OMB’s Commitment to Data Quality: Too Little, Too Late

by

Last week, OMB Controller Danny Werfel appeared on a panel discussion on potential Recovery Act applications for 2012. In the course of the discussion, Werfel also expressed concerns about the DATA act (which Sunlight supports), and that the White House was unwilling to embrace it. He specifically stated:

DATA Act takes guidance writing away from OMB and gives it to a newly stood-up entity, the commission, which creates a new layer of regulation, two sets of rules and additional regulatory complexity.

You can get a good rundown of Werfel’s criticisms of the DATA Act and a rebuttal of them over at the Data Transparency Coalition blog, but in my opinion, taking the responsibility for data quality away from OMB is one of the best things about the DATA act. They have proven time and again that their commitment to quality spending data is superficial at best. Here are some examples.

Item 1: The Missing Data Quality Dashboard

In an OMB memo dated April 6th, 2010, Jeff Zientz outlined the plan for a publicly available data quality dashboard that measured each agency’s progress towards improving their spending data quality. Designed to be a contribution to the Open Government Directive plans, the memo required incremental improvement in the data from each agency, culminating in 100% of awards being reported on time, completely and accurately by the fourth quarter of FY 2011. Additionally, the memo announced the creation of a USASpending.gov Change Control Board, which was to oversee the collection and presentation of the data on USASpending.gov. In fact, it seems it would have many of the same responsibilities as the FAST board in the DATA Act.

To date, we have found some submissions from agencies that outline their broad plans to improve the quality of their data, but have not seen any evidence of the dashboard, nor any public reference to the Change Control Board. And the data certainly hasn’t improved much in the last two years, much less achieved 100% accuracy, timeliness and completeness. OMB seems to be upset that Congress has adopted their idea, but has chosen to implement it themselves since OMB never did.

Item 2: (Not) Using the Financial Audit Process to Reconcile USASpending Data

During a hearing of the House Committee on Oversight and Government Reform in March 2011, Werfel stated that part of the problem with the quality of the data is that the public spending information is not tied in to the financial statements that undergo audits. We agree that having two divergent reporting streams is another driver of the data quality problem. Because this data is not audited, there is very little incentive to ensure that it is correct.

At the Recovery Act panel, Werfel contended that legislation was not necessary for OMB to move the ball forward with agencies in regards to the quality of their spending data. Earl Devaney, former chairman of the Recovery and Accountability and Transparency Board, also present at the panel, disagreed with him, saying, “The government moves fastest when Congress acts and there are concrete data sets. If we leave it up to the agencies, it will never get done.”

Funnily enough, Werfel seemed to agree with Devney’s sentiments at the Oversight Committee hearing mentioned above. In response to Rep. Cummings’ question:

… Is it, from what you can see, problems stemming from folks who are just not doing what they are supposed to do? Is there a bigger hammer that needs to be hammered? …

Werfel responded:

… To answer your question, I think a bigger hammer is needed. I think what you have today is a dichotomy that was reported on the first panel that is accurate. We have what I would argue is a very robust financial statement audit process that exists today … We have 20 of the 24 major agencies in Government receiving a clean audit opinion. A lot of effort goes into scrutinizing, to the 10th decimal point, the numbers that go on our balance sheets and our other basic financial statements. And we have developed a very robust process in response that is moving forward and achieving important things in terms of financial reporting reliability. That robustness does not exist with respect to the information as reported on USAspending.gov. The spend information, as we call it, is not completely wired into the financial statement audit process. We think that we need to look at that audit process and that reporting model to potentially realign some of that audit scrutiny around spending information. I think you would see a difference in results if agencies felt the accountability of an auditor’s eye on these issues.

Werfel (and by extension OMB) seem very cognizant of steps that could be taken to raise the level of compliance from agencies. This makes it even worse that they have not been able to take these steps to accomplish improvement in the data in the last six years. And what’s more, they have actively obstructed our attempts to find out more about the quality of the data via the FOIA process.

Item 3: Obstructing Public Attempts to Learn About the Quality of USASpending Data

After examining the data quality of grants in USASpending.gov in our Clearspending project, we wanted to extend that analysis to contracting data. The only source of data we knew of were the “FPDS Data Quality reports”. These reports have been mandated by OMB for the last four years and require agencies to detail what amount of their contract spending was in USASpending.gov as well as various data quality metrics about each field in the database. OMB delayed responding to our FOIA for these reports for over a year, until they were questioned about it by the House Oversight Committee. Then they responded with an extremely truncated version of the data, and for only one year — they refused to provide data for additional years. However, we were able to get this data by filing a FOIA with each individual agency, suggesting that there were no legitimate objections to the FOIA being fulfilled, just simple obstructionism by OMB.

The reforms posited by the DATA Act and the creation of a commission to implement them are right in line with what OMB has advocated for in the past. They appear to now oppose those changes because they are no longer slated to be the ones in charge of them. That’s not a good enough reason to oppose a bill that contains not only useful reforms for spending data, but for government data in general.