February 1, 2000

Mr. John T. Spotila
Office of Information and Regulatory Affairs
Office of Management and Budget
Dwight D. Eisenhower Executive Office Building
17th Street & Pennsylvania Avenue, N.W.
Room 350
Washington, DC 20503

Dear Mr. Spotila:

In November of 1998, in connection with the FY 1999 appropriations law, Congress instructed oMB to promulgate a data quality regulation by the end of that fiscal year, reminding OMB that this was a requirement of the Paperwork Reduction Act of 1995. Five years have now passed since enactment of the Congressional mandate in the Paperwork Reduction Act, and the September 30, 1999 deadline in the appropriations statute has also passed, yet no proposed standard or regulation has been published.

To help the process along, CRE has drafted the enclosed "Model Notice of Proposed Rulemaking on Data Quality" for use by OMB in complying with Congress' mandate. Attachment I. This Model NPRM is in the form of a typical NPRM, containing preliminary and supplemental information (including a detailed discussion of the legal precedents that serve as the basis for the proposed rule), as well as the text of the Proposed Rule itself

In drafting the data quality definitions in the Model NPRM, CRE paid close attention to Congressional language specifying four distinct components of data quality. CRF. examined a range of available precedents and definitions for these (‘our components before drafting the definitions set forth at section 3.b of thc Proposed Rule,

Another issue addressed in the Model NPRM concerns remedies. Once again, Congress has provided guidance by specifying that OMB’s regulation must require that each agency establish a petition process to enable stakeholders to correct inaccurate or misleading data. In drafting the petition procedure contained in the Model NPRM, CRE addressed three key concerns.

  • First, CRE conducted a study of existing petition processes at a key agency, EPA, to determine the petition ‘case load’ presently administered. Attachment 2, The CRE study demonstrates that EPA presently administers upwards of 140 petition procedures. Clearly, the petition mechanism selected by Congress provides a time- and resource-efficient mechanism for resolving most data quality disputes.
  • Second, CRE considered the question of "administrative law standing," i.e., who should have access to the petition process, and who, if anyone, should be excluded? As is amply set forth in the legal discussion in the Model NPRM, CRE examined a range of judicial, statutory and regulatory precedents in crafting a provision that would provide access to petitioners with a concrete stake in the use of controverted data, while at the same time preventing ‘nuisance petitions’ by those with no real stake.
  • Third, CRE has raised the question of who should be the ultimate arbiter of data quality disputes. The tentative proposal adopted in the Model NPRM is to make agency decisions on data quality petitions subject to judicial review under the Administrative Procedure Act.

CRE recognizes that the enclosed draft is just that, a first draft designed to demonstrate that the task is doable and to serve as a basis for initiating a stakeholder review. Accordingly, after obtaining initial comments and revisions from parties familiar with the data quality issue, CRE will publish the Model NPRM on its website and undertake an extensive public comment process. CRE will then prepare a summary of the comments received by CRE’s "interactive public docket," make revisions to the Model NPRM based on the comments, and transmit a revised comment to OMB and other concerned parties. We note that, in connection with the data access regulation, CRE undertook a similar exercise, posting an 80-page data access proposal and commentary on its website and reporting on the public comments received through the interactive public docket.

It has been argued that the task of regulating Federal data transactions is an immense one, and that limited agency resources prohibit OMB from issuing a rule. Yet Congress has already mapped out a relatively simple and straightforward approach in its mandate in OMB Under the congressional approach, the OMB regulation is to contain three elements: definitions of the four components of data quality, a petition process and the requirement that rulemaking agencies to issue conforming regulations.

The continued absence of a data quality regulation places the entire American public at risk. All public stakeholders - including citizens with health or environmental concerns, workers with occupational safety concerns and small businesses with compliance or enforcement fairness concerns - are being deprived of their congressionally-recognized and mandated right to a workable regulation that would clarify when data is fit for use in making decisions affecting the daily lives of these stakeholders.

In the connection it should be emphasized that all parties - the Federal Government as well as regulated businesses and the public at large - suffer from the present situation in which data infirmities go undetected arid unresolved. Bad data make for bad regulatory decisions. Establishing a date quality regulation does not benefit any one group or interest as against another. Rather, it creates a level playing field that will improve the quality of Federal regulations in virtually e very field.

We understand that the deadline for promulgation of a data quality rule was October 1, 1999. We also note that the data access rule has been promulgated. Now that the data access rule has been issued, we hope that OMB will be able to devote itself to thc data quality issue. As is set forth above, OMB is faced with two major tasks: drafting a proposed rule and substantial public comment review. To assist in these tasks. CRE will post the Model NPRM on CRE’s interactive public docket (at www.TheCRE.com). after which, as mentioned above, we will analyze and summarize the comments received through the CRE interactive docket, and prepare recommendations based on the same.


Jim Tozzi