors. Finally
and most important, USCIS built an in-house team with technical
expertise who could correctly procure and manage IT projects.
Following ELIS, USDS created the Flexible Agile Solutions
for the Homeland (FLASH) procurement vehicle to encourage a
more diverse set of vendors to compete. FLASH failed, but other
Federal agencies later refined the process to successfully
contract with smaller vendors. These agencies include Centers
for Medicare and Medicaid Services (CMS), Office of Personnel
Management (OPM), and--wait for it--VA.
VA has a procurement vehicle called CEDAR, which
prequalified four service-disabled, veteran-owned small
businesses to receive task orders for IT development and
delivery. There are talented tech teams within the agency who
are working to get it right, but these teams are small and are
often not consulted when it comes to these large-scale
procurements.
The way forward for VA is clear. Like USCIS did, they must
free themselves from the large vendor contracts that are not
serving them or their users, and relying on the expertise of
their own digital service teams, work to bring on smaller,
responsive vendors for discrete pieces of work.
I would like to close with a positive story.
Dr. Williams is a psychiatrist at a VA in a large
Midwestern city. When she first came to VA, she was delighted
to find that VA's case management system made her work easier,
unlike her previous experiences using Epic and Cerner.
At VA, care is provided holistically, which means
practitioners work together across specialties to provide the
best care for veterans, and the IT system facilitates this. Dr.
Williams once met with a veteran who was in the process of
moving, as vets often do. Because her meds were packed in one
of numerous cardboard boxes, she would stopped taking them and
had become actively psychotic. Using the VA's existing case
management system, Dr. Williams was able to look up the
patient's history, place an order with the onsite pharmacy for
the missing medication, arrange a home health nurse to assess
the veteran's new home and create a system that would remind
her to take her medication, enroll her in an outpatient program
so she could get evaluated the next morning, and arrange a van
to transport her to a clinic near her home.
The system allowed Dr. Williams to do all of these things
in less than 30 minutes. The vet was able to stay out of the
hospital and get on a path to health.
This is how government systems should and can work, but you
can not get there with a small pool of vendors who take
advantage of government's lack of technical know-how to line
their own pockets. VA has the chance to do something wonderful
for our Nation's veterans. They should take every opportunity
to get it right and serve our veterans as intended.
Thank you.
[The Prepared Statement Of Hana Schank Appears In The
Appendix]
Mr. Rosendale. Thank you, Ms. Schank.
The written statement of Ms. Schank will be entered into
the hearing record.
We are now going to proceed forward with questions, and I
will recognize myself for 5 minutes.
Ms. Oakley, let us start with the data points that you
collected.
According to your report, the VA's spending on IT contracts
rose from $4.2 billion in 2017 to $6.5 billion in 2021, but the
number of contracts dropped by more than half, from 4,831 to
2,263.
Why did this happen?
Ms. Oakley. What we--what we recognized in these trends was
that VA is awarding less new contracts, and they are issuing
modifications or task orders on existing contracts more
frequently. That would contribute to the decline in new
contract awards.
That being said, the value of new contract awards has gone
up pretty significantly over the past couple of years. VA seems
to be relying on its existing mechanisms as opposed to putting
in place new contracts to be able to award this funding.
Mr. Rosendale. Ms. Oakley, you also found that the number
of companies working for the VA, which you just referenced,
dropped from 1,247 companies in 2017 to 873 companies in 2021.
Can you explain how that happened?
Ms. Oakley. Absolutely. I mean, this is a common trend
across the Federal Government. The number of companies that are
doing business with the Federal Government has been decreasing
significantly over the past--10 to 15 years. VA's total
decrease is about 25 percent in the number of vendors that are
doing business with VA.
That makes sense from the perspective also of VA's efforts
to put in place things like category management, where they
combine requirements for common products and services on
specific contract vehicles to be able to get the best price and
save money, operating as one buyer across the VA and across the
Federal Government. That is part of that trend governmentwide.
Mr. Rosendale. I understand that. It sounds like you try to
utilize, in the private sector, the economies of scale.
Ms. Oakley. Yes.
Mr. Rosendale. You have got all of these things, and you
are trying to get it so that there is one--one person or three
or whatever that are handling that. In the private sector, that
is used as a tool to drive costs down----
Ms. Oakley. Yes.
Mr. Rosendale [continuing]. and increase efficiencies.
Meanwhile, I think we are seeing the exact opposite here.
How much of it is due to the marketplace and how much of it
is due to the VA's decisions and their practices?
Ms. Oakley. I do not have data on how much is due to the
marketplace for sure, but I definitely know that is the
intention of VA using contracts like T4NG or the NASA SEWP
contract, is to kind of get that preferred pricing, that better
pricing that comes from having that data that shows, hey,
NASA's paying this much and Health and Human Services (HHS) is
paying this much, we should be paying this much as well too, as
opposed to each individual contracting officer going at it
alone.
You are right----
Mr. Rosendale. Again, we continue to go back to the intent,
and the intent is to utilize the economies of scale. When we
look at the hard numbers and we see that the contracting went
up from 4.2 billion to 6.5 billion----
Ms. Oakley. Yes.
Mr. Rosendale [continuing]. clearly this is not working.
Ms. Oakley. Well, you know, the majority of that increase
from 2017 to 2021 is really driven by those large-scale IT
modernization efforts that we have talked about a number of
times. VA has undertaken a significant number of IT
modernizations all at the same time that are really driving
those increasing costs. Financial Management Business
Transformation (FMBT), Electronic Health Record Modernization
(EHRM), those kinds of things are really pushing a lot of that
cost increase.
Mr. Rosendale. One other question. Do we have any type of a
breakdown--do you have access to any type of a breakdown to
show exactly the difference between the cost that was
contracted for initially and had it been intended to be the
total cost and how much has been added on as extras through
change orders, through modifications of any form or fashion?
Ms. Oakley. Yes. There are two different issues at play.
For the large-scale modernization type efforts, we would be
able to more easily figure out kind of what that delta is
between what we expected it to cost and what it is currently
costing.
For contracts like T4NG or, you know, NASA SEWP or
whatever, there is a upper ceiling limit for those contracts,
and so task orders can be continued to be awarded on those
contracts until they hit that ceiling. And then a new contract
needs to be negotiated at that point in time.
They are kind of two different issues at play.
Mr. Rosendale. Very good. Thank you very much.
I will yield now 5 minutes to Representative Cherfilus-
McCormick.
Ms.Cherfilus-McCormick. Thank you, Mr. Chairman.
As is common now in Federal contracting, small companies
are acquired by large government contractors, which can
potentially lead to conflict of interest. This has certainly
been the case for our--from our observations of the T4NG
contracting vehicle which is being used by the VA and
administered by the Technology Acquisition Center.
GAO's testimony and recent reports have highlighted that,
as dollar amounts excuse me have increased for IT contracting
at VA--thank you--at VA, the recipients of the awards has
decreased.
My question is for Ms. Oakley.
Ms. Oakley, from your observation, has VA done a sufficient
job to manage the number of companies that were on T4NG's
vehicle?
Ms. Oakley. Yes. I guess answering the sufficient aspect of
that, I think, is a little bit difficult for me, because I do
not know what the expectation would be for the number of
companies that should be on a contract vehicle like this, but
T4NG is set up such that there are opportunities for on-ramping
companies, and mainly those opportunities are targeted at small
disabled-veteran-owned businesses.
I think those have been provided, and we have seen a growth
in the number of contractors on T4NG since its inception to
what the current contract is as well too.
I think that that is definitely something that we have
seen. Then there is a pretty even split between small
businesses and large businesses on those contracts.
The other thing that I would mention in terms of T4NG is
that one of the biggest benefits is saving administrative time,
getting contracts turned around quickly, because that is what
those types of contracts are intended to provide, that
efficiency. I think that is where you are seeing that drive in
use on T4NG, is that ease of use for contracting officers who
have--at VA, have to deal with, for example, applying the rule
of two to every single procurement that they make.
Ms. Cherfilus-McCormick. Why does the government use
category management, and what benefits has the government
gained from such strategies?
Ms. Oakley. Yes. Category management, as I mentioned, is a
way for the government to leverage its buying power and act
like more of a single buyer of common products and services,
and so these are things that are commonly available
commercially. Our work over the years has focused on what
benefits category management has provided. In 2021, we reported
that the government has saved about $30 billion just in 3 years
through its category management efforts. It does drive savings
across the Federal Government.
We have not done that work specifically for VA, but we have
some work coming for you all in that regard over the next year.
It just provides the data specifically across the government to
be able to make those purchases.
One thing I will say that we have found to be a continuing
challenge is that defining of requirements. We want to look at
category management as a holistic thing, not just the
contracting approach. Being able to work more effectively to
define the requirements of what you want through category
management will enable agencies like VA to better get what they
need through these types of efforts.
Ms. Cherfilus-McCormick. We have heard of several IT
products and software, including software for blind and low-
vision VA employees, where VA is not utilizing an enterprise
contract. Is there any benefit to VA making employees purchase
these items and services individually?
Ms. Oakley. I am not familiar with that contract for sure.
I think we point out one of the challenges that we have
identified for VA acquisition management is contracting officer
workload, and one of the things that we have seen over the
years is that VA has not been operating as a consolidated
enterprise in a number of areas.
What that has resulted in is contracting officers having a
higher than normal workload, issuing individual contracts and
task orders for their own purposes. I think to the extent that
VA can operate from a more holistic perspective, with an
organization as big as VA, significant savings could be
achieved.
Ms. Cherfilus-McCormick. Thank you so much for your
testimony.
I yield back.
Mr. Rosendale. Thank you very much.
Representative Mace, you are recognized for 5 minutes.
Ms. Mace. Thank you, Mr. Chairman.
I want to thank the ranking member for being here today.
I would like to thank both of you for showing up and taking
our questions this morning, because, as you know, the VA
refused to show up. Just appreciate your effort to explain to
us these--sort of the process and how it works.
Ms. Schank, my first job out of college at The Citadel was
at Andersen Consulting. Late nineties, Y2K, Common Business-
Oriented Language (COBOL), C++, Sequel, Hypertext Preprocessor
(PHP), the whole nine yards, that is where I got my start.
Really appreciate your perspective on government contracts and
being here today.
The VA's entire modernization effort requires much work,
but today we are here to provide congressional oversight over
the IT contracting process, and GAO has determined the VA to be
a high-risk agency, as you all have stated this morning,
because it has among the highest obligations in number of
contract actions in the Federal Government.
This extremely high number of contract obligations requires
constant oversight to ensure the American taxpayer dollar is
being spent in the most efficient way possible and that
veterans are receiving the best services available. Sometimes,
when I am talking about IT and the VA, my head wants to explode
because of the number of contracts, the lack of oversight, the
lack of, actually, following the process and the rules by the
VA.
I sat in on a women's veterans roundtable yesterday, and I
appreciate the comments about the example, Ms. Schank, that you
shared about a veteran who did not have her medication and then
was able to get it in record time. There was a woman at the
roundtable, and she needed a--just her annual pap smear. It
took 6 weeks--6 to 8 weeks to get it scheduled. Once she got it
scheduled, it would take another 4 weeks to get her results.
You know, we still have such a long way to go to care for our
veterans, and we can not even get their records, you know,
electronically, or they can not.
I come from a long line of veterans in our family and, you
know, I hear about these issues all the time. I hear success
stories, but I also hear about a lot of failures. We have--one-
third of all veterans in the state of South Carolina live in
the district that I represent, so we hear from vets day in and
day out.
Ms. Schank, my first question to you this morning is: How
do we increase competition with some of these contracts, how do
we push the VA to do the right thing in this regard, to do
better, more faster?
Ms. Schank. There are really two key points.
Ms. Mace. Uh-huh.
Ms. Schank. One is that, as I said initially, the
contracts--these giant, monolithic contracts will never succeed
and they will never serve veterans or the VA. The only way is -
and you know from your experience at Andersen, that was how
things were built in the nineties, but the industry has moved
on.
For the vast majority of the private sector, they are not
building stuff that way anymore. It is not a giant waterfall
process with a big bang turn-on date. Everybody has learned
that the way you build tech successfully is to chunk it up into
small, individual pieces, and then build incrementally.
The first piece is not writing these giant contracts and
allowing--which is going to allow more vendors to compete, but
also is the right way to build technology.
Then the other piece is that, because government has
outsourced all of this technology for so long, there is a real
lack of tech expertise and tech fluency. I am not talking about
people who can code. I am talking about people who have some
understanding of how technology works and the decisions that
they are making and how that then impacts the end user.
Most agencies--I would say almost all--are very, very thin
with top senior people who have some degree of tech fluency.
Part of that is--well, a large part of it is that technology
has only evolved in the last, let us say, 20 years. If you were
somebody who was interested in going into government, you are a
policy person, and you know about policy, and you know about
law, and you might not know about technology. The world has
changed.
In the USCIS example that I gave, Leon Rodriguez, who was
running that project, had been a policy person his entire life.
He was running USCIS. He did not have tech fluency, and so when
contractors would come to him and say, Should I do A or should
I do B, he did not have the full understanding of how that was
actually going to play out for users and for the project.
He talked about that he wished he had a technology
translator who could say to him, okay, so this is what they are
asking you, and here are the ramifications.
That is not a position that exists in government. These
agencies are slowly trying to staff up, but it is lacking.
Until that skill set exists in Federal Government, these
contracts are just going to continue.
Ms. Mace. Thank you, Ms. Schank.
Mr. Chairman, I yield back.
Mr. Rosendale. Thank you, Representative Mace.
Now on to Representative Self. Thank you very much.
Mr. Self. Thank you, Mr. Chairman.
First to Ms. Oakley: Is this issue that you are describing
here across the Federal Government as a whole? Is this just a
microcosm of what you see?
Ms. Oakley. I think to some extent, yes, right? I mean,
overall, as I mentioned, we have seen a decrease in the number
of companies that want to do business with the Federal
Government. We see continuing mergers and acquisitions amongst
all the major Federal contractors.
Ms. Shank was exactly right. When a small innovative
company is there, they get bought up by the big guy.
Mr. Self. That is part of the American Dream?
Ms. Oakley. That is what happens, yes.
Mr. Self. A lot of this you ascribe to COVID-19 IT. Was
that a one-off or is that translatable into the post-pandemic
era? Is this something that was a one-time shot?
Ms. Oakley. Yes. A big majority of the spending increases
that we saw was due to addressing the COVID-19 pandemic and
providing technology and that kind of thing.
We did run the numbers for Fiscal Year 1922, because our
report was done in fiscal year 1921, and it seems as though the
trend is continuing. We have not been able to get behind the
data yet because we just did it to support this hearing, but it
seems as though VA's trend of increased IT spending is
continuing.
Mr. Self. I would like to see that----
Ms. Oakley. Okay.
Mr. Self [continuing]. when you get the handle on the
numbers.
Ms. Oakley. Absolutely.
Mr. Self. For Ms. Schank, we have this problem, I think, in
Department of Defense (DOD), for instance, the Defense
Industrial Base, procurement in general. I realize that is
manufacturing versus service. That is IT versus manufacturing.
You have told us what VA can do.
What, in general, can Congress do, though, because, as I
said, you know, selling your small company to a large company
is part of the American Dream. We are not going to stop that.
How does Congress--what would you recommend that Congress
do to--to fix these issues? We have talked about the issues.
Now, you are talking to a legislative branch. What can we do
that does not inhibit the American Dream and yet moves in the
direction that the two of you both have indicated would be
positive?
Ms. Schank. One of the big issues is that there is no
accountability for these large companies.
Mr. Self. Uh-huh.
Ms. Schank. Acquiring things is fine. As--you know, yes, a
lot of people build a company so that it can get bought, and
that is fine.
Mr. Self. Right.
Ms. Schank. The problem is that they--they are not held--so
these companies fail, and then the contracts are renewed. It is
the same 10 players. When you talk about--I mentioned
healthcare.gov, Oracle. You know, I talked about ELIS failing,
IBM.
Mr. Self. Okay. Okay. How would we do that? What does
accountability at the congressional level look like----
Ms. Schank. So----
Mr. Self [continuing]. in your expertise?
Ms. Schank. Yes. There are not--in these contracts, there
are no outcomes that are specified. In government, delivering
the product is the end. You do not have to deliver it well. You
do not have to deliver it in a way that improves accessibility.
In the private sector, there are metrics that you would use
to say, did this achieve our goals? In government, the goal is
very often just launching the system and not how well it
performs and what the outcomes are.
So holding----
Mr. Self. Do you recommend something other than indefinite
delivery, indefinite quantity contracts?
Ms. Schank. Yes. I think that, in general, contracts that
specify outcomes are successful, and that is been proven. There
are multiple smaller contracts where government has specified
you need to, you know, move X number of people out of beds and
into--homeless people out of beds and into shelters, or, you
know, into their own homes, and that is how we will decide if
this was a success or not, not is the product on and
functioning.
Mr. Self. We have got accountability. We have got outcomes.
Okay. Here comes the tough question. The Electronic Health
Record (EHR) contract, is that even breakable into smaller
contracts?
Ms. Schank. Absolutely. I mean, even as I was kind of just
doing some digging around how people use the system, I started
thinking like, oh, okay, so you would--in a normal scenario,
you would break this up into the scheduling contract and the--
so you can break these things up into individual pieces,
discrete pieces. Every big-tech project is--I do not want to
say breakupable, but it can be divided that way.
Mr. Self. Ms. Oakley, what is your opinion of that?
Ms. Oakley. I agree completely. We have done a lot of work
on the practices of leading companies that do product
development for cyber products as well as cyber physical
products, and we are completely in line with Ms. Schank in
terms of these companies take an iterative approach in
development. They measure their expectations for their pieces
that they are chunking out. They call them minimum viable
products. It is like, okay, we can achieve this in this
timeframe, and it provides a useful capability to a customer.
That is how they do their work.
Mr. Self. My time is up. Mr. Chairman, I yield back. Thank
you.
Mr. Rosendale. Thanks very much, Representative.
It sounds to me as we are sitting here talking that if we
had smaller pods or components which communicated with each
other seamlessly, that therein is the other component, that
they communicated with each other seamlessly, that we could get
these systems working. The problem is we have people that are
putting the requirements out that are not really familiar with
the application, okay, how it is going to be utilized, or as I
love to quote Dwight Eisenhower, farming looks very easy when
you live a thousand miles away from a cornfield and use a
pencil for a plow. They are just completely out of touch with
what is actually going on.
The other thing is that, while we talk about the American
Dream--and I really agree with Representative Self that it is
to take a small company, grow it, sell it to another company,
or just grow it big by acquiring other smaller companies. In
the private sector, that does not happen unless you produce a
successful product. That is not what is happening with the VA
right now.
Ms. Oakley, you determined that 10 companies receive nearly
half of the VA's IT contracts during these 5 years, worth $12
billion. What are those companies, and is that typical in a
federal agency?
Ms. Oakley. Yes. I have a list here. The top 10 companies
are Cerner Government Services; Denali Holding; Booz Allen
Hamilton; AT&T; Four Points Technology; Leidos; Veterans Tech,
LLC; Liberty IT Solutions, which was actually acquired by Booz
Allen Hamilton; Nester Consulting; and Accenture.
I do not have the governmentwide data to know the
concentration for contractors governmentwide, but we did crunch
some numbers for you all in terms of professional and support
services contracts, and it seems to be about the same, about 40
percentish of all obligations are concentrated amongst 10
different contractors.
Mr. Rosendale. Ms. Schank, how does this concentration
affect the results we see on VA IT projects, and do you think
that that is healthy?
Ms. Schank. The concentration is absolutely the opposite of
what you would want to see for tech development. The issue is
that there are fewer--and it goes back to, again, why are these
giant contracts being written? The only people who can - and
the only companies that can compete for these giant contracts
are these same giant companies, which is how you keep getting
these repeat offenders.
In the private sector, most places have brought this work
in-house, so they are not hiring these giant companies.
Government made the decision not to bring that - and has not,
thus far, brought a lot of that tech talent in-house, so they
are the last man standing when it comes to these giant
contracts.
Mr. Rosendale. Let us look at the top three companies. The
first is Oracle Cerner. EHRM has clearly been a disaster, and I
expect to see significant savings squeezed out of that
contract. A system we have not talked much about is the H.R.
Smart, which is decades-old human resources software from
Oracle, PeopleSoft, that VA is abandoning.
I am not going to disclose VA's cost estimate to purchase a
new H.R. system, but it is significant.
In your experience, how well do such large, one-stop shops
for different kinds of software--for example, medical records
and human resources--tend to perform compared to specialized
software companies?
Ms. Schank. Not well. These companies operate by building
in their own licensed--their own proprietary tools, and then
they live off of the licensing contracts, and they are able to
lock themselves in so that no agency can free themselves from
this contract, and they have to keep just re-upping--modifying
whatever the tool is that they have bought to try to serve
their users.
A smaller vendor who is specialized would build it right
the first time--the first way. They do not already have an
existing thing that they are trying to cram in to work for this
particular product.
Mr. Rosendale. Let me ask you something. If we look across
the VA system, we have got 179 medical facilities. Do you think
it would be more effective and efficient to try and develop a
electronic health records system and also a management system
for the facility, or is it going to be more efficient to
develop a system for each one of the components--for the
pharmacy, for the records--and then, either which way, that is
got to be seamless so that they can work across the spectrum?
Ms. Schank. Well, as my co-person here testified, you want
to build a minimum viable product. This is how technology is
developed. What you are describing is the correct way to
develop technology, where each one of these things would be
developed separately, with the understanding that they would
roll up into one large system.
They all have to do different things. They might have
different users. Their goals are different. The outcomes, in
theory, if there are outcome goals, they would be different for
each one of those individual pieces. They should be developed
separately, tested, and then rolled up into a larger system.
Mr. Rosendale. I understand that, but I am trying to get to
the--is it the site with all its components that should have a
separate system that can communicate with the other sites or is
it each of the components within that site, whether it is
scheduling, whether it is prescription drugs? Should each of
those components be different or should it just be the site
itself?
Ms. Schank. Each of those should be developed individually.
Mr. Rosendale. Okay.
Ms. Schank. Yes.
Mr. Rosendale. Thank you very much.
Representative Cherfilus-McCormick, I yield for 5 minutes.
Ms. Cherfilus-McCormick. Thank you, Mr. Chairman.
Ms. Oakley, as you have noted in your testimony, VA
acquisition management remains on this year's high-risk list.
VA acquisition management has been on the list since 2019 and
in Fiscal Year 2022. VA obligated 56 billion for goods and
services.
Before we dive into the IT acquisition component of this, I
would like to ask you about the progress that you feel the VA
has made since last year's report, if any.
Ms. Oakley. Yes. We did report progress that VA made. In
particular, we focused on leadership commitment. We do see a
commitment from, for example, the chief acquisition officer in
the organization to making improvements with regard to VA's
acquisition management efforts.
We also noted improvement with regard to their action plan.
That is what we say is required to be able to outline the steps
that you are going to take to improve in this area.
VA has also implemented a number of our recommendations
over the past couple of years, so they are moving in the right
direction.
The action plan itself does outline a number of really
significant efforts that have to occur to be able to drive
improvement and, no pun intended, but now we just need to see
action on the action plan.
Ms. Cherfilus-McCormick. Thank you.
Also, Ms. Oakley, I know that, in the report, GAO makes
reference to VA's efforts to implement its new acquisition
framework. We have now seen the completed negotiation for the
continuation of the Oracle Cerner contract, and we have heard
that the VA has plans to move forward with a contract for
supply chain modernization sometime soon.
Are you aware at all of how the VA has used this
acquisition framework to manage and prepare these acquisitions?
Ms. Oakley. I am not aware, on the EHRM program, how the
acquisition lifecycle framework or the draft framework
contributed to how they negotiated that contract.
The framework is broader than just the contract. It is the
approach to the program. I would hope that as that framework
rolls out, that VA takes some significant steps to adjust the
management and oversight and execution of that program to be in
line with what is required of their draft framework.
With regard to the supply chain modernization, as we
understand it, VA is trying to use the supply chain
modernization as a testbed for the acquisition lifecycle
framework. I do not have any information, nor have I reviewed
any documentation, that would give me an indication of how
indeed they are doing that.
VA recently provided us some information on the supply
chain modernization efforts that are moving forward. I will
look forward to seeing how those principles from the
acquisition management framework are being applied, and we will
definitely assess that.
Ms. Cherfilus-McCormick. I think we are all looking forward
to hear about that.
Our committee has received a number of concerns from
industry on the current plans for supply chain modernization,
as well as concerns for inside of the VA on their current
strategy. We have learned from the past decades of failed IT
modernization attempts that, at the very least, a project that
is of scale should have costs and estimates, a defined scope
and a schedule.
Following on your work on high-risk lists, are you aware if
the VA has completed any of those requirements for the supply
chain modernization?
Ms. Oakley. At the last hearing I was at, I think Mr.
Christy testified that, no, they had not in terms of especially
the cost estimate.
At that time--and I will continue to reiterate that a good
cost estimate, a defined scope and schedule, manageable
requirements, are all keys to establishing a sound business
case for moving forward with a program a related acquisition.
I would expect and hope that VA has done that. As I have
cautioned before, VA often puts action ahead of strategy. We
recommended that VA do a comprehensive supply chain management
strategy to be able to drive the approaches that it is going to
take for contracting for this. That is not complete yet, but
that acquisition is in the process of moving forward for the
actual system itself. That gives us a little bit of pause of--
is this the cart before the horse again?
Ms. Cherfilus-McCormick. Well, I have one more quick
question. Would you recommend VA move forward with a contract
competition without those basic items?
Ms. Oakley. I guess I would say that our work has found
that those items are critical to that business case that I have
talked about, and if you do not have a sound business case--
this is where we have seen those failures that we have been
talking about in the past. If they are not in place, I would be
hesitant to suggest that this was a sound proposition.
Ms. Cherfilus-McCormick. Thank you.
Mr. Chairman, I yield back.
Mr. Rosendale. Thank you, Representative.
Representative Self, I recognize you for 5 minutes.
Mr. Self. Thank you, Mr. Chairman.
I will just point out that in your written testimony, you
noted that the VA has implemented 24 IT recommendations since
2010. They have not implemented 22. Now, I understand you have
no authority to force implementation, but I just want to note
they have not done 22.
Now, do you agree with Ms. Schank that in-house
management--project management I think is the way she would
describe it--is the way to go?
Ms. Oakley. I think it is necessary that the government has
in-house expertise to be able to effectively oversee these very
complex procurements, for sure.
Mr. Self. Okay. Now, for VA, let us remember, they used to
do their own construction.
Ms. Oakley. Right.
Mr. Self. I believe the overruns on one particular--I have
forgotten where it was--Colorado or somewhere--construction
project management was taken away from VA----
Ms. Oakley. Uh-huh.
Mr. Self [continuing]. and given to the Corps of Engineers
for this very reason.
Ms. Oakley. Yes.
Mr. Self. Now, I realize it is different, but do you stand
by your in-house recommendation knowing that VA could not
manage their own construction projects?
Ms. Oakley. I think maybe we are talking about two
different things.
Mr. Self. Uh-huh, we are.
Ms. Oakley. I would say that there was a pendulum that has
swung, right?
Mr. Self. Uh-huh.
Ms. Oakley. Like, the government used to do everything
itself, right? Then, at a period of time, it swung to hands
off, the contractors are going to be doing this kind of thing.
I think there is a happy medium that requires the
government to have expertise to understand what it is buying,
what it is getting, and what its customers need to be able to
put in place contractors, the private sector, to be able to do
those things. I think it does not mean that VA has to go and
build these systems itself. It means VA has to know what it
wants and to be able to determine if it is getting what it
wants.
Mr. Self. Okay. Ms. Schank, would you describe to me what
this in-house office looks like?
Ms. Schank. Of tech expertise?
Mr. Self. Correct.
Ms. Schank. It is really that program management layer. The
issue is that the distance between policymakers and what gets
delivered is too great, and you can not have that in
technology. That distance needs to be shrunk. There needs to be
a senior level of people who have tech fluency who can manage
these projects, not, you know, in like a code review kind of
way, but just in a general strategy, here is what we want our
users to be able to do.
Mr. Self. Is this at the VA level? Is this at the Veterans
Integrated Services Network (VISN) level? What level are you
discussing?
Ms. Schank. It is at the VA level, you know, in the CIO's
office, let us say.
Mr. Self. Okay. Again, so you think that this is action--
and I will ask both of you--action that Congress ought to take,
to redesign this infrastructure?
Ms. Schank. I think that this is not an issue that is
unique to VA. This is across every Federal agency.
Mr. Self. Uh-huh.
Ms. Schank. I am not an expert on what Congress can do, but
if there are levers that Congress can use to ensure that there
is senior--that there is a senior level of tech fluency at
every agency, that would change things immensely.
Mr. Self. Ms. Oakley.
Ms. Oakley. I guess I would say some of the things that
Congress can do is support workforce initiatives that bring
highly skilled folks like that to the government. I think we
focused on this in the past with cybersecurity, for example.
Those are the kind of actionable things that Congress can
support to say, okay, VA, we need you to get this expertise in-
house to be able to execute these programs for us to get what
we want out of them.
Mr. Self. Okay. Mr. Chairman, I yield back.
Mr. Rosendale. Thank you very much, Representative.
I still want to highlight some of the issues that we are
facing and the amount of money that has been thrown at this
problem. We talked about Oracle a moment ago, but, Ms. Schank,
the No. 2 vendor is Dell, which VA mainly uses as a
clearinghouse to buy different companies' software. No. 3 is
Booz Allen Hamilton. Here is a partial list of the major
contracts Booz Allen Hamilton has held:
The Veterans Benefits Management System, which cost about
$1 billion. It was rolled out roughly a decade ago and is in
widespread use, but it has limitations that the VA is still
struggling with.
The Forever GI Bill IT system changes that failed and
delayed veterans' housing stipends in 2018, that cost about $12
million before the plug was pulled on that program.
The Program Management Office for the Defense Medical
Logistics Standard Support (DMLSS), DMLSS, supply chain
project, which spent at least $230 million before it failed
last year.
The EHRM Program Management Office, which has produced
terrible results and has cost at least $728 million.
We are talking about massive amounts of money that have
been thrown at IT.
Ms. Schank, in your experience, is it normal for a company
with this track record to appear and reappear on so many
different projects, and why does this continue to happen?
Ms. Schank. Unfortunately, it is absolutely normal in
government that this happens. This happens time and again.
There was a very public example in Rhode Island where the State
had a public benefits application that they were rolling out.
They had contracted with Deloitte. It had been under
construction for a long time, cost overruns. It failed, and the
company that they hired to fix it was Deloitte. In part, that
is because it is such an enormous--again, there are only a
certain number of--the way that contracts are currently
written, there are only a certain--small handful of companies
who can meet the requirements and can bid on those contracts.
What you get is just these repeat offenders. There is no
accountability, and they continue to get hired.
Mr. Rosendale. Yes. Again, if we compare it to the private
sector, while it is the American Dream to grow a big, large
company, that usually comes with success because you have had
satisfied customers, not all of these failures that have been
demonstrated, at someone else's expense.
Ms. Oakley, I understand agencies do annual past
performance evaluations for every contract, and those ratings
tend to be wildly out of step with the actual outcomes of the
projects. Why is that, and how does this perpetuate failed
projects?
Ms. Oakley. We have definitely reported in the past that
there tends to be a lack of frank performance information
provided in some of these systems. I have not dug into the
reasons for that. If I were a betting person, I would say that
contracting officer workload probably drives a bit of that
phenomenon.
For VA, in general, our work showed that, indeed, they
actually, overall for IT contracting, had about a 96 percent
rate of actually entering the performance information into the
system, which is better than some agencies. We did not assess
the quality of that information as well too.
If I can just comment really quickly--the government plays
a role in ensuring success of the programs. When we are talking
about the contractors and their performance, when they are
being approached with these gigantic programs like EHRM, like
FMBT, like the supply chain modernization efforts, that are
trying to be the be-all end-all of everything, that is the
problem, right?
The requirements are unattainable, and so these contractors
sign up for these things, because it is money. You are signing
up to do these contracts. The requirements just are not
executable in the way that they are structured, and so then it
kind of snowballs from there.
That is where the smaller, more manageable chunks of
requirements and capability are so important for the government
to approach contractors with, because you would get that
performance and then you would not be locked in to staying with
that contractor for the continuation.
Mr. Rosendale. Sure. Sure. Also--this committee has heard
about the Veterans Health Information Systems and Architecture
(VistA) system and how the veterans facilities, the VISNs all
across the country still had that as their backup, that it is a
systemwide program that everybody is still falling back on. I
would like to try and figure out how we can somehow incorporate
any additional components that are missing from VistA that we
are trying to provide through Oracle.
Representative Cherfilus-McCormick, I will recognize you
for 5 minutes.
Ms. Cherfilus-McCormick. Thank you, Mr. Chairman.
Yesterday we had a full committee hearing on issues with
tracking of supplemental funding provided by Congress during
the pandemic. The inspector general recently released a report
stating that because of the 30-year-old financial management
system, which has not been updated, the manual journals were
required to document spending. This, of course, provided
problematic because of the lack of guidance provided to
employees on how to document the spending. The Financial
Management Business Transformation program is meant to update
the system, but, to date, our committee has seen increased
timelines proposed for implementation at (Veterans Health
Administration) VHA.
My biggest concern is--with the program now is its
interdependency that has been identified by VA with the supply
chain modernization initiative finalizing their plans for a
future solution. The interdependency between large IT
modernization programs seems to me to be a major risk factor
for the successful IT acquisition.
Ms. Oakley, who is responsible for coordinating the timing
and planning of these large acquisition programs?
Ms. Oakley. Yes, I mean, there is a number of different
officials that are responsible, which I think is part of the
problem. In theory, the chief information officer is
responsible for reviewing and approving VA's entire portfolio
of IT acquisitions, and so they should have that insight.
That being said, the CIO is not an acquisition
professional. You have the chief acquisition officer that also
has to be able to understand what is going on and ensure that
the contracts are being structured and executed in line with
the strategies that the CIO has put in place.
As we have kind of talked about before in prior hearings
and we have reported a number of times, there is a bit of a
fragmentation of leadership within VA for managing and
executing large acquisition programs.
Ms. Cherfilus-McCormick. Now, are there any other entities
that you believe play a major role in those large contracts
also?
Ms. Oakley. I mean, well, chief information officer, chief
acquisition officer, the Chief Financial Officer (CFO),
obviously, is going to have to play a role from a budgeting
perspective. At the highest level, those are the folks in the
positions that, in theory, should be driving those decisions.
Ms. Cherfilus-McCormick. Do you have any recommendations on
how we can streamline so they are working together instead of
the fragmentation that you proposed--that you talked about?
Ms. Oakley. Yes, I mean, they have devised a number of
governance boards where, I think as I have said in the past, it
is kind of an approach of we are going to all just agree and
hopefully it works out and make decisions that way. Having that
kind of belly button person who gets to say, no, I have taken
all your inputs from all the experts and now we are going to go
this way, I think, is an important thing.
One of the bills that we talked about in the last hearing
was that chief management officer bill that you had talked
about, that had been proposed. That is a role that that
position could play in making sure that those investments are
coordinated.
Ms. Cherfilus-McCormick. Has the VA been receptive toward
addressing the fragmentation?
Ms. Oakley. I think to some extent. The kind of development
of those governance boards and whatnot and through the
acquisition lifecycle framework, the draft acquisition
framework that they are hoping to put in place, that does drive
a little bit more accountability and responsibility for
decisions throughout the acquisition lifecycle process, which
goes beyond just the contract awarded.
I think they are taking steps to bring more rigor and
discipline to the approach. We have not seen wholesale change
quite yet.
Ms. Cherfilus-McCormick. Thank you so much.
Mr. Chairman, I yield back.
Mr. Rosendale. Thank you very much, Representative.
Representative Self, I recognize you for 5 minutes.
Mr. Self. Thank you, Mr. Chairman.
I think we are getting to the bottom-line issue, which is a
governmentwide issue. Not only U.S. Government, it is every
government. How do we instill accountability? We can go around
and around on this. The fact is the government has no profit
motive, because in private sector, you know, you have a
measurement in place. You either make a profit or you go out of
business. We do not have that in government.
We have rent seekers, people who want the advantage, to get
all of the government contracts that you ladies have been
describing.
I still am wrestling with how do we put accountability into
this. We can add a layer of bureaucracy, which we have
suggested here, but because we have no profit motive, I am
still wrestling with how we get to an outcome-based system,
because we are spending a whole lot of taxpayer dollars here.
I have not heard of anything yet that gives me the
confidence that we have any outcomes, because we have used the
term, but I think this committee ought to try get at the
baseline here what we need, and that is outcomes.
In an IT system, it looks to me like you would have a
list--maybe Ms. Schank could weigh in on this--you would have a
list of outcomes that would be generic enough, that would be
across the systems, without saying you got to have this field
in this form.
I would be interested in that, if you could give us an
idea, because to me that is the bottom line. How do we get to
accountability outcomes without just adding a layer of
bureaucracy and saying that is going to fix the problem? I do
not think it ever fixes the problem for us to simply add a
layer of bureaucracy.
Ms. Schank, would you care to comment?
Ms. Schank. Yes. There are well-defined outcomes for how
these systems should perform that are used in the private
sector. Those are always human-centered outcomes.
The humans at the end of the system, are they able to
achieve what they need to achieve within a reasonable amount of
time? Sometimes that could be something like saying no one
should wait more than 5 minutes to find out, you know, if they
have an appointment or nobody--and it is not only the IT, it is
also, you know, nobody should wait on the phone for more than 5
minutes before talking to a person at wherever, right?
If you center the humans in the process, in thinking about
the outcomes, the end goal of these systems is to serve people.
Are they served efficiently, correctly, and accessibly?
Mr. Self. That is okay. That is below congressional level.
You know, that is in the VA level. I guess I will continue to
wrestle with it.
Ms. Oakley, would you care to comment? Please.
Ms. Oakley. Yes, I would just say it may not be necessarily
below congressional level because you can require things of VA.
When she is talking about human-centered, our work on product
development of leading innovative companies shows that the
customer is involved from the very beginning and the customer
is providing input and feedback and responses that then shape
the continuing effort.
It is not just do we get to the end and does it meet their
needs, right? That would be too late to do anything about it.
When you are talking about what could be required of VA, it is
how does VA document in any of its acquisitions how the end
user, the customer, however you want to define it, is involved
in the process all along the way to ensure that we are
establishing the right requirements, we are focusing on the
correct minimum capabilities that we can get out there first
based upon what is needed, and that in the end it is going to
provide a useful outcome to those folks.
Mr. Self. I would like for you to provide the ideas to my
staff, because I do not want Congress in the project management
business. At the same time, I am wrestling with how we get to
this discussion. How do we get there?
Mr. Chairman, I yield back.
Mr. Rosendale. Thank you very much, Representative.
I would like to go and dig into a little bit more about the
VistA system. I understand that you are not necessarily, you
know, professionals on VistA, but you sure understand the basic
technology.
Does either one of you know how long this VistA system has
been in place?
Ms. Oakley. I think at least since the eighties.
Mr. Rosendale. Since the eighties. Okay. That one has
certainly been tried and tested, shall we say. A lot of the--a
lot of the warts have been worked off of it.
Is it accurate to say that all of the current VA facilities
use this system around the country, including--including those
that are currently being forced to use the Oracle system? Is it
true that they still have VistA running in background?
Ms. Oakley. I believe that is true, yes.
Mr. Rosendale. Okay. I mean, to me, that should be
startling to everybody in this city to think that we have a
system that was developed in the eighties that is basically the
primary system that we are using for electronic health records,
and yet we are spending billions of dollars with another
contractor, with another vendor that has not provided us the
results that we need. Yet this one is still working for a
fraction of the cost.
Does anybody know what components are necessary to fulfill
that seamless, operation that we were talking about earlier so
that everybody can still communicate with each other? Do you
know what components are missing from VistA that we were trying
to achieve by bringing in Oracle to replace them?
Ms. Oakley. I do not know that specifically. I can get that
answer for you for the record. I do know that we have talked
about the medical supply chain and the challenges with using
those types of systems for the medical supply chain, and it is
antiquated. It does not provide the right kind of data. It does
not give real-time information to be able to manage a supply
chain at any given medical center, let alone nationwide like we
saw in the COVID-19 pandemic.
There are various aspects of VistA, I think, that affect
the efficiency and effectiveness of operations within the
medical centers from finance to contracting to you name it.
Mr. Rosendale. The very prescription drug issue that you
spoke of earlier, Ms. Schank, it is my understanding that that
is one of the main components that we were having problems
with, the Oracle system, that the pharmacies within these
veterans facilities, they were having major, major problems
with that and risking lives and the health of our veterans.
Ms. Schank. I am not aware of exactly what the issues are
with VistA that requires the upgrade or the replacement. I do
know that if you have an immensely complex system that users
like, you want to try to keep as much of that functionality as
you can. The idea that a contractor would come in and just
throw out a system that has good pieces to it is sort of
anathema to how you develop technology.
Mr. Rosendale. Maybe we should actually be talking to the
farmers that have a plow instead of the people that are using a
pencil and finding out exactly which components are missing and
maybe we could introduce those into the VistA system and we
could actually deliver something that is working for not only
the facilities but a lot better delivering benefits and
services to the veterans.
Ms. Schank. Absolutely.
Mr. Rosendale. Okay. Thank you very much.
I will recognize Representative Cherfilus-McCormick for 5
minutes.
Ms. Cherfilus-McCormick. Thank you, Mr. Chairman.
Ranking Member Takano has introduced, and I have
cosponsored, two bills this Congress, one which would require
independent verification and validation of large IT programs
like FMB and supply chain, another which would establish an
undersecretary for management at the VA.
Mr.--Ms. Oakley, you testified at our legislative hearing
on both of these bills, which were very helpful, and we look
forward to incorporating GAO's comments into our final version.
Has GAO identified any other issues with the VA acquisition
management that needs to be addressed with legislation?
Ms. Oakley. At this time there are not any outstanding
matters to Congress that we have identified with regard to
acquisition management. I think we need to see VA take action
itself first to really address those root causes of its
acquisition management problems. Those root causes are outlined
in that action plan I mentioned. Again, taking action on the
action plan will get VA moving in the right direction and then
I think illuminate any areas that the Congress needs to act on
that VA can not do itself.
Ms. Cherfilus-McCormick. Given recommendations that are
outstanding with VA around acquisition management, what should
be the priority for VA to address making sure near-term
programs like FBT and supply chain modernization do not suffer
the same fate as previous attempts?
Ms. Oakley. Yes, so I am just going to put a little plug
in. I think our priority recommendations letter to VA will be
issued next week, and in that letter we outline what the
recommendations the Comptroller General of the United States
thinks are the highest priority for VA to address.
There are several that are related to acquisition
management, including addressing the acquisition workforce
challenges that we identified in a report last year, and
developing and executing the comprehensive supply chain
management strategy as well. Then we have some recommendations
related to actually assessing the acquisition function within
VA to understand where improvements need to be made.
Those are just three that are outlined in that priority
recommendations letter.
Ms. Cherfilus-McCormick. Thank you so much, Mr. Chairman. I
yield back.
Mr. Rosendale. I thank the representative.
Representative Self.
Mr. Self. Thank you, Mr. Chairman.
As we know, the EHR system is paused. I have asked
previously the cost to upgrade VistA versus Cerner.
Can you walk me through the steps, Ms. Oakley, to cancel
the Cerner contract? How difficult is that? What are the steps?
What would that take?
Ms. Oakley. Well, the government can cancel any contract at
its convenience at any time. The process for doing that, I
think, would obviously involve some pretty significant steps,
but it is well within the government's rights to be able to do
that for sure.
I can get you for the record the actual process and what
that would take for sure just to give you a little bit more
detail on that.
Mr. Self. Is there a huge buyout?
Ms. Oakley. There is usually a limitation of liability on
every contract that should be outlined in that contract to give
a sense of what the government would be on the hook for. I am
not even going to make a guess because I have not looked at
that EHRM contract recently. I can get back to you on what that
termination liability might be.
Mr. Self. Okay. Thank you, Mr. Chairman. I yield back.
Mr. Rosendale. Thank you very much, Representative.
Thank you very much. The panel is excused from the witness
table.
With that, I will yield to Ranking Member Cherfilus-
McCormick for her closing statement.
Ms. Cherfilus-McCormick. Thank you so much, Mr. Chairman.
I appreciate the testimony and answers from our witnesses
this morning. I think it is--I think it was a good start to a
much larger discussion to continue with VA this Congress.
Competition in contracting is a positive thing, and we need
to incentivize this at the VA. Where dollar amounts have
increased, they have continued to see poor performance.
Employees, veterans, and Members of the Congress have had
enough of the current lack of success, and I look forward to
working together to ensure we have more positive outcomes in
the future.
Thank you so much for your testimony.
Mr. Chairman, I yield back.
Mr. Rosendale. Thank you very much, Representative
Cherfilus-McCormick.
I want to sincerely thank our witnesses for joining us this
morning.
Ms. Schank, I know you had to travel to get here, and I
appreciate you putting that effort forward.
Ms. Oakley, you have appeared before this committee roughly
a dozen times, and I certainly appreciate your dedication.
I expect the VA to provide the best-qualified witnesses,
upon request, when we resume this discussion of this topic in
the future. I intend to stay focused on these issues for as
long as I have the privilege of chairing this committee.
Not only are the veterans unable to see the results from
billions of dollars that are invested in IT, but the VA is
becoming an increasingly difficult place for most companies to
do business.
Government contracting is supposed to be a level playing
field. However, the VA's become a notoriously difficult,
arbitrary client. We want the best companies with the most
innovative ideas working for the VA. Those companies are hardly
hurting for work, and I am sorry to say they are looking
elsewhere.
This situation is guaranteed to produce more bad outcomes,
and I expect to hear how the VA's IT and contracting leaders
are going to fix it.
With that, I ask unanimous consent that all members have 5
legislative days to revise and extend their remarks and include
extraneous material.
Without objection, so ordered.
The hearing is adjourned.
[Whereupon, at 9:22 a.m., the subcommittee was adjourned.]
=======================================================================
A P P E N D I X
=======================================================================
Prepared Statement of Witnesses
----------
Prepared Statement of Shelby Oakley
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Prepared Statement of Hana Schank
Thank you for holding this hearing today, and for inviting me to
offer my expertise on the shrinking vendor base for IT contracts, and
why this makes it increasingly difficult for the federal government to
serve the American people the way policymakers intend.
I've been a technologist for my entire career, first in the private
sector, and now in the public sector. I started out in the `90's,
working for Andersen Consulting, building the large-scale systems that
are now in various states of decrepitude, across both the public and
private sectors. After launching some of the first websites and systems
for Fortune 500 companies, I started an IT consultancy, which I ran for
15 years.
At that time I was exactly the kind of small business that the
government should want to work with - low-cost, efficient, and
specialized. But while I landed some small government contracts, I
never even considered bidding on a major piece of work. I'd heard that
government RFPs required a huge lift to respond to, and the few that I
did consider bidding on were clearly written by people who didn't know
what they were asking for.
Because I ran a women-owned business, I was often asked to be a sub
on a contract, by larger businesses who wanted the competitive
advantage that gave them. But when these businesses approached me, it
was never to do anything that would have a large impact on the project.
It was usually a hand-wavy, ``we'll throw you a bone''--type of
arrangement. I almost never agreed to sign on as a sub.
After Healthcare.gov failed to launch (Oracle was one of the major
contractors), I joined the United States Digital Service (USDS), hoping
to bring my hands-on technical skills to government. But like many
USDS-ers, I immediately grasped that the root of all government system
failures was not a lack of hands-on tech talent, but procurement.
When I was at Andersen in the `90's, we were one of a large pool of
companies competing for these contracts. But over the years, the
industry has consolidated - Andersen Consulting has become the mega-
company Accenture - and the requirements that government lays out for
contracts have become increasingly onerous, requiring specialized teams
who do nothing but respond to government proposals. As a result, only
the very largest firms have the capability to compete. According to the
GAO, at VA alone, the number of IT contracts increased by $3B over the
last four years, while the number of contractors fell by more than 25
percent. As of 2021, only 30 contractors accounted for 75 percent of
all of VA's IT obligations.
These big players often don't play fairly, gaming the system to
elbow smaller companies out of the way, often by acquiring them. They
also account for every single one of the headline-grabbing large-scale
Federal IT disasters.
At USDS I worked on fixing ELIS, the Nation's immigration case
management system. USCIS started working on ELIS in 2005, during the
George W. Bush administration. Seven years into development the system
was such a mess that it had to be scrapped. The development of
gargantuan systems often takes much longer than anyone expects and
involves multiple types of failure. These failures are not limited to
the United States. In 2011, the UK killed a 4.6B system
that had been in development for nine years, meant to streamline the
national health system's record keeping. In 2019, after 9 years of work
and $2.2B, the Canadian federal payroll system's migration to a new
platform failed so spectacularly that thousands of Canadians went
without pay for weeks.
By 2015, when USDS came onto the ELIS project, they found that the
contractor - IBM - had seized an undue amount of control over the
design of the system. They'd designed ELIS so that it relied heavily on
IBM products, even when those products did not benefit the system. As a
result, ELIS took five times longer than paper to move applicants
through..
Today, ELIS is stable and functional. The way USCIS got there was
three-fold:
-First, they extricated themselves from their monolithic
contract with IBM
-Next, they carved out smaller, discrete projects with clearly
articulated outcomes goals. These contracts mostly went to
smaller IT vendors.
-Finally, USCIS built an in-house team with technical expertise
who could correctly procure and manage IT projects - something
that had been lacking, and which led directly to ELIS's initial
failure.
Coming out of the work on ELIS, USDS created the FLASH procurement
vehicle to encourage a more diverse set of vendors to compete. FLASH
unfortunately failed, but other Federal agencies later refined the
process successfully, including digital services teams at CMS, OPM, and
the VA.
Veterans Affairs has a successful procurement vehicle called CEDAR,
which pre-qualified four service-disabled, veteran-owned small
businesses to receive task orders for agile IT delivery support in
application development, product management and user research. Larger
vendors' business models incentivize writing proprietary software into
any system they develop, whether those pieces of functionality serve
the end user or not. By breaking contracts into smaller pieces, it
becomes more challenging for large vendors to game the system this way.
CEDAR has a task order limit of $10M, which the VA has learned is too
low. As a result, they are working on a new vehicle with no limit,
called SPRUCE.
Consolidation and a lack of competition is always bad for the user,
and when it comes to federal IT contracting, the end user is the
American public. I want to make clear that these bad contracts are not
because people in government aren't trying hard enough. Often,
government has no choice but to go with a vendor that has failed
repeatedly. After the failed rollout of a system for public assistance
built by Deloitte, the state of Rhode Island renewed the company's
contracts. Similarly, VA may choose to proceed with Oracle Cerner
because it is easier to keep building the boat you were working on than
to turn it into a plane.
Going forward, it needs to be easier - and incentivized - for
agencies to write smaller contracts and hire a smaller, more diverse
pool of IT vendors. Additionally, vendors must not be rewarded for
failure. Today, most contracts simply require that a product be
delivered, not that it be usable or efficient. Government must demand
better.
Finally, I want to touch on cost savings. When IT fails it is
expensive, and we see cost overruns into the billions of dollars.
Bringing senior tech talent in-house, while potentially expensive as a
line item, would likely lead to tremendous cost savings as there would
be people who could advocate for building the right thing the right way
the first time. VA has already started this process. There are talented
tech teams within the agency who are working to get it right, but these
teams are small, and are often not consulted when it comes to large
scale procurements.
I'd like to close with a positive story. Dr. Williams is a VA
psychiatrist in a large city. When she first came to VA, she was
delighted to find that VA's case management system made her work
easier, unlike her previous experiences using Epic and Cerner. At VA,
care is provided holistically, which means practitioners work together
across specialities to provide the best care for veterans, and the IT
system facilitates this. Early in her career at VA, Dr. Williams met
with a veteran who was in the process of moving, as vets often do, so
she had put all her belongings in cardboard boxes, along with her
medication. As a result, she'd stopped taking her medication and had
become actively psychotic. Using the VA's existing case management
system, Dr. Williams was able to look up the patient's medication
history, immediately place an order with the onsite pharmacy for the
missing medication, arrange a home health nurse to assess the veteran's
new home and create a system that would remind her to take her
medication, enroll her in an outpatient program so she could get
evaluated the next morning, and connect with a social worker who was
then able to arrange a van to transport her to a clinic near her home.
Because the system allowed Dr. Williams to do all of these things in
less than 30 minutes, the veteran was able to stay out of the hospital,
and get on a path to health.
This is how government systems should work, and can. But you can't
get there with a small pool of vendors who take advantage of
government's lack of technical know-how to line their own pockets. VA
has the chance to do something wonderful for our Nation's Veterans.
They should take every opportunity to get it right, save money, and
serve Americans the way it is intended.
Thank you.
[all]