Written by Lisa Miller      Posted February 7, 2008

Lisa Miller is President and CEO of Dynaxys, a technology and business service provider headquartered in Silver Spring, Maryland. She has extensive experience in the public sector, including managerial positions at the Department of Housing and Urban Development (HUD) and the Federal Savings & Loan Insurance Corporation. She has also been a Director in the Washington office of PricewaterhouseCoopers, where her client list included HUD, the Federal Deposit Insurance Corporation, and other government and nonprofit organizations.

In the 2006 Technology Issues for Financial Executives survey (sponsored by the Financial Executives Research Foundation and Computer Sciences Corporation), nearly 20% of respondents identified upgrading or replacing legacy systems as a critical technology concern. But only about 1% cited evaluating the adoption/use of XBRL. How ironic then that these two critical issues, holding down opposite ends of the concern spectrum, would be paired up in a highly successful endeavor that puts a federal government agency at the forefront of innovative applications of XBRL General Ledger (XBRL GL).

It’s true and it’s a real success story that other federal agencies as well as organizations still strapped with legacy systems could benefit from. In short, the Federal Housing Agency (FHA), an agency within HUD, used XBRL GL to create interoperability between disparate legacy financial feeder systems and the department’s ledgers and control systems.

A federal agency an early adopter. Here’s how the FHA used XBRL GL combined with best practices to standardize data flowing between multiple systems and move responsibility for data integrity and utility from the consumer to the producer.

The FHA has as many as 20 different financial feeder systems supporting its insurance operations. They were all built to support commercial, not federal, accounting standards. It was critical that FHA find a way to improve the interfaces, data management, business processes, and controls between these commercially based systems and the FHA general ledger. At the same time, the agency needed new interfaces to other federal systems to improve internal controls and ensure the agency had current, accurate, and complete financial and procurement information.

This project focused on two subsidiary ledgers that were commercial, transaction-based systems: one for loans and one for properties that report collection and disbursement results for more than $3 billion in assets held by the government. To transform the data from a commercial chart of accounts into the government’s standard, the FHA controller had been using a data warehouse (to receive and translate the financial information) and complex posting models before uploading the data into the FHA general ledger.

Translation errors and the effort required to resolve them produced an unending nightmare. The monthly closing typically took 7 to10 business days. Year-end close took between 30 and 45 days. Inefficiencies, rising expenses, and frustrations ran high.

Thanks to the freely available XBRL GL framework, FHA was able to turn this familiar situation around. With the XBRL GL taxonomy, one can represent any data found in charts of accounts, journal entries, or historical transactions (whether they are financial or nonfinancial) without requiring a standardized chart of accounts to gather information. With XBRL GL a business does not have to change the way it represents data. And better yet, XBRL GL facilitates data exchange among software applications.

XBRL GL presented an obvious solution to FHA’s data problems:
First, XBRL has emerged as an effective framework to reduce the cost and effort required in information exchange. It was designed to allow financial data to be part of an information supply chain.
The XBRL taxonomy (i.e., dictionary) defines a common language that specifies the tags (words) to be used, their semantics (meaning), how they are defined (types of data, structure, and relation to each other), and the rules and formulas they must adhere to.
Further, XBRL provides a metadata structure that can be used across an organization. (XBRL delivers on the best practice dictum: Provide the data once — use it many times.)
At each step in the supply chain, data need to be viewed, analyzed, and manipulated for a variety of purposes while its integrity is kept intact. XBRL achieves this and other fundamental benefits such as accuracy, consistency, efficiency, reuse, flexibility, traceability, and visibility as data move through the supply chain.

In the FHA project, a metadata structure for data exchange was established so that legacy data could be standardized once and reported to multiple systems. The metadata structure also provides enhanced data integrity. Data are validated before they are transmitted. XBRL was used to standardize collecting, obligating, and disbursing data coming from the various financial feeder systems. Data were filtered or summarized based on the system they were feeding, and then sent to other systems. Performing full audits is relatively easy now because the audit trail for the detail is linked to the data through tags.

With a more reliable, accurate, and consistent crosswalk, the FHA improved cash management by eliminating posting errors and the subsequent rework or complicated after-the-fact adjustments. Reconciliations are done on the spot. Month-end and year-end closing of the books takes days not weeks. And post-closing adjustments or other rework is a thing of the past.

Today, FHA transmits clean, sound data from its legacy systems. Data validity has been consistently measured at 100%. The data warehouse middle layer is gone, and data flow directly from the source system into the general ledger, improving data management practices.

This project represents not only the first use of XBRL GL in a federal agency, it represents the first use of XBRL GL for data interchange in any organization in the United States. One might assume: If it can be done at HUD, it can be done anywhere.

Add to Del.icio.us | Digg this

This entry was posted on Thursday, February 7th, 2008 at 9:37 am and is filed under General, . You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.