Advocate for Federal Data Standards Cites Analytics Opportunities

by   |   September 24, 2012 6:21 pm   |   0 Comments

IMG 1689 150x150 Advocate for Federal Data Standards Cites Analytics Opportunities

Hudson Hollister of the Data Transparency Coalition

WASHINGTON—If the federal government would only adopt data standards, it could avoid risky grants, catch more cases of fraud earlier and streamline regulatory reporting,  saving taxpayers and the government both time and money.

That was the message that Hudson Hollister, executive director of the Data Transparency Coalition, delivered at the Predictive Analytics World Government Conference held here on Sept. 17 and 18.

Noting that representatives from several federal offices of inspectors general were in attendance, Hollister stressed that data standards would enable government to better police federal grants and contracts, prevent misuse of government funds and avoid scandals. He explained various data standards that are being proposed or adopted, cited a few examples of how government is starting to use them and reviewed legislative attempts to mandate standards.

Today, there are large differences among federal agencies in how they track payments to contractors and grantees, he said. Standard data formats would mean not only that data from one agency could be easily compared to another, but also that any given agency could perform a wider range of data analytics on its own information more easily.

Hollister reviewed different types of what he called “identifiers.” One of the benefits of data standardization, he said, is the legal entity identifier, which would make it possible to compare, compile and match data collected by different agencies. This identifier is being developed by U.S. Treasury’s Office of Financial Research, he said. If the U.S. Securities and Exchange Commission and federal government purchasing officials were using such an identifier, they could automatically match up a public company’s SEC reports with its application for a federal loan, for example.

Related Story

Analytics, technology companies lobby for data standard and federal spending portal.

       Read more»

Such a comparison could have identified that a loan to the startup solar panels maker Solyndra was high risk, he said. Solyndra filed for bankruptcy protection in 2011, about two years after winning approval for $527 million in federal loan guarantees.

“By tying together filings that companies make with different entities, the legal entity identifier will open up a new world of predictive analytics possibilities,” he said. The first legal entity identifiers are expected to be available in March, according to Hollister.

Another type of identifier is the uniform award ID, which was developed by the Recovery Accountability and Transparency Board that has tracked spending of federal stimulus funds since February 2009. The uniform award ID was the first time that an identifier has been used to track grants and contracts, Hollister said. “The result was a far lower fraud and error rate than we see across the federal government as a whole.” In July testimony before Congress, the controller of the Office of Management and Budget said that agency was studying the feasibility of developing a uniform award ID for broader use in the government.

Markup languages and data models are another area of potential standardization. The SEC and the Federal Deposit Insurance Corporation (FDIC) have already adopted XBRL, a variation of extensible markup language (XML), for financial reports they require. In addition, a standards organization called the Object Management Group is working with the Enterprise Data Management Council to develop the “financial industry business ontology,” a data model that could standardize how complex financial instruments are expressed.

That model could enable both government and industry to track things like derivatives and mortgage-backed securities without having to “hunt through the legalese of thousands of contracts from hundreds of entities,” said Hollister. For regulators, “that means that predictive analytics can get into the terms of contracts. They can get into the instruments that were at the center of the financial crisis.”

Hollister also ticked off various legislative efforts to mandate data standards. In addition to the Digital Accountability and Transparency (DATA) Act, which was re-introduced in the Senate on Sept. 21, he listed the following:

Standard Business Reporting Legislation: Regulatory agencies often request the same information, but it’s reported in different ways and so cannot be shared. For example, 80 percent of what’s reported to the Bureau of Economic Analysis is also reported to the SEC, he said. With Standard Business Reporting, companies would only report this information to the government once. It has already been adopted in Australia, Netherlands, the U.K. and Belgium, but it will probably be a decade or more before the U.S. adopts it, Hollister said.

Financial Industry Transparency Act: An attempt to add this provision for creating financial data reporting standards to the Dodd-Frank Wall Street Reform and Consumer Protection Act failed. The Data Transparency Coalition will continue to push for such a standard, he said.

Public Online Information Act: Bills currently in both the House and the Senate would require that any data that is public be posted online in a standard format.

With the possible exception of the DATA Act, Hollister held out little hope for much movement on these mandates in the near term. “There hasn’t been much notice of the need for data standardization either in Congress or the White House,” he said.

Tam Harbert is a freelance writer based in Washington, D.C. She can be contacted through her website.





2f4030c6 41e3 4507 9efb 8be1e3aad05d Advocate for Federal Data Standards Cites Analytics Opportunities



Tags: , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>