Editor’s Note: OMB opposes the Data Act noting that it “fails to embrace transparency as it currently exists in OMB’s standard way of how it issues guidance,” see OIRA Watch for more about OMB’s views. The Data Act contains exemptions from the Paperwork Reduction Act, the Data Quality Act and the Administrative Procedure Act, key “good government” laws that regulate the regulator. Placing federal data functions beyond the scope of the good government laws, and their public participation and redress provisions, would diminish federal transparency and accountability along with the quality, utility and integrity of the data being collected and disseminated.
From: The Washington Post/Federal Eye
Data mining by private companies could save the government billions if a proposed government transparency act is made law, according to presentations at a forum in Washington on Tuesday.
The vendors, including Microsoft, Teradata of Dayton, Ohio, SAP, the German software giant and MarkLogic of McLean, showed the types of data that they said could be mined from accurate and comprehensive records if the Senate passes the Digital Accountability and Transparency Act.
The act, proposed by Rep. Darrell Issa (R-Calif.), and passed by the House in April and now pending in the Senate would establish an independent board that would track all federal spending. The board would maintain a Web site similar to usaspending.gov, which the Sunlight Foundation says shows $1.3 trillion in spending discrepancies.
Issa and Senate co-sponsor Mark Warner (D-Va.) say the bill could prevent the waste and abuse of federal money.
If government reporting forms were standardized and streamlined, every cent could be tracked, the companies argued.
Teradata suggested consolidating datasets from across several federal agencies, which the company could analyze. Once it’s pooled, federal employees, congressional members and the public could review the data, which the company said could help flag questionable spending or people attempting to game the federal procurement system.
Bill Franks, Teradata’s chief analytics officer, used the grant process as an example of datasets that could be pooled across multiple federal agencies.
Perhaps, he said, an applicant that has been rejected by one agency applies to another. Under the current system, if one agency found a grant recipient fraudulent or wasteful, it might not be able to flag the problem to another agency, Franks said.
But if agencies pooled grant data, a fraudulent grant applicant could be found immediately, he said.
Teradata implemented such a pooled system in Michigan, where 21 agencies have integrated data. Figures show that $1.8 billion was recouped tax amounts from eight states using Teradata’s system: Arizona, Iowa, Maryland, Missouri, New Jersey, Ohio, Oklahoma and Texas.
Teradata’s clients also include Bank of America and Wells Fargo, which integrated data systems after mergers and acquisitions
“The idea of having 12 or 15 banks with disparate systems is not much different from the government,” Franks said.
Some of the companies at the forum worked on the recovery.gov Web site and used it as a model.
IPHIX chief executive Gianluca Garbellotto showed an existing program that uses a Google search application to retrieve data on recovery.gov. A similar plug-in could be used on a new federal spending Web site, which could contribute to ease of use for federal employees, congressional members and the public, he said.
J.J. Kirkpatrick of Level One Technologies in Maryland Heights, Mo., said DATA Act reporting could be done in under five minutes by importing accounting data either automatically or a few clicks after an e-mail alert is sent to an awardee.
Elder Research, based in Charlottesville, Va., has already partnered with the U.S. Postal Service’s inspector general to use predictive analysis techniques to identify contracts that may be prone to fraud. The firm uses internal Postal Service documents and 35 metrics to run its analysis. Antonia de Medinaceli, director of fraud analytics at Elder, said three-fourths of leads the model generates result in audits and investigations by the inspector general. De Medinaceli said Elder’s model demonstrates the possibilities of access to internal records. If they ran their model based on data available on USA Spending, they would only have seven applicable metrics with less accuracy, he said.
Kirkpatrick, director of marketing for Level One Technologies, said a streamlined and standardized reporting process for recipients of federal money could have prevent government spending scandals by flagging questionable spending that agency inspectors general could then look into. He cited the recent General Services Administration spending scandal, where an agency manager named Jeff Neely authorized more than $800,000 in spending on a 2010 employee conference in Las Vegas, exceeding the amount set aside for that conference.
“There will always be Jeff Neelys in government,” Kirkpatrick said. “But with the DATA Act we can shine a lot on their activities.”
Leave a Reply