Skip to content

Hearing on the Use of Technology to Better Target Benefits and Eliminate Waste, Fraud, and Abuse

April 19, 2012











APRIL 19, 2012



Printed for the use of the Committee on Ways and Means

GEOFF DAVIS, Kentucky 


RICK BERG, North Dakota
TOM REED, New York
TOM PRICE, Georgia
DIANE BLACK, Tennessee


JENNIFER M. SAFAVIAN,  Staff Director and General Counsel
JANICE MAYS, Minority Chief Counsel






Panel 1:

Ms. Donna Roy
Executive Director, National Information Exchange Model (NIEM), U.S. Department of Homeland Security

The Honorable George Sheldon
Acting Assistant Secretary, Administration for Children and Families, U.S. Department of Health and Human Services

Panel 2:

Mr. Robert Doar
Commissioner, Human Resources Administration, New York City

Ms. Ginger Zielinskie
Executive Director, Benefits Data Trust

Mr. Darryl McDonald
Executive Vice President, Teradata Corporation

Mr. Campbell Pryde
President and Chief Executive Officer, XBRL US



Thursday, April 19, 2012
  U.S. House of Representatives,
Subcommittee on Human Resources,
Committee on Ways and Means,
Washington, D.C.

The Subcommittee met, pursuant to notice, at 10:03 a.m. in Room 1100, Longworth House Office Building, Hon. Geoff Davis [Chairman of the Subcommittee] presiding.

[The advisory of the hearing follows:]


     *Chairman Davis. Welcome to the most recent installment in the Subcommittee’s ongoing efforts to promote the standardization of program data within the cross public benefit programs.

     One of the key purposes for today’s hearing will be to review the progress that has already been made with relevant agencies and outside experts, as well as consider next steps.

     It was just 13 months ago that this Subcommittee held its first hearing on the use of data matching to improve customer service, program integrity, and taxpayer savings.

     One of the initial factors driving this effort was the rising level of Federal improper payments which peaked at $125 billion in fiscal year 2010.

     It was about much more than that.  It really became about making Government work smarter, faster, and more efficiently for both beneficiaries and taxpayers.

     We should expect more from Government.  We should expect it not to operate from an era before the personal computer existed, but one from the 21st Century using integrated technologies and information systems, not unlike the private sector uses today in competitive industry, and also most consumers do within their homes today in terms of how they manage their finances and their purchases.

     Companies like Google and Facebook have grown based on their ability to help customers search, access, share and interpret data.

     It is long past time for Government to use those same sorts of tools to improve the services we offer to help the less fortunate and the taxpayer dollars spent on them.

     In September, the President signed our bipartisan and bicameral Child and Family Services Improvement and Innovation Act, which codified into law for the first time in history our data standards language for all child welfare programs.

     This was followed by another example of bipartisanship when the same language was included in the Middle Class Tax Relief and Job Creation Act of 2012, and applied to the Unemployment Insurance and Temporary Assistance to Needy Families programs.

     We hope to expand this effort through H.R. 4282, which was recently introduced by Representative Berg and Ranking Member Doggett and co‑sponsored by the entire Subcommittee.

     This bill would apply the same data standardization language to the Child Support Enforcement Program under our jurisdiction.

     With enactment in three programs already, Congress has sent a clear bipartisan message that it wants to improve the operation of public benefit programs through data standardization.

     H.R. 3339, the Standard DATA Act, would apply this provision more broadly to all programs within the Subcommittee’s jurisdiction.

     We believe this bill is firm but flexible so that those with the most knowledge of how these programs operate can come together to craft a better path forward.

     We also need to hold the Executive Branch accountable for its role in modernizing the operation of Government programs.  That is why we have asked several key agencies, the Department of Health and Human Services and the Department of Homeland Security, to testify at this hearing about their progress and plans for the future.

     We plan to continue expanding this effort across other programs that serve similar populations so we can achieve a truly beneficiary-centered approach to how these programs are run.

     We also need a more complete view of how many people receive benefits under multiple programs, which will help to better target benefits to people in need, as well as improve understanding of program effectiveness, especially given our budget challenges ahead.  We simply have no choice but to take these steps to make sure programs operate more efficiently and to provide benefits to those most in need and only to those in need.

     We look forward to all testimony from both of our panels. We also look forward to continuing to work in a bipartisan manner to improve how public benefit programs serve the children and families who depend on them, while assuring taxpayer dollars are used efficiently and effectively.

     Now I would like to turn the microphone over to my friend and partner in this effort over the last year and a half, the Ranking Member, Lloyd Doggett from Texas.

     *Mr. Doggett.  Thank you, Mr. Chairman.  Thank you for the great personal interest that you have shown in this whole data matching effort and for the partnership we have enjoyed in working as you mentioned on child welfare legislation on data matching for Temporary Assistance for Needy Families and for Unemployment Insurance Programs.

     It is important to improve the Administration of public programs through technology.  It benefits taxpayers and it benefits those who rely on those programs.

     Abuse of Government programs, whether by multinational pharmaceutical companies or a person collecting unemployment insurance is never acceptable.  It denies the benefits that are needed to others, and it unjustly treats the taxpayer.

     I hope we can continue to make progress on standardizing data that would include incorporating existing non‑proprietary standards such as XBRL, that has the support of the Texas Society of Certified Public Accountants.

     I think the better use of data matching and data sharing across agencies improves our efforts to reduce fraud and abuse.  We want to get these benefits to those who need them and who are eligible for them.

     My interest in this is not only about program integrity but seeing that programs satisfy their legislative purpose.

     This is particularly true when there are so many families in my home State of Texas and across the country that struggle to maintain their footing.

     In Bexar County, for example, one in four children are poor.

     We must find a way to ensure that our most vulnerable citizens are provided with the assistance that they need to weather economic storms, and improving the ability to do this through better data is one important aspect of outreach.

     Those efforts are now underway in Philadelphia, through a collaboration between the State of Pennsylvania and the Benefits Data Trust, a non‑profit organization from whom we will hear this morning.

     The Benefits Data Trust has been able to use state data to determine if an individual who is receiving assistance from a certain program is likely eligible for assistance in another.  It is true that there are too many duplicative programs, but the total benefits provided by all of them, if delivered efficiently in the most effective way, probably do not begin to meet the needs that are out there.

     Identifying and connecting low income individuals to the services for which they are eligible allows the organization to work in a cost effective manner while also dramatically improving the outcomes of those individuals who are to receive benefits and our economy in the process.

     I think this is particularly relevant to the current attention on the delivery of food assistance through SNAP.

     A report released recently by the Department of Agriculture found that SNAP benefits helped to reduce national poverty in a significant way.

     These benefits play an important role in providing critical assistance that I saw recently at the food bank on the west side of San Antonio.

     Mr. Chairman, I look forward to hearing from today’s witnesses and continuing our work together to improve the outreach, the efficiency, and the effectiveness of these vital public initiatives through the use of technology.

     Thank you so much.

     *Chairman Davis.  I thank the gentleman.  I want to remind our witnesses to limit their oral statements to five minutes, please.  However, without objection, all of the written testimony will be made part of the permanent record.

     Our first panel this morning, we will be hearing from Ms. Donna Roy, Executive Director, National Information Exchange Model or NIEM, U.S. Department of Homeland Security.

     The Honorable George Sheldon, Acting Assistant Secretary, Administration for Children and Families, U.S. Department of Health and Human Services.

     We happen to stand together behind President Obama as he signed into law that very first piece of data standardization language on September 30 of last year.

     We are excited to have you here.

     Before we move on to our testimony, without objection, I would like to change pace for one moment and recognize Dr. Price, Chairman of the Republican Policy Committee, who is a member of the Subcommittee, and able to be here with us today, who would like introduce a witness from his home state who will speak on our second panel today.

     *Mr. Price.  Thank you, Mr. Chairman.  I apologize for going out of order and I appreciate the indulgence of the Chair.

     I wanted to take this opportunity, not knowing the schedule that is going to break as the morning proceeds, to introduce Darryl McDonald, who is going to be on our second panel.

     Darryl is the Executive Vice President of Teradata Corporation, a wonderful corporate citizen and job creator in Georgia and in my District.

     Teradata is responsible for providing strategic direction for their products, solutions and services, and is among the world’s largest companies focused solely on analytics and data warehousing.

     Teradata is a growing business with over 8,000 employees worldwide.

     Mr. McDonald has deep roots in Georgia, attended the University of Georgia, and can give a good “go Dogs” with the best of them.  He currently lives in the 6th Congressional District, and his testimony on the second panel today will highlight how both business and Government can benefit from technology and ultimately save the taxpayers money.

     I am pleased and honored to introduce and welcome Darryl McDonald to our committee.  Thank you, Mr. Chairman.

     *Chairman Davis.  Thank you.  With that, Ms. Roy, please proceed with your testimony.



     *Ms. Roy.  Chairman Davis, Ranking Member Doggett and members of the Subcommittee, thank you and good morning.

     My name is Donna Roy.  I am the Executive Director of the Information Sharing Environment Office in the Department of Homeland Security’s Office of the Chief Information Officer.

     I also serve as the Executive Director of the National Information Exchange Program or NIEM.  I have had the privilege of holding this position for the past three and a half years.

     I appreciate the opportunity to discuss with you today how NIEM helps Government agencies enhance mission performance, gain efficiencies, and reduce costs associated with exchanging information across IT systems.

     NIEM is a federally supported Government‑wide initiative that helps communities of people with common interests connect and exchange information in order to successfully and efficiently accomplish their missions.

     NIEM is not a system or a database, nor does it transmit, store or engage in operational data storing.  Rather, NIEM provides the tools, the training, and importantly, the community driven support to assist users in adopting a standards based approach.

     It is fitting that we are having this discussion today on the seventh birthday of NIEM.  The program began on April 19 in 2005 when the CIOs of DOJ and DHS signed an agreement to establish the interagency program.

     This action brought to fruition the work started by half a dozen or so practitioners at the state and local level of government who had worked to build a national model for sharing law enforcement and homeland security information.

     Today, all 50 states and 18 Federal agencies are committed to using NIEM in some capacity and at varying levels of maturity.

     NIEM is now recognized and used by international partners, such as Canada, Mexico, and member countries of the European Union.

     Despite NIEM’s growth over the past seven years, the program has maintained at its core a strong commitment to community involvement.

     In my opinion, this is one of the main reasons why NIEM stands apart as an example of a Government collaboration at its best.

     Whether NIEM is involved in a situation to protect citizens, respond to disasters, promote public health, or provide needed support services to children and their families, the lack of standardization in information exchange across this broad landscape of systems, agencies, and jurisdictions creates challenges in providing effective and efficient citizen services, often leading to fraud, waste and abuse.

     The challenge is clear.  How do we connect the wide array of systems across the whole of Government while supporting an appropriate privacy and security framework flexible enough to accommodate the diverse laws, regulations and policies across the United States.

     The majority of the work within NIEM is accomplished within a strong and active community of volunteers at the state and local level.

     The NIEM community stewards a common vocabulary and a mature framework to facilitate information exchange.  NIEM breaks down agency stovepipes and creates opportunities for agencies to share information quickly and effectively without rebuilding systems.

     As an example of success, NIEM has helped the State of Colorado connect child support systems with judicial processing systems, improving the speed of administrative case processing and execution of child support orders.

     The project reduced manual data entry, eliminated most of the paper forms, and improved data reliability.  In effect, NIEM helped Colorado provide faster, cheaper, and better services to its citizens.

     In the difficult fiscal times facing the Government today, it is core systemic improvements that will save scarce financial resources, improving the effectiveness in Government and ultimately making our country safer.

     NIEM is one of these rare systemic improvement opportunities.

     As a former Marine who possesses a sincere passion for good government, I am continually impressed and energized by the level of support across the NIEM community.

     On a more personal note, as the daughter of a woman who aged out of the system, and as a former foster mom, I am particularly touched by the positive changes NIEM is making in the lives of the most vulnerable in our society, which are the children and the families that support them.

     Again, I appreciate the opportunity to testify before you today, and I look forward to answering any questions you may have.

     [The prepared statement of Ms. Roy follows:]

     *Chairman Davis.  Thank you, Ms. Roy.  We wish a happy birthday to NIEM as well.

     The Chair now recognizes Secretary Sheldon for his testimony.



     *Mr. Sheldon.  Chairman Davis, Ranking Member Doggett, members of the Subcommittee, thank you for the opportunity to testify about the development of standardized data exchanges and the use of technology to better target benefits and eliminate waste, fraud and abuse.

     This is my second opportunity to testify in front of this Committee.  I testified two years ago as Secretary of the Department of Children and Families in Florida on this Committee’s efforts to provide expansion of the IV‑E waiver concept.

     I thank you for the legislation that passed this year.

     I would like to acknowledge the leadership of Chairman Davis in this area.  Like you, I believe that these efforts will lead to better targeting of benefits to eligible households, and at the same time, will reduce fraud, waste and abuse.

     Much work is occurring in this regard already.  Clearly, much more needs to be done, and the hearing today furthers this critical conversation.

     I have a deep personal commitment to the effort you are promoting because I experienced firsthand as a state administrator what a difference improved data sharing can make for those on the front line delivering services.

     We all recognize the problem that we are trying to tackle. Valuable information which could support more accurate eligibility and benefit determinations, and more thoughtful case planning is siloed among multiple systems across related but not fully integrated programs.

     Even in cases in which organizations are allowed to share information, uncertainties about legal requirements, cultural differences, and misperceptions about privacy requirements too often stymie efforts to exchange information, even when the benefits are obvious.

     We face this situation in the State of Florida, for instance.

     The first point I would like to emphasize is the solution is not purely a technical one. The initial hurdle is to promote a new way of thinking, a cultural exchange that promotes the sharing of information.

     In Florida, we worked across eight state agencies to develop what we called a “Children and Youth Cabinet Information Sharing System.”

     Rather than develop the initiative from start, we piggy backed on the success of another initiative that the court system had in place for severalyears.

     The court system was a multi‑agency data sharing system.  In this manner, we not only benefitted from their technology platforms, but we were also able to incorporate the lessons that they had learned.

     I emphasize our use of the piggy back in Florida because we are taking that same approach at ACF.

     I will talk more about NIEM later, but as you know, we know a good idea when we see it.  It does not make sense to pay to replicate systems we already own or have invested in heavily.

     I believe that the Committee’s initial emphasis on data exchange is wise.  While I am not discounting the need for investment to support systems at the Federal, state and local levels, there are significant returns that we can realize by improving data sharing within the systems that we currently have.

     Implementing our system in Florida was a big step forward.  It saved time, saved labor, reduced errors, and provided real time access to information.  It allowed participating agencies to maintain the control and security that they wanted over their own data.

     This is another point I want to emphasize because it demonstrates that significant gains can be made without compromising privacy, which I know is a critical concern of everyone in this room.

     Most importantly, it supports better decision making because it provides much needed real time up to date data and data exchange capability on a vulnerable child or vulnerable family.

     My experience in Florida also confirms that improved data sharing can lead to lower improper payments.

     In Florida, we determined eligibility for TANF, SNAP, Medicaid and Refugee payments through an integrated automated process.

     While five years ago we had one of the worst food stamp error rates in the country, in the last three years of my tenure there, we achieved the lowest error rate in the SNAP program for three consecutive years, an unprecedented achievement.  By 2010, the error rate was under one percent.

     The point I am making is that an appropriate use of technology in terms of sharing our data elements can really lead to an overall improvement of the system.

     Thank you, Mr. Chairman.

     [The prepared statement of Mr. Sheldon follows:]

     *Chairman Davis.  Thank you very much, Mr. Secretary.  I would like to move on to questions now.

     What we are talking about today is not about new data but about using existing data more efficiently and effectively.

     Secretary Sheldon, nearly every one of your programs provides a congressional or annual report, most of which come to this Committee.

     I have two here, which I will not enter into the record out of mercy for the transcriptionist, but we have some examples I would like to show you.

     We have pages up on the screen right now, quite detailed information, such as how many beneficiaries there are and how dollars are being spent.

     However, beyond those broad facts, the data is not usable in helping to improve how the programs perform.  Variance metrics, understanding the quality of work, the ability to drill down inside of those.

     For example, what I would like to share with you and members on the Subcommittee from these pages is a recent Child Welfare Annual Report on States’ Planned and Actual Expenditures, which was requested by Congress.

     I see that many of the same forms are used throughout, but one thing is how overall the data lacks consistency.

     The pages also appear to be PDF copies of the original forms submitted in paper form by states, making many of them illegible, frankly.

     What does your agency do with this information, and do you think this sort of report is useful for your staff, congressional staff, or outside experts?

     I want to qualify my remark.  I am not doing this to put you on the spot so much, but as we have discussed, we have this common problem we are trying to work around, and I would like to hear your thoughts on steps to fix this.

     *Mr. Sheldon.  I appreciate this.  The report is based on congressional direction of how we should be sharing this information.  I do think this data is important not only to us but I think it is important to the field.

     Clearly, as you have identified, these PDF files are not very user friendly.

     Bryan Samuels, who is our Commissioner for the Administration for Children, Youth and Family, actually identified this difficulty about a year ago, and began working with the staff to try to get a more uniform system of reporting from the states.

     In addition to that, it is my view that at some point, assuming resources are available, we need to be moving from hardcopy reporting that can only be shared in a PDF format to a much more user friendly, state friendly, and services friendly reporting system.

     It is something that has been identified, and I appreciate you bringing it to my personal attention, but I would report that Commissioner Samuels has identified this and is attempting to work on it.

     *Chairman Davis.  I believe the purpose of the report is to understand state variations in planned and actual expenditures on child welfare so policymakers can know what works better to protect children.  That is an important goal.

     How do you expect this and other reports will change with the enactment of our data standards provision that was signed into law?

     *Mr. Sheldon.  I think the data standardization exchanges are an excellent opportunity to begin addressing some of these issues.  As I indicated in my prepared comments, a lot of progress has been made, but there is still a lot more that needs to be done, and I think this is a good example.

     With respect to NIEM, we have established an interoperability project management team headed by Joe Bodmer, who is behind me.  I am very optimistic about the direction they are taking and the impact it could have on these kinds of reports.

     *Chairman Davis.  Thank you, Mr. Secretary.  Ms. Roy, NIEM is not a standard but a way of setting parameters for data exchange so information can flow more freely.

     I recognize this pushes beyond NIEM, but can you tell us more about the burden on the agencies and departments, especially small local governments, to transition to the structures developed by NIEM, both in the short and long term?

     Feel free to use examples outside of human services, if that would be helpful in answering the question.

     *Ms. Roy.  Thank you for asking that question, Chairman.  We believe we hold true in the NIEM program.  One basic foundational tenet is that we provide technologies, tools and support to reduce the burden for our state and local adopters to make sure the technology we employ is consumable given the reduced budgets and restraints they have at the state and local level.

     As an example, in this PDF, it would be as simple as putting XML behind it embedded within this PDF that would make this a useful document.  That would provide a very low burden to employing this type of solution.

     The Recovery Act included recipient reporting from the state and locals in the form of an Excel spreadsheet, and in the back of that Excel spreadsheet, XML data that could be used to exchange information and to add up and total up the dollars there.

     I don’t believe that everything that we do is a high cost IT implementation within the state and locals.  We have really reduced that burden by providing tools and ways for them to use XML in a manner that is at a reduced cost and a quicker time for implementation.

     *Chairman Davis.  Very good.  Thank you very much.  With that, I would like to yield to Mr. Doggett, the Ranking Member of the Subcommittee, for five minutes.

     *Mr. Doggett.  Ms. Roy, NIEM as it currently exists, I know, has been instrumental in facilitating the exchange of data.  What are some examples, additional examples, of how it is currently being used in the Federal Government?

     *Ms. Roy.  NIEM has realized a significant increase in adoption in the Federal Government.  In the Department of Homeland Security, where my day job is, I can provide some examples.

     NIEM is being applied in the Disaster Assistance Improvement Program from FEMA to harmonize the 60 forms that one could apply for in a disaster situation.

     Amazingly enough, 80 percent of the data on those 60 forms is the same, and yet, we make our citizens fill out those forms again and again.

     The FEMA Disaster Assistance Program is aimed at harmonizing that and allowing citizens to fill out most of that information once and pass that information to the agencies to ensure that those forms are pre‑populated.  To me, that’s good Government.

     The Citizenship and Immigration Services in the Department of Homeland Security has applied the use of NIEM in the E‑verification and self check programs, allowing faster citizen services in a critical service that we provide.

     Outside of the Department of Homeland Security, the Department of Justice has a strong program around index and law enforcement information sharing.

     True banner best practice child from the beginning of the NIEM program.

     There are really great cases at the state level based on some at the Federal level.  In Colorado, New York City, and Massachusetts, NIEM is being applied in child and family services.  The really great part about NIEM is that people can use it.  It is free to use.  I never have to know about it.

     I often do not know about it until I hear the good news such as the New York Times article around the New York City’s HHS‑Connect successes.  They were able to provide streamlined citizen services to their citizens

     As an example, an 18 year old mother of an one year old son who arrived in the city’s homeless intake center.

     She had been removed from her own mother’s house by ACF in September because of neglect.  By typing her name into this new interface, the case worker was able to within minutes find her birth certificate, her baptism certificate, and her mother’s driver’s license to help her document her identity and housing history which homeless services requires.

     That took about 45 minutes out of a six hour intake process because of the work New York City’s HHS‑Connect has done to bring together their information using the NIEM standard.  Again, we did not know they were doing it.  We had no anticipation that NIEM would be used in human services when we started the program.

     These are all great sorts of use cases around community adoption and the community involvement in the use of a program originally started for law enforcement and homeland security information.

     *Mr. Doggett.  Thanks so much to both of you for your leadership.  We have a vote underway.  I will hold the rest of mine.  Thank you.

     *Chairman Davis.  I thank the gentleman.  I apologize to our guests and to our witnesses that a vote intruded upon the proceedings right now.  We will be back in 20 minutes.

     I would ask Ms. Roy and Secretary Sheldon if you would please consider remaining.  I believe some of our members do have some questions for you before we move on to our second panel of witnesses.

     Appreciate you being here today very much, and very excited about the progress we have made over the last year and a half working together.

     With that, the Subcommittee stands in recess for 20 minutes.


     *Chairman Davis.  The hearing is back in session.  We will continue with our questioning of our first panel.  I would like to recognize Dr. Boustany from Louisiana for five minutes.

     *Mr. Boustany.  Thank you, Mr. Chairman.  This is a very important hearing, and I know you have had a long interest in this interagency cooperation and how we can use data to drive this.

     Ms. Roy, you have been involved in all this, trying to push this from the inside.  Why is it so hard to reach consensus on standardization?  Please describe some of the difficulties you have encountered going forward with all this.

     *Ms. Roy.  Thank you, Congressman, for asking me that.  If it were easy, I believe it would not be so hard, such a big push to get people to adopt.

     Most of the objections we get are related to education and the time to adopt something new.  A lot of misperceptions, it is hard to use or hard to adopt another standard.

     We believe when we have a cogent conversation around the re‑use potential, those conversations get easier to have.  Sometimes we have the objections of not invented here, it is not my XML standard, as an example, so it is hard for me to sort of diverge and listen openly to a better way of doing business.

     For the most part, it has been about outreach and making sure we have enough outreach activities out there to apply what we think are the good aspects of a program like NIEM out there to 3,000 counties, 50 states.  As an example, in Homeland Security, 18,000 police departments.  It is a pretty big country, a lot of people to get to.

     *Mr. Boustany.  Magnitude of order.

     Secretary Sheldon, do you want to comment on that?

     *Mr. Sheldon.  Yes.  I saw this at a state level, where the agency that I ran, mental health, substance abuse, TANF, child welfare, domestic violence, and everyone is comfortable with their little silo; they are safe in it.

     I think as I indicated in my testimony, we know how to do the technology, quite frankly, but it is the culture of sharing that has to happen.

     I have been meeting with several state commissioners.  I think there is a growing realization at the state level as there are declining resources that there is a benefit to data exchange standardization.

     I met, for instance, with the Secretary of the Department in Virginia just last week.  They are doing some extremely innovative things in Virginia, and I think other states are too, but Virginia may very well be one of those states that could be an example to other states.

     What they are attempting to do is not just add standardization but a growing sharing of information across lines, and  ultimately, I think this is going to be where we are headed when we get to the point where we are not just dealing with the human services domain, but we are also dealing with the health domain as well as with the education domain.  Those three are so intricately linked.

     *Mr. Boustany.  Right.  It is interesting.  There is an irony I have experienced in my time in Congress.  Late in 2005, I went to Iraq.  On the way back, I stopped in Landstuhl.  I was talking to the chief medical officer at the hospital in Landstuhl where our wounded warriors were evacuated.

     What was interesting was the fact that we had good data for battlefield to Landstuhl, but when our guys left the Department of Defense and went over to the Veterans Administration, which consisted of a whole different database, there was a lack of communication and lost information, and a lot of frustration on the part of our veterans.

     I returned in 2008 to see what changes were made, and basically, nothing had changed.

     The irony is here we are, an Administration driven to implement health information technology to improve quality and collaboration, integration of health, and yet our own departments cannot seem to get it together.

     I share Chairman Davis’ intense interest in this issue because as he said in his opening statement, with our budgetary constraints and all the other things we are dealing with, this should be a no‑brainer, albeit a very difficult task, given the magnitude and cultural issues.

     That is why I asked the question, to try to drill down a little further into what specific problems you are encountering going forward on all this.

     If you could continue to work with us on these things and help us understand where the road blocks are, what specifically is occurring.  We can pass legislation, but actual implementation and getting this right is a whole different story.

     With that, Mr. Chairman, I would yield back.  Thank you.

     *Chairman Davis.  I appreciate your comments.  The Chair recognizes Mr. Reed for five minutes.

     *Mr. Reed.  Thank you, Mr. Chairman.  Thank you to our witnesses for testifying today.

     Ms. Roy, I am going to direct a lot of my conversation to you, as I do not think we have ever had a member of the Homeland Security team come here and testify before on human resources issues and others.

     I really want to delve into the concerns that NIEM potentially is out there for purposes of intelligence gathering.

     Can you dispel for me any of the issues that people are expressing, that this is confidential information data that could potentially be exploited or concerns about privacy and issues like that?

     *Ms. Roy.  Congressman Reed, those are very good questions.  Thank you very much for asking them.

     NIEM is just a format for data as it moves.  It does not imply that you are moving it anywhere.  It does provide a common vocabulary for someone who wants to send data to someone who wants to receive it or more appropriately, network effect of multiple people who want to receive it in a way they can understand it.

     Early when the HHS started to see the value of NIEM and have the conversations, I have read some sort of blogs out there that the CIA would get your health records, sort of had to chuckle on that one.

     NIEM does not move the data.  NIEM provides the format for which the data moves.

     In combination with NIEM are the implementations within a Federal, state, local or private sector agency who put in place security, privacy, and other aspects of the IT system that protects the data.

     NIEM is not the data itself.  It is just the format for which we agree upon sharing the data.

     The format had applicability across more than intelligence domains.  It is the homeland security, the emergency management, the justice, education, transportation.

     There are a lot of domains where Donna Roy is Donna Roy, and the description of Donna Roy as a person, I have a name, date of birth, sex, all of those things are consistently the same across multiple domains across our Government, that is where NIEM provides a significant amount of value.

     *Mr. Reed.  Excellent.  Mr. Sheldon, do you have anything to offer from HHS’ point of view?

     *Mr. Sheldon.  I think the agencies are still going to be able to control their confidentiality requirements.

     When we moved in this direction in Florida, we had what we called three joint application design sessions.

     The first two sessions were with programs and addressed what kind of information should be put up, what would be useful, and what would be useful across agency lines.

     Then in our third session, we brought in the general counsels of those agencies to address privacy issues.

     Many of the ideas that were brought up during the programmatic piece of those design sessions were included in our ultimate application.

     There were a few that did not make it because of confidentiality concerns, and then we developed memorandums of understanding between all the agencies who were participating.

     There was a memorandum of understanding between the sender agency and the receiver agency.

     I think as we move forward, we are going to have to do that kind of collaboration in order to break down some of these walls.

     I also think there is a lot of misconception about confidentiality requirements and what is confidential.  For instance, if I have a child who is in the child welfare system, and that child is in care, it would be very helpful to the state child welfare agency to know how that child is doing in school.

     Yet, that information, under the current FERPA statutes, is not shareable.

     I think with respect to data sharing you need a process where access is limited to those who have a need to know.  Obviously, when you deal with a child in the child welfare system, the extent to which you can interface with what is happening with that child in the education system, is better for the child.

     Those are the kinds of things I think on an ongoing basis we are going to have to work through.

     I am very confident from a technology standpoint we can do this.  I do think we have to break down a lot of misconceptions that exist.

     *Mr. Reed.  I would agree with that sentiment, the misconception about the ulterior motives potentially that are out there floating in the blogosphere and everywhere else.

     I appreciate that and I look forward to working with you to accomplish that.

     *Mr. Sheldon.  If I might, Mr. Chairman.

     *Mr. Reed.  Please.

     *Mr. Sheldon.  We had a taskforce in Florida after the Virginia Tech killings, and what we found is that the folks in the colleges and mental health arena were not communicating with the people who had a need to know if in fact there were some dangers.

     That is the kind of ongoing dialogue, I  think, institutions are going to have to have in order to not just make sure that we are covering fraud and abuse but we are also protecting individuals at the same time.

     *Mr. Reed.  I appreciate that.  My time has expired.  Mr. Chairman, I yield back.

     *Chairman Davis.  I thank the gentleman.  The Chair recognizes the gentleman from North Dakota, Mr. Berg.

     *Mr. Berg.  Thank you, Mr. Chairman.  Thank you for being here.  I apologize for the disruption with the voting.  It is one of those days probably more typical than not.

     As you know, this Committee has been working a lot on data standardization and the programs within our jurisdiction, child welfare, TANF, and the Unemployment Insurance.

     As you heard Ranking Member Doggett and others, we co‑sponsored a bill to do the same thing in child enforcement, and really on a global basis.

     My question for Ms. Roy is on the bigger picture, as you look around all the things that Government is involved with, are there other agencies and programs that are looking at standardization and doing some things that kind of pop out as other examples?

     *Ms. Roy.  We are in the business of being an available-to-use public service.  I wish I knew all the good news around the communities and the agencies that were adopting NIEM.

     Again, as I stated, I find out when I hear it in the press or someone bragging about a good news story.

     That being said, we are seeing a lot of positive movement in the human services arena in Colorado, New York City, in Massachusetts.  There is an exponential growth factor every time one of those states shares its best practices.

     It is a little bit of a wild card with this community driven program, but they foster conversations with a state near them or with someone else that is interested.

     We will see this network impact of a community that is very, very involved.

     At the Federal level, as I mentioned, we have almost all of the Federal agencies doing something with NIEM.  In particular, new domains, and around education, around transportation, are encouraging, because they have an impact on this cross community information exchange.

     A child is a child.  He goes to school.  The child might be part of a welfare program and might get school lunch programs.

     NIEM was built to actually sustain that cross business transactional support for data exchange.

     *Mr. Berg.  Thank you.  Just one quick follow up. How do we encourage agencies to move in this direction?  Is there something we can do?

     If you said there was one thing that could be done by Congress to encourage that, what would it be?  Other than more money.

     *Ms. Roy.  We appreciate support in getting the word out, that NIEM is a tool for your constituents to use for better IT connections, for better sharing.

     We appreciate increased outreach on this.  We are getting great support from the Federal CIO Council for the Federal agencies.  Every time we see a Federal agency adopt, we see a ripple effect in the state and local agencies that also exchange information with the Feds.

     I would say continued outreach and understanding that this is a valuable program that can help significantly increase mission performance.

     *Mr. Berg.  Thank you.  I will yield back.

     *Chairman Davis.  I thank the gentleman.  I would like to welcome now a distinguished member and long-serving member of the Ways and Means Committee, the Chairman of the Subcommittee on Health, the gentleman from California, Mr. Herger.  Thank you for joining us today.

     *Mr. Herger.  Thank you, Mr. Chairman.  I appreciate an opportunity to sit on the panel.

     Mr. Sheldon, today there has been a lot of discussion about the positive of information sharing, but I want to bring it back to an issue that is easily overlooked and potentially more important.

     I want to discuss the safeguards that should be in place to ensure that the personal data of program beneficiaries is kept secret.

     As you may have heard from recent news reports, California’s Department of Child Support Services lost computerized storage devices that were being transported as part of a disaster recovery task last month.

     These devices held the personal data of 800,000 parents, guardians, and children, compromising their names, addresses, driver license numbers, Social Security numbers, health insurance, and employer information.

     The news reports of the incident are extremely troubling.  We do not have much information other than the devices went missing somewhere between Colorado and California.

     I want to make sure this security lapse is taken seriously by the Government agencies involved as by the families whose personal information has been lost.

     I would like to know what safeguards states operating child support enforcement and other Federal programs are required to have in place to protect personal data.

     Has HHS looked into the incident in California, and if so, did California’s Department of Child Services violate any of the safeguards that should have been in place?

     *Mr. Sheldon.  Congressman, I could not agree with you more.  Our child support enforcement efforts nationally have had a 15 year unblemished record in terms of protecting data.

     We do take what happened in California extremely seriously.  We immediately reached out to California when it came to our attention to determine, number one, the nature of the loss, what information was lost, and the level of data exposure.

     We are still working with them.  I have asked that we go back and look at our safeguarding protocols we are requiring of states.  I have been assured there are a lot of protocols in place in terms of making sure that states comply with confidentiality.

     I want to make sure we are doing everything we can in this area.  We will continue to work with California.

     The information that we have gotten back to date indicates, I think, there were five canisters lost.  One of those has been regained.  We requested that the state immediately identify the individuals, notify those individuals of the potential breach, and that is ongoing as we speak, I believe.

     *Mr. Herger.  Mr. Sheldon, could you tell me, what does your agency plan to do going forward to prevent this type of lapse from happening again?

     *Mr. Sheldon.  Well, the first thing that we are doing is we are working to make sure that states are meeting the current safeguard requirements that we have in place.

     We currently are doing routine monitoring of states.  We will increase that monitoring on an ongoing basis.

     I think quite frankly we also are identifying through a lot of those routine efforts if there is any unauthorized use of information.

     We are employing technical assistance teams to work with states in this arena.

     I share your concern because I think this was a serious breach, and we are trying to double our efforts in working with states and holding them accountable.

     *Mr. Herger.  Mr. Sheldon, I appreciate your taking this seriously.  Again, it is difficult to comprehend how something like this could happen.

     *Mr. Sheldon.  This was, as I understand it, an emergency disaster exercise.  To have this happen on that kind of exercise is in itself problematic.  The purpose of those exercises is just the opposite.

     I would point out the state did have back‑up information, so there was no loss on the part of the state. Exposing confidential information of individual citizens, however, I do think, is cause for serious alarm.

     *Mr. Herger.  Well, again, I appreciate that. It is very important, as you can understand, number one, to understand how this could happen in basically a mock up trial.  Imagine what would happen if there was an actual emergency.

     I think you can understand how it is very important that we check and double check and make sure this does not happen again.

     Again, I thank you very much.

     *Mr. Sheldon.  Mr. Chairman, what I would like to do is get back with you as we proceed so you know exactly what steps we are taking as it relates to this particular breach.

     *Mr. Herger.  I would appreciate you doing that.  Thank you very much.  Mr. Chairman, thank you.

     *Chairman Davis.  I thank the gentleman.  I want to thank both of our witnesses, Ms. Roy and Secretary Sheldon from the Administration, in helping us understand the issue further.

     If members have additional questions, they will submit them to you in writing, and we would appreciate it if you would also give us a copy of that so we can submit it into the official record.

     I want to thank you again.  We look forward to continuing to work with you on improving data standardization to reduce costs and improve performance in the Government.

     With that, this concludes the first panel of the hearing.  I would appreciate it if the second panel would come forward.  Thank you very much.

     Now we will go ahead and begin the second panel.  I appreciate our witnesses being here today.

     We will hear from Mr. Robert Doar, Commissioner, Human Resources Administration for New York City.

     Ginger Zielinskie, Executive Director of Benefits Data Trust.

     Mr. Darryl McDonald, Executive Vice President of Teradata Corporation.

     Campbell Pryde, President and Chief Executive Officer of XBRL US.

     Appreciate you taking time out of your schedules to join us here in Washington.

     Mr. Doar, would you please proceed with your testimony?



     *Mr. Doar.  Thank you and good morning, Chairman Davis, Ranking Member Doggett, and members of the Committee.

     I am Robert Doar, head of New York City’s largest social service agency, HRA.

     Due to the structure set up in New York State, I help to manage over $39 billion in resources in an array of programs, including TANF, SNAP, Medicaid, and child support enforcement.

     Mayor Michael Bloomberg has made it a top priority to break down silos between city agencies, so that we can improve the quality of services to the millions of participants in our programs, and to properly administer and protect city, state and Federal tax dollars.

     He drew from his work in the private sector and saw we needed a better computer system to manage eight health and human service agencies with 80 different case management systems serving more than three million recipients and reporting to different state and Federal oversight agencies.

     In 2008, we embarked upon HHS‑Connect to basically create a system to share information across multiple services, multiple agencies, so we could better serve our citizens and clients.

     A key was the strong leadership provided by the Deputy Mayor for Health and Human Services, Linda Gibbs, and the emphasis placed on this program on all commissioners in human services and city government.

     Presently, there are two major functions of HHS‑Connect that are underway.  Client access to information about benefits and services and worker access to better data and realistic view of the client.

     ACCESS NYC is the client portal.  It is a screening tool to self screen for more than 30 city, state and Federal human services benefit programs.

     Applicants can also apply on line for SNAP, school meals, and senior citizens and disabled rental increase exemption programs.

     Recently, Medicaid renewals were added to the site, and since we began, 100,000 SNAP applications have been submitted on line.

     To better assist clients on the worker side, we created Worker Connect, to share information among agency workers.  It is a secure read only web based application that allows select city workers to access a limited set of information from multiple data sources through one point of entry.

     The basic technical elements are a common client index and a document manager.  The common client index is the initial process to link the client’s identity within any of the participating agencies.

     The identifier is used to data mine details such as benefit information, case composition, and employment history, and display them to workers.

     The document management is simple but has been very useful across agencies, an electronic repository of documents submitted by HRA clients when they apply for our benefit programs.

     On the program side, the key benefit of HHS‑Connect is how the information can be used.  In our child welfare agency, they regularly use Worker Connect.  Child welfare workers use it to quickly identify and locate children and guardians they have difficult finding but are known to other city agencies.

     Often times, the reports child welfare workers receive from a central registry are missing critical identifying information, especially when they are called in by anonymous sources.

     Worker Connect has alerted staff to the identity of household members and the existence of another parent or other children who may be at risk.

     It has also been useful to our homeless agencies’ intake facilities, where families showing up at the facility often do not have easy access to much of their documentation.

     My agency also uses Worker Connect as an additional tool to identify fraud, abuse, and improper payments within public assistance programs.

     We use it for eligibility verification, to quickly and accurately identify inconsistencies, particularly unreported income, between information provided to us on applications and that which has been submitted to other agencies.

     In the future, we hope to use it to help identify patterns of potential fraud and abuse within the public assistance programs we administer.

     I need to be clear that information sharing was not undertaken lightly.  There is an over arching requirement that access to information is only granted in compliance with all applicable laws and regulations.

     We strictly adhere to Federal, state and local laws governing the protection and use of confidential records maintained by our social services agencies.

     I believe there is definitely a role for the Federal Government to help in this area so that every other state or city trying to identify what data can be shared is not overwhelmed with legal analysis and forced to recreate the wheel time and again.

     I also believe that we need to be careful to recognize that the sharing of data is not synonymous with the sharing of eligibility rules.

     Although we strive to make sure that low income individuals and families have appropriate access to benefits for which they are eligible, we need to be mindful of unintended consequences.

     Every program has different standards for how to consider resources and income and ultimately determine eligibility, and many of these differences are appropriate given the different goals of the programs.  Therefore, discussions as to what extent programs that share common data or clients should share the same eligibility standards needs to be approached strategically.

     I thank the Committee and the Chairman for all of your work on addressing this important issue, and for moving this important agenda forward.

     Thank you.

     [The prepared statement of Mr. Doar follows:]

     *Chairman Davis.  Thank you, Mr. Doar.

     Ms. Zielinskie, you are recognized for five minutes.



     *Ms. Zielinskie.  Congressman Davis, Congressman Doggett, members of the Subcommittee, thank you so much for the opportunity to testify today and for your ongoing work in regards to the Standard Data Act.

     I would like to start today by sharing a quote with you from one of our clients, a 58 year old unemployment insurance exhaustee who was eligible for more than $200 a month in SNAP benefits.

     “I was so shocked when I received your call.  In the past three weeks, I have gone through three failed job interviews and I felt like a failure.  This benefit amount is equivalent to my monthly mortgage payment.  I am so glad you did not let me drop through the cracks.”

     Benefits Data Trust is a national not for profit organization committed to transforming how people in need access public benefits.

     We have successfully completed over 280,000 benefit applications on behalf of low income Americans through the use of data sharing strategies to target outreach and streamline the application process.

     Maximizing private sector targeting outreach strategies, Benefits Data Trust has been able to utilize more than 20 different targeted Federal and state government agency data sources to conduct national, statewide and regionally-based outreach.

     We all know that data driven strategies can and should combat fraud and create efficiencies in the verification process.

     I would like to share with you today five key points on how and why, right now, without prohibitive investments in technology, data sharing can and should be used to increase access to public benefits for the people who need it most.

     First, data sharing strategies can create vast opportunities to conduct targeted, cost effective outreach.

     Federal and state agencies can share enrollment data internally, across departments, and with business and not for profits to generate targeted outreach lists of millions of individuals who are highly likely eligible and not enrolled in benefit programs.

     For example, in 2010, working with the Pennsylvania Department of Labor and Industry and Department of Public Welfare, BDT identified 80,000 unemployment insurance exhaustees commonly known as “99er’s,” who were likely eligible and not enrolled in SNAP.

     The three entities developed comprehensive data share agreements and a process to generate an automated monthly file of new exhaustees eligible for this outreach.

     The targeted outreach achieved initial response rates of 25 percent.  In comparison, standard direct marketing delivers response rates of closer to one percent.

     Results show that many individuals we help are not familiar with the safety net system at all, and have never needed help before.

     Nationally, in the last year, more than 5.5 million people have exhausted their unemployment insurance benefits.  This illustrates the tremendous opportunity and responsibility we have to help folks grappling with the recent economic recession.

     Second, data sharing strategies can streamline the applications process and create express lane eligibility opportunities.

     In Philadelphia, Benefits Data Trust receives an automated monthly data file of individuals 60 and older who were recently enrolled or were re-certified for Medicaid and not enrolled in SNAP.

     Since income, residency and Social Security number or non‑citizenship status were just verified by the same department, an individual does not need to provide proof of these elements on their SNAP application.

     Therefore, eligible individuals are able to apply for SNAP in one phone call without having to provide any additional documentation.  This dramatically streamlines the application process for the applicant and the verification process for the administering agency, reducing the cost of outreach, application assistance, and eligibility determination.

     New enrollment or recertification in Medicaid, TANF, heat assistance, or the earned income tax credit approval, is an opportune moment to help people access other benefits they need and create express lane eligibility opportunities.

     Third, increasing access to public benefit programs for people who are eligible, especially seniors, helps individuals, strengthens our local economies, and reduces long term national health care costs.

     Fourth, data driven strategies can cut outreach and application assistance costs by more than 70 percent.

     Fifth, there are several actions that the Federal Government can take to make it easier for Government, business and not for profits to utilize data sharing strategies.

     The Federal Government can take a proactive approach by setting guidance around standard data share and data security provisions.  This will immediately provide Government agencies, business and not for profits clarity on how to share and protect data.

     Continuing to fund technical upgrades and integrations of state systems as well as projects that utilize data driven approaches to outreach and enrollment will continue to enhance improvements in this area.

     I would like to end with one last quote from an 82 year old woman whom we were able to apply for both food stamps and the low income subsidy.

     “I do not even know how you found me.  My husband died of Alzheimer’s and we both worked all of our lives.  His treatment ate up all of our money.  Thank you so much.”

     [The prepared statement of Ms. Zielinskie follows:]

     *Chairman Davis.  Thank you very much, Ms. Zielinskie.

     Mr. McDonald, you are recognized for five minutes.



     *Mr. McDonald.  Good morning, Chairman Davis and distinguished members of the Subcommittee, and thank you for the opportunity to testify here today.

     I am Darryl McDonald representing Teradata Corporation.  For 30 years, Teradata has redefined the lead of database technology and the use of advanced analytics.

     Among the nearly 1,400 Teradata customers, companies such as eBay, Wal‑Mart, Wells Fargo, Caterpillar, and AT&T, have learned how to recognize data as their most valuable asset by transforming that data into useful information.

     These same approaches can be applied to the Government.

     My testimony today will focus on how applying advanced analytic solutions to massive data sets or big analytics helps Federal agencies meet complex, large scale mission demands despite unprecedented budget cuts and infrastructure.

     It is important to remember that big analytics are not futuristic or incomprehensible.  In fact, one of the best known example is from Major League Baseball, the 2002 Oakland A’s.

     As described in the book “Moneyball:  The Art of Winning An Unfair Game,” and the subsequent movie and feature film, the A’s used data analytics to more effectively compete against other teams using less money.

     A lesser known example is from one of our state government customers found in Michigan.  Since 1996, the State of Michigan has been creating their enterprise data warehouse which supports reduced health care costs and a 25 percent reduction in administrative costs.

     The Michigan enterprise data warehouse has also aided the state in many ways, including doubling the recoveries for Medicaid fraud, moving from last to first in child immunization rates, and identifying more than $70 million in fraudulent child care.

     Imagine the benefits achieved by Michigan at the scale of Federal Government.

     Giving Federal decision makers the ability to utilize all the data on hand to find underlining insights is essential for effective risk management, cost containment, and mission success.

     Rather than making weak decisions with data sampling, big analytics utilized all the available data to enable fact based decisions.  Big data analytics cuts analysis time from weeks or months to near real time and enables continuous improvements to meet changing technology, regulatory and mission needs.

     As one of the world’s largest creators and consumers of data, the Federal Government will see its long term future of success or failure linked to how well it addresses big data.

     Teradata has numerous and varied success stories of helping clients integrate and understand and leverage big data at the Federal and state levels, within and outside the U.S.

     For example, the U.S. Department of Agriculture’s Risk Management Agency has successfully applied data and analytics within the Federal Crop Insurance Program, by bringing together disparate data to identify and combat fraudulent claims, the agencies have saved the American taxpayers approximately $838 million in improper pay outs from 2001 to 2011, with cost avoidance estimated at $1.5 billion.

     Another example is the U.S. Transportation Command.  The Legacy information systems used by Transportation Command limited the visibility to historic and current shipment and arrival detail.

     Today, data from 33 different systems is integrated to improve decisions with near real time visibility across the Department of Defense.

     A final example is the Centers for Medicare and Medicaid Services and the CMS Data Dashboards.  Completed within five weeks, the initial Dashboard was launched early because among other reasons, the historical claim data was housed in a single system.

     With many organizations seeing tremendous benefit, a logical next question is how do we motivate more Government agencies to adopt big analytics?

     In that respect, Mr. Chairman, the timing of this hearing could not be better.  Teradata strongly supports H.R. 3339 and respectfully requests Congress to pass this legislation.

     Not only would this legislation improve cost effective delivery of essential services to millions of Americans, but it would also create a stepping stone for other Government agencies to understand and acquire the benefits of big analytics.

     Consider the lessons from Moneyball.  If analytics can change a hundred year sport like baseball, think of the possibilities for Government.

     Once again, Mr. Chairman, thank you for your continued leadership on this issue, and Teradata stands ready to support you and your colleagues.

     Thank you.

     [The prepared statement of Mr. McDonald follows:]

     *Chairman Davis.  Thank you very much, Mr. McDonald.

     Mr. Pryde?



     *Mr. Pryde.  Chairman Davis, Ranking Member Doggett, and members of the Subcommittee, thank you for inviting me here to discuss the use of standards to improve Government reporting.

     I am Campbell Pryde, President and CEO of XBRL US, a non‑profit organization established to support the implementation of standardized business reporting by Government and business through marketplace collaboration.

     I applaud you and the other Members of Congress who are striving to make Government more effective and efficient by using data standards.

     I will briefly discuss one such standard, the Extensible Business Reporting Language or XBRL.

     XBRL is a data standard that is used to communicate financial and performance related data by both business and Government.

     The objective of XBRL is to enable compatibility and comparison of the data that is being standardized.

     Shipping containers are an useful analogy to demonstrate the importance of data compatibility.  The standard shipping container revolutionized the way that products were transported from the manufacturer to the consumer.  It increased the speed of delivering products to market and reduced perishable waste.  It reduced transfer costs between ships, trucks and rail.

     No longer did it separate products, have to be manually loaded onto trucks, reset onto ships, and off loaded onto trains.

     This also drastically reduced loss through theft.

     It ultimately changed the design of ships and trains to accommodate the new shipping containers allowing both ships and trains to carry far more cargo.

     Finally, it reduced storage costs because cargo is moved more quickly and it can be stacked at storage facilities.

     All of these gains were the result of developing a global standard for shipping containers.  The manufacturers, the shipping companies and the transportation companies realized that they could drastically reduce handling costs with a simple standardized solution that was compatible between trucks, trains and ships.

     Today’s management of financial reporting data in Government resembles the transport industry before the introduction of the shipping container.

     The implementation of data standards such as XBRL can improve the speed, reduce the transfer costs between systems, allow more data to be moved and reduce storage costs by allowing storage in one format.

     In addition to compatibility, data standards also facilitate the comparison of data. By standardizing financial and performance data, it can be quickly compared.  This allows data to be easily aggregated and disaggregated, giving the users the ability to disaggregate or drill down into information, providing improved transparency and the ability to compare, for example, the relative performance of recipients of Federal funds.

     It is important, however, that everyone uses the same standard.  Standards, just like electrical plugs, are standardized within countries, but unlike the shipping container, the electrical plug standard is not global.  This lack of standardization requires the need for adapters and transformers to use the same product across countries.

     The use of different data standards across agencies will result in the same problem.

     Some recent legislation has suggested the use of XML as a reporting standard.  XML is a flexible data standard that like electrical plugs can use different formats to deliver electricity, or in this case, information.

     Allowing agencies to use their own variants of XML to communicate financial reporting data will result in data compatibility issues between agencies and will minimize the ability of taxpayers or policy makers to efficiently analyze such information.

     XBRL is a specific data standard and ensures that reported financial information is compatible and comparable across all agencies.

     Moreover, XBRL eliminates the need for adapters or transformers to transfer the comparable financial data between users.

     Standards like XBRL will ensure that financial and performance information can be transported in a cost effective and timely way from creation through to analysis.

     This will enable you as policy makers to monitor and track Government spending as well as use up to date financial information to make appropriate future allocation decisions.

     It will also allow the Federal, state and local agencies and other recipients of Federal funds to analyze how those funds are used.

     Most importantly, using data standards  like XBRL to communicate this information will reduce costs, increase speed, increase transparency, and increase the effectiveness of taxpayer dollars spent.

     XBRL US has vast experience and expertise in XBRL development through our work with the SEC and the FDIC.  We are ready and available to help Congress and those in the Federal and interested state agencies in this important initiative.

     Thank you again for having us here today, and I look forward to answering your questions.

     [The prepared statement of Mr. Pryde follows:]

     *Chairman Davis.  Thank you very much, Mr. Pryde.  I appreciate all of your testimony.  We appreciate your perspectives on this important topic.

     Before we move to questions, I would also like to insert a letter from the Kentucky Society of CPAs into the record, without objection, so ordered.

     [The information follows: The Honorable Geoff Davis]

     *Chairman Davis.  When we started this data standards effort, it was based on a question about what was going on in programs under the jurisdiction of the Committee.

     For me, from a personal perspective, to give you an idea, coming from the private sector, working with data integration in a retail business to business environment, early in the standardization with electronic data transfer, before the Internet fully stood up, and then saw the beginning of that coming in, I sat down with the Subcommittee and I asked a question as the new Chairman, how do I get a cost rule up?

     For those in the audience who might not be familiar with that, the idea of saying if I take what would be called a “vendor master,” similar to what the gentlewoman, Ms. Roy, from NIEM talked about with data matching across standards, trying to find out exactly if we had the means of understanding who was on all of what programs with one given record, similar to some of the initiatives Mr. Doar is taking.

     The answer was that we cannot do that.  That literally was the genesis moment from years of discussion about data standardization to move forward and talk about this.

     I appreciate all of you being here.  We knew there was a lot of program overlap within our programs that creates additional cost in overhead, takes away taxpayer dollars, and does not help beneficiaries.

     We also knew there were serious questions about program effectiveness that current data could not successfully answer, especially when there are multiple programs that are involved, and the data standardization effort that we put in place will eventually get us to the point of being able to address these questions and concerns, we believe.

     So far, we have made considerable progress with child welfare, with Temporary Assistance to Needy Families, with Unemployment Insurance, and hopefully soon, child support enforcement.

     However, if we are going to take a beneficiary-centered approach to better using data standards in the Administration of public benefit programs, it needs to be the most complete view, and that probably means programs outside the jurisdiction of this Subcommittee, like food stamps, Medicaid, housing, Medicare and Social Security.

     Mr. Doar, I will start with you.  As we look to expand this effort, from your experience, what human service programs should we seek to include in this effort?

     *Mr. Doar.  Well, the one we struggle with the most is schools.  Those may be beyond your jurisdiction. 

     *Chairman Davis.  We would be glad to bring that under the jurisdiction.

     *Mr. Doar.  The data concerning children in the New York City Public School system is very firm.  They are not participating, and they feel constrained, that they cannot participate.

     I think data concerning enrollment and what is happening in schools would be good for us in social services and it would be good for them in education.

     That would be the area that we are most frustrated by at this point.

     Child support enforcement is great.  We have food stamps.  We have Medicaid.  We have cash assistance.  We have housing.

     The one that we struggle with is schools.

     *Chairman Davis.  I appreciate your perspective.  My wife and I have volunteered with kids and families on the edge for over 25 years.

     The one consistency, particularly now, is our oldest daughter is a school teacher dealing with at-risk children, and I am sure it is the same phenomenon in New York as in our small little part of the world, moving to multiple schools sometimes within a year, even within the schools, they cannot share information about the kids which creates a real educational challenge as well, not to mention the lack of connectivity.

     Ms. Zielinskie, your organization works primarily with seniors, correct?

     *Ms. Zielinskie.  That is correct.

     *Chairman Davis.  What are the most common combinations of programs that seniors are usually eligible for but they may not receive?

     *Ms. Zielinskie.  Sure.  To help seniors reach and maintain economic security, we really need to look at housing, health care, and basic living costs.

     Obviously, housing is the biggest indicator of whether or not an individual will be able to reach and maintain economic security.  That would also include heat assistance or LIHEAP.

     Second, we talk about health care.  Obviously, if we take a look at poverty among seniors and actually include health care costs in that calculation, millions of seniors are in poverty and struggling to meet their health care costs.

     In addition, there is also the cost of food.

     Combining those programs as we take a look at helping low income seniors in America is critical.

     *Chairman Davis.  Mr. Doar or Ms. Zielinskie, either one, is there certain program data that you wish you had that could further streamline your organization’s efforts?

     *Ms. Zielinskie.  Yes.

     *Mr. Doar.  In my case, the other area that is a constraint is the Social Security Administration.  While a lot of what we have done with HHS‑Connect was limited because some data we get is a result of previously established matches with the Social Security Administration, and to the extent that we got data from them through a match, there are very strong prohibitions against our ability to share what they gave us outside our area.

     That is a problem.  That leads HHS‑Connect to not be complete, so a worker could look and see if there is any information, but could not be sure that by it not being there on a particular case that it is not there somewhere, if we got it from SSA.

     That would be another area where this initiative could be even more successful if that kind of Federal guidance allowed us to go in that direction.

     *Chairman Davis.  Thank you.  Ms. Zielinskie, briefly.

     *Ms. Zielinskie.  Sure.  That being said, I think the Medicare Improvement for Patients and Providers Act or commonly known as MIPPA, has made great strides in how Social Security and the Low Income Subsidy data has been able to be shared with the states.

     I think when you start talking about challenges relating to access points of which data is available, it gets down to talking with states and lawyers about what is able to be shared and how it can be used.

     It is not necessarily the data set so much it is an opportunity for access.  Obviously, Medicaid is going to be the most telling in terms of if we are really talking about reaching the poorest individuals to receive support.

     It is more about guidance around how different agencies, business and community partners can share data while also protecting an individual’s privacy rights.

     *Chairman Davis.  Great.

     *Mr. Doar.  Mr. Chairman, can I just add also, when we put Medicaid data on HHS‑Connect, we only put the fact that they are eligible or receiving, they are a Medicaid recipient.

     Claims data is also very strictly limited in our ability to share.  In many contexts, particularly with very vulnerable people, people who are high users of Medicaid, knowing the extent to which they are taking advantage of Medicaid provider services and those services are being claimed on their behalf would be very helpful in both limiting costs and getting better care.

     *Chairman Davis.  I appreciate the perspective that both of you have shared.  I would say we met early on with the Inspector General, the Social Security Administration, and I think in all of the agencies we met with, the leadership understands the importance of sharing.

     There may be some issues that have to be addressed statutorily that would legitimately protect privacy but allow enscripted data be shared with those who are receiving the service as well.  That will be a continued discussion as this implementation moves forward.

     With that, I would like to recognize the gentleman from Washington State, the former Chairman of this Subcommittee, Mr. McDermott, for five minutes.

     *Mr. McDermott.  Thank you, Mr. Chairman.  I want to commend Mr. Davis.  We are going to miss you because this issue of data is a huge problem.

     I go back to the days in the State of Washington when we tried to put welfare/mental health data on a database so that when a mental patient showed up in this clinic and then showed up in that hospital, then showed up across the state, we would have some way of someone knowing what had happened before.  We ran into all kinds of problems.

     I also have spent a lot of time in trying to get data in the military medical system and the Veterans Administration medical system to talk to one another.

     You have a proprietary system in the military that does not talk to the publicly developed system in the VA.

     A guy gets blasted in Afghanistan and gets put out of the Army and into the VA, his medical records ‑‑ if you go to the hospital in Seattle, the doctors are sitting there with two computers.  One with the military system and one with the VA system.  It is absolutely insane.

     This data thing, it would make better health care, it would make better all of our social systems, so I commend Mr. Davis for bringing this.

     What I am interested in, Ms. Zielinskie, is this.  You are talking Pennsylvania.  How wide across the country, does your organization work in other states doing this, or are there agencies in all the 50 states?  How is it working to use this, what you are trying to do is looking for people eligible for benefits.

     *Ms. Zielinskie.  Correct.

     *Mr. McDermott.  You are using whatever data system.  Tell me what is going on in the rest of the country.  You are our only window into the 50.

     *Ms. Zielinskie.  Thank you.  Thank you for the question.

     We have actually been able to conduct outreach nationally, and we have been able to conduct outreach in other states in addition to Pennsylvania and are exploring opportunities in some of our partner states.

     I think to your point, different states have different make‑ups or the puzzle of how their data sits.  The information that we seek to gain access to is not terribly deep.

     It is very possible to pull automated files once you start getting to the right data folks that are in the different state departments.

     It crosses agencies and the landscape are different, so where food stamps sits and where Medicaid sits – how it is set up in New York City is not necessarily how it is set up in Pennsylvania or Virginia or New Mexico.

     *Mr. McDermott.  Let me ask a question, a specific question.  It is an issue you brought up, which is one that has troubled me because of this Committee.

     You have the people who have had middle class experience, and they lose their job, and 99 weeks later, they come to the end of that and they have nothing.

     *Ms. Zielinskie.  Yes.

     *Mr. McDermott.  They have no idea of the social service system, number one.  Number two, they are too embarrassed to go over to the welfare office and see if they qualify or wherever you have to go on this food stamp stuff.

     How does it work for states trying to find those people or most states just saying if we do not hear about it, we do not need to worry about it?  How are they dealing with the 99er’s who are out there eating into their 401(k)’s to keep a house but are eligible for food stamps?

     *Ms. Zielinskie.  I think there is a wide variety in what different states are doing.  I do not have necessarily that information here, but would be more than happy to get back to you about other efforts.

     I know there are a lot of other national advocacy organizations engaged in talking about 99er’s.

     To your point, about 45 percent of the folks that we help with the unemployment insurance exhaustee outreach project are over the age of 50.  Exactly what you are saying– they are spending down their small nest egg.  They never thought they were going to be in the situation that they have found themselves in.  These are the individuals that are in great risk of foreclosure.

     How it all connects is very integrated.  I almost pulled a quote of an individual who is an MBA, and for two years, he has been searching for a job. It was a challenge for him to overcome the stigma and pride and call us. Obviously he is a hard worker if he was able to get an MBA– and now all of a sudden he is in need. Overcoming that stigma is something that I think illustrates how important it is to not only think about data in terms of fraud and creating efficiencies, but also that it is absolutely critical to think about it in terms of how we can help folks that are teetering on falling into poverty.

     *Mr. Doar. I would just add the programs that we run in New York City are well known to the populous.  There is lots of promotional activities that go on.  Food stamps, USDA is running advertisements promoting food stamp benefits.

     The number of people who take advantage of these benefits are quite large now in the City, food stamps and Medicaid particularly.

     In the case that you talk about, people who exhausted unemployment insurance tend to turn to health insurance, public health insurance first, and perhaps food stamp benefits, and then last, TANF.

     The other thing is that in New York City, the Mayor continues to have an expectation that people who are struggling should work and personal responsibility matters.

     We have to be careful about the extent to which we are overly promoting the receipt of public assistance as a substitute for personal responsibility and employment.

     *Mr. McDermott.  Could I ask one more question?

     *Chairman Davis.  Briefly, yes.

     *Mr. McDermott.  Yesterday, we eliminated the Social Services Block Grant.  How much of that comes to your Department and what is it used for?

     *Mr. Doar.  Well, at HRA, the Adult Protective Services Program is a program that serves our most, most vulnerable populations.  We have about 9,000 people who have been determined unable to care for themselves in any way.

     That is a program we are going to have to run, and people are going to have to serve, 75 percent of the funding for that program in HRA comes from the Federal Government.

     To the extent that we get less SSBG money for Adult Protective Services, the Mayor will have to make up the difference, and that is a fiscal burden on us, and it is not a population we can neglect, it is a particularly vulnerable population.

     I think from the Mayor’s perspective, if he had a choice of Federal programs that are in need of reduction, that might not be the first one, particularly as it affects Adult Protective Services.

     *Mr. McDermott.  Thank you.

     *Chairman Davis.  The gentleman’s time has expired.  The Chair now recognizes Mr. Paulsen from Minnesota for five minutes.

     *Mr. Paulsen.  Thank you, Mr. Chairman.  To the panel, there was recently an announcement for this year’s Human Services IT Conference, it is called ISM, and the announcement listed sessions on the usual new service delivery models, emerging technologies, and general best practices.

     What caught my eye and drew a little bit of attention, I think, was the session description that said this “The era of big data has arrived at ISM.  State and Federal agencies are looking to advanced data analytic capabilities, improvement predictive modeling, to reveal patterns of behaviors and outcomes that were previously buried in mountains of data.  ISM 2012 Data Analytic Sessions show how big data can turn into big discoveries.”

     Mr. McDonald, how are big analytic efforts different than what our agencies are doing now, and do you agree this is what agencies should be doing in the future?  Is this the direction to go?

     What is the difference between data and data analytics today versus 30 years ago?

     *Mr. McDonald.  Sure.  The answer is yes.  I think what is happening is the amount of data is doubling extremely fast, just from traditional organic data.

     With all the new digital data that is being created and leveraged across the different agencies, you can imagine the data that they are having to wrestle with is overwhelming.

     You really have to start looking at instituting business rules and analytics to help mine that data based upon the kinds of goals that you set and the outcomes you want to get, and more importantly, how do you push that out to the front line Government workers so they are able to make decisions more quickly on things that cross in front of them.

     I would say today people are overwhelmed with data and we have to use technology to take that overwhelming aspect out of it and try to give them the actions they should take based upon what is sitting in front of them, and again, whether it is private corporations or Government agencies, they are all struggling with this.

     I would say what has happened is the amount of data has changed over the past 30 years, but more importantly, the complexity of it has changed as well.

     As you talk about this unstructured to big data, it really is marrying up new data types with traditional data types which says you have to understand how to read and integrate both, but more importantly, take that confusion out of it, simplify it, and give people on the front lines the types of activities and actions they should take based upon what rules you set in place and the outcomes you want to get.

     *Mr. Paulsen.  Let me ask this for the rest of the panel because we are going to find a lot of information obviously in these mountains of data that are out there.

     Ms. Zielinskie, I think in your testimony you mentioned that we are going to be able to connect individuals to benefits, they may not have been getting these benefits before, we are going to be able to tie that together.

     How about the converse?  Will we be able to make strives down the road to make sure only, for instance, people that are eligible for programs like low income people that are able to get welfare benefits get those benefits, that programs that require work activity, can we ensure that actually happens?

     Can we tie the data in that fashion to make sure we are going to see some progressive results there?

     *Ms. Zielinskie.  I think there is absolutely opportunity to be comprehensive in how we use data.  We certainly do not want to have people who are not eligible for these programs receiving them.

     That being said, we want to make sure that people who are eligible are able to access them so they can get themselves out of poverty and to self sufficiency.

     Certainly there are accuracy components of using data.  If we look at, for example, SNAP in Pennsylvania, there is a payment accuracy challenge of one‑tenth of one percent.  There is a fraud issue of one‑tenth of one percent.

     I think that it is critical that when we are talking about data that we do not just talk about it in the frame of fraud.  We also talk about it in terms of making sure we are helping the right people in a comprehensive way so they can get out of poverty.

     *Mr. Paulsen.  Mr. Doar, in terms of waste, fraud and abuse?

     *Mr. Doar.  Yes, we absolutely see the benefits of this kind of sharing of data to see the inconsistencies in income reporting and work status and household status that would allow us to be sure our programs are being expended correctly.

     The most exciting use of big data from our standpoint in the City is the potential use to identify households where there is a potential child welfare issue in the future.

     You could use data to find what are risk factors in advance of unfortunate circumstances happening.

     That is where I think right now we are trying to find what tells a child welfare protective services worker, what do they need to know that would lead them to say this is a more potentially at risk family than another family, so we can prevent tragedies from occurring.

     *Mr. Paulsen.  Thank you, Mr. Chairman.  I yield back.

     *Chairman Davis.  I thank the gentleman.  The Chair now recognizes Mr. Berg from North Dakota for five minutes.

     *Mr. Berg.  Thank you, Mr. Chairman.  I am not sure exactly where to start, but it keeps coming back to me that nothing measured gets managed, and I think out there, nothing that is measured, it will not get managed unless it is measured, but it will not get managed unless it is looked at.

     That is what I cringe at, just the mountain of information, a lot of really great information that would really help us as policy makers make decisions when we get to the point where we are going to make decisions.

     I am looking at kind of the big picture.  My question really relates to standardization.  Again, kind of boiling this down, we are going to look at a program and we make a big effort to really look at that program, but within our scope, there are so many other programs that affect that individual that is within that program, and yet we are just kind of looking at that program somewhat in isolation.

     We are looking at this international child support law and what it is doing.  One of the things we are looking at there is tapping into the National Directory of New Hire’s.

     We are kind of looking at that as a way to expedite a lot of information and hopefully lower the cost of a lot of these programs.  That information will be right there.

     I think that is ultimately what we want to do, have some uniformity and make sure we are delivering those very efficiently and cost effectively.

     Maybe it is kind of a bigger question, but the question is how would standardization help us when we are reviewing and evaluating programs?

     Is it worth the effort to move towards standardization because that benefit is going to be there?

     It is kind of a question for each member of the panel.

     *Mr. Doar.  It would absolutely help us.  One of the theories of the post‑welfare reform world was that we have a combination of programs that can serve as work supports.  We want to know if people are working and taking advantage of SNAP, public health insurance, perhaps other forms of assistance.

     It is good to know how those programs work together for an individual client, and whether they work together, for how many.

     The analytical opportunities that we can see if we get greater standardization across programs so we can know who is a multiple program user and who is not and why not.

     I absolutely agree about the National New Hire’s database.  I come from the child support program.  It is a tremendous resource and has greater potential than it is currently being used for.

     There is no question that standardization will allow us to see not just how these programs work by themselves, but how they work in concert with other programs and with employment, which is I think what we are all trying to do.

     *Ms. Zielinskie.  Thank you for the question.  The primary focus of our work has been on older Americans.  There has been great work done with the Elder Index, which does exactly what you are talking about.

     I think absolutely standardization around what an individual needs to reach and maintain economic security is critical, as we explore how to support our older Americans.

     That being said, we also need to take into consideration that the cost of living in New York City, for example, is probably different than in North Dakota.

     *Mr. Berg.  Some parts of North Dakota get pretty expensive.

     *Mr. Zielinskie.  The Elder Index does that.  Wider Opportunities for Women and the National Council on Aging are also engaged in a lot of work around economic security for older Americans.

     I guess a caution as we look at standardization is that it is critical to also need to look at the environment in which those individuals are trying to thrive.

     *Mr. McDonald.  I would just add that I think the standardization has to be the starting point for trying to take this complexity out of the programs and agencies and the investment in that will reduce the amount of future investment that you have to deploy on programs for them to be able to share information and get value out of that information, both from a cost savings standpoint, but more importantly, how to reallocate it to the right programs and the right missions.

     I think it is just fundamentally the right place to start.

     *Mr. Pryde.  I think one of the advantages is you do not get the data stuck in separate systems.  If everyone is using a different system, there is no way these systems can talk to each other.  If we have standards across an agency or among agencies, one system can talk to another seamlessly.

     I can request some information, that request can go down to all those systems and pull it on the fly, which is not the case where you have different computer screens to log into each different system.

     *Mr. Berg.  Thank you.  I will yield back, Mr. Chairman.

     *Chairman Davis.  I thank the gentleman.  The Chair now recognizes Dr. Boustany from Louisiana for five minutes.

     *Mr. Boustany.  Thank you, Mr. Chairman.  Under our Subcommittee’s jurisdiction, we have two programs, SSI and Unemployment Insurance, that both have been deemed high-error programs by the Office of Management and Budget.

     In fact, in fiscal year 2011 alone, they combined to account for $18.3 billion in improper payments.

     Mr. McDonald, I read through your testimony and you had a lot of detail in there referencing the work that you all have done in Michigan with regard to fraudulent child support.

     I would like you to elaborate for the Committee a little more about that work and what steps could we take at the Federal level to implement some of your best practices?

     *Mr. McDonald.  Sure.  I think what the State of Michigan decided to do was build a platform that integrated its information across all their different programs.

     They knew by doing so they would not only provide better customer satisfaction to those people that are using the programs, but they would link that information to find fraud and abuse, and then take that extra money and apply it back to those programs that were under funded.

     Think about the simplicity if I come and get a new driver’s license but I have not been paying my child support, instantly, you are able to connect those two, and start trying to recoup those benefits, right?

     Just the simplicity of connecting different programs and agencies’ information gives you that instant insight of first, being a better provider to those taxpayers, but secondarily, catching fraud at the moment it happens instead of having to wait a year later to determine if we have been over paying or under paying in certain areas.

     I think it is very common practice on the business side and I think it is now becoming an opportunity in the Government to start looking at that same benefit they can reap by consolidating the data, standardizing the data, and giving access and interoperability to the different agencies in a quick fashion, so they can catch things both proactively and defensively.

     *Mr. Boustany.  Thank you.  At a time when we know we have a number of duplicative programs with a lot of overlap and we are dealing with budgetary problems, looking at the effectiveness of these programs is critically important.

     It seems to me what you just described would also help us in our oversight role, sort of paring down, figuring out how to most efficiently use those resources.

     *Mr. McDonald.  I would agree.  I think too often people ask about how much money do you need, and I think not enough is put into what will be the impact or return of that investment.

     In most cases, what we are seeing, especially like the State of Michigan, they spent a few million dollars, but they got $15 for every dollar they invested.

     Who would not like that formula for investing, to try to take costs out and return that money back, right?

     As you can see, understanding the impact, but also everyone now operates under a similar fashion.  Whether you are Bank of America or CMS, in reality, projects are 12 month based, they are program based, you have to spend money and see that return in a quick fashion, but you have to have a view as to where am I going to be in three years, am I incrementally improving the overall process versus this program approach and a 12 month approach.

     *Mr. Boustany.  Thank you.  From the time the first set of data standards were included in the child welfare programs to the next iteration of TANF and Unemployment programs, the language went from suggesting a standard like XBRL to suggesting XML.

     Some of the concerns that were expressed was that XBRL was too expensive or too complicated to implement, and the feedback we received was that we should only recommend that the agencies implement XML.

     Mr. Pryde, could you respond to those criticisms for us?  Is it too complicated or too expensive?

     *Mr. Pryde.  It depends on what you are trying to do.  XBRL is specifically designed for reporting financial and performance data.  It is not designed for reporting information like biometric information or information on human resources cases.  It is specifically done for financial reporting.

     To use XBRL for sharing information about a child’s condition or operational issues, XBRL would not be appropriate in that case.  It would be silly to do that.

     There are other standards that would be more appropriate in those cases.

     However, if you are going to report financial reporting information like how much is this agency spending, how much are we spending on this program, what period are we spending it, how much is wasted and what is the cost, et cetera, then XBRL is an appropriate standard to use.

     If you ask that question, let’s use XBRL to standardize every single piece of data, yes, that would be correct, it would not be appropriate to do that.

     *Mr. Boustany.  I thank you.  Mr. Chairman, I yield back.

     *Chairman Davis.  I thank the gentleman.  The Chair now recognizes Mr. Reed from New York for five minutes.

     *Mr. Reed.  Thank you, Mr. Chairman.  Thank you again to the witnesses on the second panel.

     Being from New York, I want to spend a little time with my fellow New Yorker here.

     If we could, I am very interested.  I read your testimony, in particular, pages six and seven, talking about the lessons you learned as you implemented the programs in your agency in New York City.

     I want to kind of delve into a little bit more detail as to what you could offer us from your experience in implementing that, in particular, and also the interaction you referenced, the problem with Social Security numbers, the Social Security Administration.

     Obviously, they are such a large agency with tons of data that needs to be part of this conversation, in my opinion.

     I am very interested in your point of view on what lessons you learned that we can learn here in Washington, D.C.

     *Mr. Doar.  Well, one of the lessons we learned is that the direction from the top had to be emphatic, that the lawyers had to do everything they possibly could and be as imaginative and as persistent in trying to make legal and appropriate, not inappropriate, arguments for allowing the sharing of data.

     Bureaucracies, as I think the previous panel mentioned, are inherently narrow and afraid, and afraid of doing something that someone may say in rare circumstances it was wrong or not consistent with policy.

     That was a very important lesson.  We really pushed our lawyers to really think creatively and set up the process where those agencies that received data had to sign a form with the commissioners committed to enforcing confidentiality and security rules on employees that violated them, and to report back to the giving agency so that we were consistent.

     We had an advantage in New York State, social services is governed by sort of an umbrella agency, and that is mine, so I could delegate to sister agencies the role of being social services and in a social services’ purpose.

     One lesson is to get the lawyers and the mayors, chief executives’ position be very strongly felt that this was something we were going to try hard to do.

     Social Security Administration data is data we use now in certain programs.  We use it based on established agreements with SSA that have taken a long time to be established, that have allowed us to get data.

     Their prohibitions against sharing of that data are very firm.  One of the things our lawyers had to do was not allow that data from that source to be used in the way that we wanted to, and that is a problem that we think we would like to work on nationally.

     Also, there are HIPAA requirements involving the sharing of Medicaid or health data, which we basically agree with.  This is private data.  This is something that should not be shared inappropriately.

     Claims data as opposed to just being on Medicaid, but actually just how much is being expended on the client in Medicaid and where is strictly prohibited, so we do not share that, and that is a problem, which we think we could make greater work with this if we had that available to our sister agencies and to ourselves in a way that it could be used.

     Finally, we would constantly come up to situations ‑‑ I am on a taskforce that looks at the gap between African American children and white children in schools.

     We know that exists.  Education wants us to work with them on that, and we want to work with them on that, but we do not share any data.

     We do not know about the use of public assistance within schools by children and they do not know it either.  In both cases, I think it would be useful.

     Those are the frustrations that have occurred.  It required real work.  It required somebody holding the bureaucratic narrow individual agencies accountable, and it ran into some barriers we just could not get through with regard to sharing data.

     *Mr. Reed.  I appreciate that.  Mr. McDonald, from your experience in the private sector, if we leave SSA out of this conversation and do not really focus on that, do you see a problem, how that could impact the development of standardization across the agencies?

     It is such a large agency and such a large amount of data.  I just want to know from your perspective, would that work in the private sector?

     *Mr. McDonald.  I think in the private sector, they have come to realize the importance of standardization across corporations.  I think through the evolution of how they have gone through mergers, acquisitions and growth, they have all had to realize that unless we standardize on whether it is financial data, product data, or customer data, how will they be able to roll these companies and report on that.

     I think no matter the size, there are enterprises bigger than many of the agencies that have done that.

     Is it doable?  Yes.  Is it required?  Yes.  I think it is a matter of putting the energy and the standards behind it that says let’s adopt this.

     I think what you will see is once you start, five years from now you will say why did we not do that ten years ago.

     It is just a matter of making that first step and getting everyone committed to this is what we have to do to get to where we want to be, so let’s stop talking about it.

     *Mr. Reed.  I appreciate that.  With that, my time is expired.  I yield back.

     *Chairman Davis.  I thank the gentleman from New York.  I just have to tell you the little huddle that was going on over here in the early part of Mr. Reed’s question, your testimony has inspired two of our subcommittee members to come up with another piece of legislation to compliment what we are doing for data sharing on the health care side, our two doctors on the Subcommittee.

     As we wrap up, I would like to ask one final question.  It is open to all to comment.  I would like to start first with Mr. Pryde, then with Mr. McDonald, on this issue of analytics and standards.

     I am going to preface this by saying coming into Government from a world dealing with implementations, focusing on internal rate of return, and the idea that I can make an investment and know, particularly for those of you who have all dealt with what you might call a return on information, a dollar spent on information improvement can  usually return itself 15, 20 or 50 times what was spent in improvements of efficiency, throughput, the agility to respond to unique customer needs because of that access to information.

     Unfortunately, the Congressional Budget Office is trapped in another era and focuses on static rather than dynamic scoring, which is our issue to address in looking at the value of this.

     Speaking of value, I would appreciate it if Mr. Pryde could talk a little bit about the importance of standards.  I am very familiar with XBRL, not from a programming standpoint, but one of the very first pieces of legislation that I introduced, the Transparency and Financial Reporting Act back in 2005, was directly built around the idea of using things like XBRL and XML to harmonize processes, bring best practices to the SEC and other related organizations, with financial services accountability.

     Also, frankly, to offset some of the huge compliance costs imposed with Legacy systems and Sarbanes‑Oxley.

     I was wondering if you could comment on the value that comes in terms of pay back by the implementation of these kinds of common standards, and then the analytics that would come from some real world examples.

     I am quite open to the private sector since the Government is generally a lagging indicator of what is happening out in the country.

     *Mr. Pryde.  Sure.  We recently issued a white paper covering how much Government reporting that is required by companies.  We can make that available to the Committee.

     It kind of backs up what Ms. Roy was saying, where 80 percent of the information was repeated information that they were reporting to different agencies, so you get a massive amount of duplication in data because no one is sharing the data, there is no way of sharing it.  You get significant efficiencies through that.

     In addition, there was an implementation that was carried out by the FDIC five years ago to collect call report data.  This information was called call report data, because they had to go around and call everyone when they submitted the data incorrectly.

     With XBRL, they could check the data on submission, that reduced all the manual costs that were involved in checking on the data and getting it correct.

     A lot of this stuff, once you standardize it, you have created a platform for automation.  Everything can be automated.

     For example, every SEC filing that is filed with the SEC we analyze.  We run 12,000 checks on it.  We find out so much information about these companies which was impossible to do before in an automated way.

     We have this information go back to companies, go back to the auditors.  You end up in an era where you create layers of things that you could never dream of doing before, which can save massive amounts of costs and can completely make redundant very time-consuming and manual processes that you had in the past.

     *Mr. McDonald.  I would just add that if you think about it, without the standards, even if you got everyone to agree that they wanted to track return on investment or the IRR for programs or whatever the right metrics are, without the standardization, it is still going to be tough to measure the effectiveness and how they are calculating it.  They are all using different methods.

     We have a couple of customers that have gone on record, the State of Michigan, and they have advertised that they save over $1 million a day.  That does not mean they are not spending the money back on the programs, but think about the effectiveness of a state saving $1 million a day and reinvesting that back into their programs.

     It is through this investment and the technology to put it all together, set standards, and be able to share this information.

     If I save $1 million in one department, I give it to another department, and they are effectively better utilizing that cash and that investment.

     They in effect are making, like you said, 15/25 percent back on that investment.

     We have had other customers like AT&T who say with our analytic platform, we saved $1 billion over a ten year period.  You ask them, were these big things.  They said no, it is little things.

     It was hundreds of thousands of little things that the front line workers were able to see and solve or fix in a quick manner that allowed AT&T to use those advance analytics, which get deployed across the way, it is operationalized in all of their operations, to save them $1 billion.

     Again, that possibility is capable here for the Federal Government as well.

     *Ms. Zielinskie.  Using standardized data can have many positive impacts.  It can help people who need services in a most comprehensive fashion; it can help reduce costs by being able to determine eligibility cost-effectively; and it can help in administering services as efficiently as possible, and evaluating programs in terms of true value.

     I look forward to working with all of you to figure out how we can prove this and really have a paradigm shift, so talking about data and how we deliver services and embracing the power of the data that the Government has is not innovation but rather the norm in terms of how we do business.

     *Chairman Davis.  Thank you.  Just as a case in point, since I am married to a Bronx girl, Mr. Doar will get to have the last word.

     *Mr. Doar.  Thank you very much.  I am from Brooklyn but we love the Bronx.

     When the Deputy Mayor started the project with the Mayor’s endorsement, we had an ROI all the way through it.  When we were talking to our partners in the technology industries, we were very clear that we are going to make an investment here with your firm or this operation, but we want to see it returned more so to the taxpayers and to our outcomes.

     The key benefit was time and energy.  We are reducing costs of staff by sharing information.  We believe that is happening.

     You should also know that part of the return on investment that the City sees is the extent to which Federal claiming sources are maximized.

     There is a part of this that involves making sure that everybody who should be getting Medicaid or should be getting food stamps does, and that will lead to greater cost across the Federal Government.

     *Chairman Davis.  Thank you very much.  I appreciate all of you taking the time to come in and share from various parts of the country.

     It has been very helpful.  I believe this may not be an issue that brings in crowds of people and emotional activists on either end of the spectrum, but I am admonished to the old baseball rule that baseball is a pretty easy game, it is just throwing and catching and hitting, and in business, it is these basic building blocks that will make the difference in the long run.

     I believe that what we are all discussing here today is at the foundation of transforming our Government in the 21st Century to be something that all Americans can be proud of.

     If members have additional questions, they are going to submit them to you directly.  What I would request is that you submit your answers back to the Committee as well so we can get them into the record so all have access to that.

     With that, thank you again.  Thank you to our guests and Committee members, and with that, we stand adjourned.

     [Whereupon, at 12:26 p.m., the Subcommittee was adjourned.]


The Honorable Geoff Davis


The Honorable Geoff Davis to Mr. Pryde

The Honorable Geoff Davis to Mr. Sheldon


Steven D. Harris