Authors
Tasha Austin, Principal, Deloitte & Touche LLP
David Mader, Chief Strategy Officer for Civilian Sector, Deloitte Consulting LLP (January 2017–Present) Former Controller, Office of Management and Budget (June 2014–January 2017)
Mekala Ravichandran, Manager, Deloitte & Touche LLP
Matt Rumsey, Independent Consultant
Contents
Introduction
Milestone legislation revamping data policies across federal agencies was enacted into law in January 2019.[1] The Foundations for Evidence-Based Policymaking Act (Evidence Act) (PL 115-435), which includes the Open, Public, Electronic, and Necessary Government Data (OPEN) Act, received overwhelming bipartisan support in Congress.[2] The Evidence Act and OPEN collectively recognize that using data is essential for the practice of evidence-based policymaking to occur. The over- whelming bipartisan support in Congress suggests policymakers are also now demanding that agencies also become better positioned to increasingly and meaning- fully use high-quality evidence to inform key decisions.
OPEN builds on previous federal open data laws by adding an expectation that the federal government’s data will be open and accessible, by default, unless there are restrictions or limits such as for protecting confidentiality and national security. The across-the-board expectation applies to government agencies collecting weather, health, crime, and even financial data. Because of the sweeping scope of the legislation, it promises to transform how agencies report, publish, and ultimately use the information.[3] OPEN is positioned to help modernize the federal government’s data processes to meet the needs of policymakers and the American public in the 21st century and beyond.
In addition to building an expectation that government data are open and accessible, OPEN establishes Chief Data Officers (CDO) and a Chief Data Officers Council (CDO Council), clarifies definitions, and requires data inventories at each federal agency.[4] These key provisions build on and reinforce existing open data efforts in government including the DATA Act, without reinventing the wheel.
The passage of OPEN and its implementation complements the Trump Administration's Federal Data Strategy developed as part of the President’s Management Agenda. The Federal Data Strategy is an ongoing effort co-led by the Office of Management and
Budget (OMB), the White House Office of Science and Technology Policy (OSTP), the Department of Commerce, and the Small Business Administration (SBA) to develop a 10-year, long-term strategy for better using government data to improve how agencies fulfill their missions, serve the American public, and use their resources efficiently.[5]
Implementing OPEN presents challenges for federal agencies, both technical and cultural. This white paper draws on the experiences of relevant stakeholders, informed by implementation of other open data laws, to identify and elaborate on the challenges for federal agencies in developing strategies and practices to support OPEN’s goals. The perspectives of a cohort of federal CDOs—those who will be at the forefront of implementing the law within their agencies—are incorporated. Finally, considerations are presented for ways to enable successful implementation of the historic OPEN Government Data Act.
A Brief History of the OPEN Government Data Act
The story of open data in the federal government is a long one, stretching back to at least 1966 and the passage of the Freedom of Information Act (FOIA).[6] More recently, the rise of the internet and new information technologies brought increased attention to, and a better under- standing of, the value of government information for oversight, analysis, and innovation. A movement began to guarantee broad public access to government information, not simply as physical documents, but as machine-readable digital data.[7]
The administration of President Barack Obama embraced the push toward open government data, taking a series of steps to promote open data activities in the federal government. The Obama administration efforts resulted in an open data Executive Order,[8] Open Data Policy guidance to agencies,[9] and an initiative called Project Open Data.[10]
The Trump administration continued a focus on improving government data management through the President’s Management Agenda and the development of the Federal Data Strategy.[11]
OPEN was originally co-introduced on April 26, 2016 by Representative Derek Kilmer (D-WA) and Senators Brian Schatz (D-HI) and Ben Sasse (R-NE).[12,13] The legislation was inspired by federal open data efforts and, at its core, aimed to codify several of the Obama administration policies.[14] Ultimately, it would do that, and more, by enshrining in law a strong open data definition and an official presumption that non-sensitive “government data assets made available by an agency shall be published as machine-readable data...in an open format, and...under open licenses.”[15]
The presumption toward open data, along with the more technical data management aspects of OPEN, will support open data efforts in a variety of domain-specific contexts across the federal government. For example, it will reinforce existing laws like the Digital Accountability and Transparency Act (DATA Act), providing additional legal backing to help open data initiatives to actually result in accessibility of the intended data.[16] It should also lay a foundation upon which to build better open data mechanisms in areas as diverse as grant reporting, and legal entity identification, for example.[17] Further, OPEN should help root out proprietary data standards and file formats that currently limit the utility of key government data assets.[18]
While the initial version of OPEN was given consideration in 2016, the ultimate momentum leading to passage occurred when former House Speaker Paul Ryan (R-WI) and Senator Patty Murray (D-WA) incorporated the bill in the Evidence Act. This step also boosts federal evidence- based policymaking efforts while providing privacy and security protections for confidential data.[19] OPEN and the Evidence Act benefited from strong bipartisan support in both chambers of Congress. The inclusion of OPEN in the Evidence Act was natural because of the shared goals for improving the American public’s access to reliable information.[20] Collectively, the law aims to vastly improve efforts to focus on transparency, with benefits for government accountability, and also enable increased data uses inside government, among researchers, and for the business community.
What Does the OPEN Government Data Act Do?
Inventories of government data to improve discoverability
OPEN provides an important boost for federal data management by ensuring that agencies are properly tracking, managing, and publishing their datasets. In the coming years, this should lead to increased data quality and higher rates of data use and reuse across government and for society.
OPEN expands efforts launched under the Obama administration,[21] codifying key data management requirements, including data inventories and strong metadata standards. The requirement for a comprehensive data inventory includes all data assets created by, collected by, under the control or direction of, or maintained by the agency.[22] The law also requires those data assets to be supported by strong metadata, which should help ensure that agency data assets are functionally useful to interested parties.[23] Finally, OPEN encourages agencies to engage with their data stakeholders, working to increase public and agency use of government data, and making government data more discoverable.[24]
Openness by default to improve access to non-sensitive data
For the first time, OPEN codifies a presumption that non-sensitive government data should be open by default, while ensuring privacy protections and other potential risks are adequately managed.[25] The law includes mandates that open data include machine-readable information with an open license.[26]
OPEN’s definition of “data” as “recorded information, regardless of form or the media on which the data is recorded” and its explicit requirement that data be disclosed if it would be available under the FOIA covers a wide range of government information.[27]
Further, by requiring data in machine-readable formats, the law directs agencies to move beyond paper documents, out-of-date filing systems, and unnecessary information silos toward data that is easier to share and use.[28]
Finally, OPEN applies licenses, or “a legal guarantee that a data asset is made available...at no cost to the public...and with no restrictions on copying, publishing, distributing, transmitting, citing, or adapting such asset,” to government data.[29] This guarantee should make government data useful beyond the agency and vendor silos that currently hold it back.
Leadership from Chief Data Officers to Prioritize Data Quality
Of course, strong data management standards and definitions will likely only thrive under strong and engaged leadership. The U.S. Commission on Evidence-Based Policymaking found that “many departments lack senior leadership focused on data management...[and] urged agencies to develop more meaningful collaboration” in that area.[30] The Commission also unanimously recommended that federal agencies establish senior leadership positions to prioritize data quality and data management activities. To this end, OPEN includes the mandate that every covered agency appoint a “nonpolitical appointee... as...Chief Data Officer of the agency.”[31] It also creates a Chief Data Officer Council. Each agency’s CDO is expected to serve on the Council.[32]
CDOs will focus on data management and coordination, ensuring that their agencies establish effective procedures and work to ensure data that is transparent, accurate, and of high quality.[33] The CDO Council will serve as a forum for sharing best practices, promoting data sharing, strengthening the data workforce, and identifying ways to improve access to government data.
By requiring each federal agency to appoint a CDO and create a CDO Council where data leaders can come together to share learned lessons, leading practices, and bridge silos across agencies, OPEN supports its strong definitions and data management structures with executive level leadership at agencies across the federal government.
Statutory Requirements for Implementation
Which offices are leading implementation?
OPEN implementation will be led by the White House OMB and supported by stakeholders including the General Services Administration (GSA), the Office of Government Information Services (OGIS) at the National Archives, the Government Accountability Office (GAO), and the growing ranks of federal CDOs.[34]
In addition, in July 2019, OMB issued guidance to federal agencies in implementing some aspects of OPEN, such as designating qualified CDOs, delineating the roles and responsibilities of CDOs based on OPEN requirements, and initial development of open data plans.[35] Beyond the legal requirements and guidance documents, however, agencies will need to determine how to best execute the obligations under OPEN to fulfill agency missions and goals.
OPEN Government Data Act Implementation Milestones
Insights from Current Chief Data Officers
The CDO role in federal government agencies will be at the forefront of OPEN’s implementation. Three CDOs, whose roles were established before passage of OPEN, shared their perspectives about implementation in the context of the broader Evidence Act, as well as ongoing efforts like the Federal Data Strategy. The CDOs represented the National Oceanic and Atmospheric Administration (NOAA) in the Department of Commerce, Department of Defense, and Department of Health and Human Services. Each provided context about their current experiences and how they see their roles evolving since OPEN’s enactment. Each CDO also identified challenges, risks, and opportunities likely to be encountered based on their past experiences as their agency implements the new law.
Our semi-structured interviews with three CDOs highlighted several specific challenges presented by OPEN implementation that collectively or individually could contribute to limited effectiveness of the new law if not addressed. First, the law places responsibilities on CDOs, though additional guidance may be needed to further explicate the authorities and how they relate to other senior government officials, in short, CDOs need a seat at the leadership table. Second, CDOs need resources to fulfill their responsibilities. Third, there is a risk that agency stakeholders may fail to see the benefits of better data management and data sharing within and across agencies. This could reinforce existing management and operational silos, or even result in new ones. Finally, the federal workforce has to evolve to fit the increasing importance of data and evidence. If these challenges are not addressed, CDOs warned that implementation will likely create nothing more than another compliance exercise for agencies, falling short of realizing the vision and goals of OPEN and the Evidence Act.
Responsibilities and Roles
While OPEN places responsibilities on CDOs, those that we interviewed expressed the need for further guidance on their authorities.[36] They were particularly concerned about falling into a compliance role, which would ultimately limit the value that CDOs are able to provide to their agencies.[37]
CDOs are well positioned to make the case for data as a strategic asset that is useful by stakeholders in and out of government. They should have a seat at the table with other senior agency leaders as decisions are made.[38]
Internal Buy-in and Data Use
OPEN and the Evidence Act have the potential to promote effective internal data use, while tearing down silos across the federal enterprise and bring data practitioners together in new ways.[39]
To ensure that OPEN’s data governance ambitions are fulfilled, agencies need to embrace their own data assets. Effective data governance is most useful to internal stakeholders if it leads to increased data use within the organization.[40]
However, it is not an easy or small task. OPEN will likely gain internal agency buy-in sustained over time most readily when implementation aligns with existing agency goals, mission requirements, resources, and is supported by respective agency stakeholders.[41]
Resources
To effectively promote the value of data as a strategic asset and increase internal data use, CDOs will need proper resources. Right now, there is a lack of obvious budget to support CDOs as they work to make change.[42]
From a longer-term perspective, it will be important to consider how resource needs change as data use increases within government and by outside stakeholders.[43]
Building a Data Workforce
While top down leadership is important, implementing OPEN and fulfilling the goals of the Evidence Act will take effort from a wide range of data practitioners. The government’s personnel systems are not currently designed to build, support, and promote a data workforce.[44] There have also been broad calls to reform government’s workforce as the talent landscape evolves, including to plan for emerging technologies that are increasingly relevant in order for government to be effective.[45] A number of changes can be implemented to this effect, including better training, more certifications, and explicit career paths for data professionals within the federal government.[46]
Considerations and Next Steps
CONSIDERATION 1
OMB and Agency Senior Leaders Should Empower Chief Data Officers.
The CDO community needs to be properly empowered to effectively implement new directives about data quality and management in government. The CDOs and these initiatives can enable government agencies to more effectively fulfill their missions and serve the American public by using data to improve outcomes. Empowerment means that both political and career leaders in agencies should view the CDOs as an important element of the agency infrastructure, should trust the individuals to help fulfill the agency mission, and should be given appropriate support and flexibility to fully executive their prescribed duties. OMB’s July 2019 guidance helps set the stage for recognizing the essential role that CDOs will play in agencies and—where they exist—the essential role some already have.[47]
Initially, agencies can take steps to support new CDOs by inviting them to participate in senior leadership discussions, integrating CDOs into decision-making processes for major regulations and policies, and recognizing the new role officially as part of agency’s C-suites. Empowerment for the important CDO role also likely means that the individuals serving in the position should be free from other obligations and responsibilities. Agency leaders can support faithful execution of the responsibilities delineated in OPEN by ensuring the CDOs are not dual-hatted officials, for example, someone who also serves as the Chief Financial Officer or Chief Information Officer.
CONSIDERATION 2
The CDO Council Should Identify and Replicate Leading Practices and Success Stories.
To view OPEN’s data management requirements as more than a compliance exercise, agencies need to believe in the internal value of their data. One way to establish and articulate the value of an agency's data is to share leading practices and success stories.
Leading practices and success stories already exist in a variety of agencies, but still remain siloed and have not been formalized across government. For example, NOAA has high demand for its open data products. To fulfill this demand and achieve scale, NOAA launched partnerships with private actors and academia, providing continued access to government data while generating added value for private stakeholders and the American economy.[48]
CONSIDERATION 3
Agencies Should Establish Cross-agency Data Governance Processes to Work Across Silos.
The emerging cohort of CDOs should rapidly build a community that can work collaboratively with other topical and functional parts of federal agencies. Agencies can support building a community of data practitioners to share leading practices and open up opportunities for cross-agency collaboration by ensuring representatives from different parts of the agency regularly convene to address important data quality and access issues. Regular cooperation through a consistent governance process can not only identify issues, but it can facilitate idea-sharing and encourage relationship building between data leaders and specialists.
The governance process will likely, itself, also result in developing a larger community of practice for data practitioners, supporting sustained culture changes consistent with the intent of the Evidence Act. Instead of treating the performance, statistics, evaluation, and data science communities as separate entities, the governance process can help facilitate dialogue about shared challenges and solutions. The CDO Council may also be positioned to provide a forum and structure to build community and ease collaboration between agencies.
CONSIDERATION 4
Agencies and Congress Should Fully Resource Chief Data Officers in Key Agencies.
In order to be successful, CDOs will need to have the necessary resources to fulfill the obligations in OPEN and the Evidence Act. Underfunding the CDOs could potentially limit their ability to be empowered change agents within agencies. In practice, a small amount of funding may likely go a long way in supporting the initial development of the new leaders. Congress and the Executive Branch should work together to appropriate and allocate sufficient resources for the new CDOs to achieve their missions.
Providing full resources also means the CDOs need to be provided the time and flexibility required to fulfill the obligations and duties prescribed in OPEN. For most agencies, assigning a part-time CDO will be inappropriate and undermine the intent and spirit of the law. Agencies should aspire to recognize the new leadership role can serve as a complement to other C-suite positions, reinforcing and strengthening the capabilities of Chief Financial Officers, Chief Information Officers, Chief Performance Officers, Chief Evaluation Officers, and other agency leaders.
CONSIDERATION 5
Agencies Should Support the Data Workforce with Trainings, Certifications, and Updated Job Descriptions.
Top-down leadership from CDOs is vital, but will need to be supported by an engaged workforce of data practitioners across the federal enterprise. While every agency’s human resources team can support the data workforce, the Office of Personnel Management is uniquely positioned to help establish and maintain a qualified workforce by embracing the role of data and evidence. This can be accomplished by creating new job classifications for data roles and enhancing existing ones. Additionally, new trainings and certifications should be explored to ensure that data practitioners are continuously improving their skills and staying up-to-date on the latest techniques.
Conclusion
The OPEN Government Data Act lays out a vision for well-managed, high quality, accessible, and useful government data. If successful, its implementation can build on lessons learned from previous efforts by expanding the scope of open government data, improving its management, and empowering those that want to make use of it both inside and outside the federal government. Luckily, steps are already being taken in the right direction, as shown by the Trump Administration’s commitment to leveraging data as a strategic asset and OMB’s collaborative, government-wide approach to data management.
However, that vision is not guaranteed. To achieve it, the new CDOs working across federal agencies should be empowered to move beyond compliance and fully leverage their data assets. Existing silos will have to come down, new communities will have to be built, and leading practices will have to be identified and shared. In the near term, the resources and capabilities should be fostered to ensure success.
References
1.Congress.gov, Actions Overview H.R.4174 -- 115th Congress (2017-2018), (accessed July 15, 2019) https://www.congress.gov/bill/115th-congress/house-bill/4174/actions.
2. Ibid.
3. Data Coalition, Foundations for Evidence-Based Policymaking Act of 2018 Fact-Sheet, 2019. (accessed July 15, 2019), https://www.datacoalition.org/wp-content/uploads/2019/05/Evidence-Act-Web-version-2019.pdf.
4. United States Congress, The House of Representatives, The Foundations for Evidence-Based Policymaking Act, TitleII-Open Government Data Act, (accessed July 15, 2019), https://www.congress.gov/bill/115th-congress/house-bill/4174/text#toc-H8E449FBAEFA34E45A6F1F20EFB13ED95.
5. Federal Data Strategy, 2019, https://strategy.data.gov/.
6. Freedom of Information Act, P.L. 89-847, https://www.govinfo.gov/content/pkg/STATUTE-80/pdf/STATUTE-80-Pg250.pdf.
7. For more on the original principles of open government data see https://opengovdata.org/.
8. Barack Obama, Executive Order -- Making Open and Machine Readable the New Default for Government Information, Washington, DC: The White House, 2013, https://obamawhitehouse.archives.gov/the-press-office/2013/05/09/executive-order-making-open-and-machine-readable-new-default-government-.
9. Sylvia Burwell, Steven VanRoekel, Todd Park, and Dominic Mancini. Memorandum for the Heads of Executive Departments and Agencies, Open Data Policy, Managing Information as an Asset (M-13-13), Washington, DC: The White House, Office of Management and Budget, 2013, https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/memoranda/2013/m-13-13.pdf.
10. Office of the United States Chief Information Officer, Project Open Data, Washington, DC: 2019. https://project-open-data.cio.gov/.
11. See Federal Data Strategy 2019 and White House, President’s Management Agenda, Washington, DC: The White House, 2018, https://www.whitehouse.gov/wp-content/uploads/2018/04/ThePresidentsManagementAgenda.pdf.
12. Congress.gov, Cosponsors H.R. 5051 -- 114th Congress (2015-2016), https://www.congress.gov/bill/114th-congress/house-bill/5051/cosponsors.
13. Congress.gov, Cosponsors S. 2852 -- 114th Congress (2015-2016), https://www.congress.gov/bill/114th-congress/senate-bill/2852/cosponsors.
14. Derek Kilmer, OPEN Government Data Act, H.R. 1770. Washington, DC: United States House of Representatives Office of Rep. Derek Kilmer, 2017. https://www.datacoalition.org/wp-content/uploads/2017/05/OPEN-Government-Data-Act-One-Pager.pdf
15. Data Coalition, OPEN Government Data Act, https://www.datacoalition.org/open-government-data-act/.
16. Digital Accountability and Transparency Act (DATA Act), P.L. 113-101, https://www.congress.gov/113/plaws/publ101/PLAW-113publ101.pdf
17. Christian Hoehner, The OPEN Government Data Act: A Sweeping Open Data Mandate for All Federal Information, Washington, DC: Data Coalition, 2016, https://www.datacoalition.org/the-open-government-data-act-a-sweeping-open-data-mandate-for-all-federal-information/.
18. Data Coalition, Press Release: Data Coalition Endorses OPEN Government Data Act, https://www.datacoalition.org/press- releases/data-coalition-endorses-open-government-data-act/.
19. Congress.gov, Foundations for Evidence-Based Policymaking Act, https://www.congress.gov/bill/115th-congress/house-bill/4174.
20. United States Congress, Foundations for Evidence-Based Policymaking Act of 2017 Report, Report 115-411. Washington, DC: House of Representatives Committee on Oversight and Government Reform, 2017, https://www.congress.gov/115/crpt/hrpt411/CRPT-115hrpt411.pdf.
21. Data Coalition, OPEN Government Data Act, https://www.datacoalition.org/policy-issues/open-data/open-government- data-act/.
22. Ibid.
23. United States Congress, 2017, 10.
24. Foundations for Evidence-Based Policymaking Act of 2018, P.L. 115-435, https://www.congress.gov/115/bills/hr4174/BILLS-115hr4174enr.pdf.
25. Data Coalition, Foundations for Evidence-Based Policymaking Act of 2018 Fact Sheet, Washington, DC: 2019, https://www.datacoalition.org/wp-content/uploads/2019/05/Evidence-Act-Web-version-2019.pdf .
26. See Title II in the Foundations for Evidence-Based Policymaking Act of 2018, P.L. 115-435. 27. Ibid.
28. Ibid.
29. Ibid.
30. U.S. Congress, Foundations for Evidence-Based Policymaking Act of 2017 Report, 15.
31. Foundations for Evidence-Based Policymaking Act of 2018, P.L. 115-435.
32. Ibid.
33. U.S. Congress, 2017, 14.
34. Ibid.
35. Russ Vought, Phase 1 Implementation of the Foundations for Evidence-Based Policymaking Act of 2018: Learning Agendas, Personnel, and Planning Guidance (M19-23), Washington, DC: The White House, 2019, https://www.whitehouse.gov/wp-content/uploads/2019/07/M-19-23.pdf.
36. Michael Conlin, Chief Data Officer at the Department of Defense, interview by author, June 11, 2019, and Mona Siddiqui, Chief Data Officer at the Department of Health and Human Services, interview by author, June 5, 2019.
37. Siddiqui, interview. 38. Ibid.
39. Dr. Ed Kearns, Former Chief Data Officer at the National Oceanic and Atmospheric Administration (now Department of Commerce Chief Data Officer), interview by author, May 22, 2019.
40. Ibid.
41. Conlin, interview.
42. Conlin, interview.
43. Kearns, interview.
44. Conlin, interview.
45. William Eggers, John O’Leary, and Amrita Datar, The future of work in government, Washington, DC: Deloitte, 2019, https://www2.deloitte.com/us/en/insights/industry/public-sector/future-of-work-in-government.html.
46. Ibid.
47. Vought, 2019.
48. For more, see https://www.noaa.gov/big-data-project.
