Honoring Two Unsung Heroes of the Federal Statistical System

statistics-conceptualPerforming public service by working in a governmental or quasi-governmental position is not an enviable thing to do in this age when such work is denigrated by politicians and others who want to “blow up the administrative state.” This column is about two distinguished individuals who have toiled for long periods of time in an area that receives attention only from those who understand the importance of data and statistics to the well-being of a democratic state. This often excludes congressional appropriators who for years have been willing to transfer much-needed funds from agencies like the Census Bureau and to continually underfund agencies such as the Bureau of Justice Statistics.

Katherine Wallman recently retired after 25 years as the chief statistician of the United States. Although recognized by international statistical bodies as a key member of their community, her position within the U.S. bureaucracy is housed in a small office of Statistical and Science Policy (SSP) within a somewhat obscure Office of Information and Regulatory Affairs, in the more visible Office of Management and Budget (OMB), which is part of the Executive Office of the President.

The U.S., unlike countries such as Canada, has a decentralized statistical system. All the major statistical agencies are located within larger departments, e.g. the Census Bureau and the Bureau of Economic Analysis (BEA) in the Department of Commerce, the Bureau of Labor Statistics (BLS) in the Department of Labor, or in Independent Agencies like the Environmental Protection Agency or the National Science Foundation (NSF). The role of SSP and Mrs. Wallman, as she was officially called, was to coordinate these agencies and provide guidance to overall governmental data collection and dissemination policy.

Katherine Wallman
Katherine Wallman
During her 25-year tenure, through three administrations, she rescued her office from the depths to which it had fallen during the Reagan years. It was during those years that I first met Mrs. Wallman. She was then the executive director of the Council of Professional Associations on Federal Statistics (COPAFS), having previously served in the government at the National Center for Education Statistics and at OMB. As a sister organization to the Consortium of Social Science Associations (COSSA), where I was first director of government relations and starting in 1988, executive director, COPAFS had been established to communicate the importance of the federal statistical system and to promote its budgets. The membership of both organizations had considerable overlap. For the next 10 years we worked together on many funding and data management issues and eventually we served on each other’s boards.

At SSP, she formalized the Interagency Council on Statistical Policy, a group of the 13 principal statistical agency heads, which discusses and coordinates policy initiatives for their agencies. Each year when the president releases that office’s budget, the SSP prepares a special Analytical Perspectives section that describes the priorities and funding for each of the agencies. Later, SSP publishes Statistical Programs of the United States Government, which provides much of the same information. In addition, an Interagency Forum on Child and Family Statistics has functioned for many years producing key indicators of children’s well-being in the U.S.

One of the major accomplishments for SSP during Mrs. Wallman’s tenure was the enactment, after decades of discussion, of the Confidential Information Protection and Statistical Efficiency Act of 2002. This law strengthens pledges of confidentiality to foster public trust in the government’s promise to hold data collected for statistical purposes in confidence, while at the same time allowing the limited sharing of business data among the Census Bureau, the BEA, and the BLS as well as updating sampling frames to achieve more consistency among federal surveys conducted by these three agencies.

From time to time, Mrs. Wallman and her agency were confronted with proposals to reshape, usually in the form of more centralization, the federal statistical system. In 2011, President Obama joined the list, which included former Senator Daniel Patrick Moynihan, of putative centralizers by proposing to consolidate four statistical agencies: the Census Bureau, the BEA, the BLS, and the National Center for Science and Engineering Statistics at NSF, as part of an idea to change the Department of Commerce into a Department of Business that would incorporate the Office of the U.S. Trade Representative and other entities. Needless to say, none of this has come to fruition.

During the many years we worked together, especially after she rejoined the government, we talked about working on the inside, e.g., at OMB, as opposed to the outside, e.g., at COSSA. Having started my Washington career at the Department of Education during the transition from the Carter to the Reagan administrations, I was reluctant to re-enter the bureaucracy. She would admonish me by insisting that inside was a good place to be because it provided more opportunities to have a direct influence on policy. In her case, I cannot disagree. She will be greatly missed.

Constance Citro

Constance F. Citro
Constance F. Citro
The Committee on National Statistics (CNSTAT) functions within the Division of Behavioral and Social Sciences and Education within the National Academy of Sciences (NAS), a quasi-governmental agency established by an Act of Congress during the Lincoln Administration. For the past 13 years, Constance Citro has served as CNSTAT’s executive director. She has announced that she will step down at the end of June.

Connie has served the statistical community and her committee well. She has led or supervised the preparation of many reports, one of the major functions of the NAS, which have influenced many federal agencies. She first came to CNSTAT in 1984 as study director for the panel that produced The Bicentennial Census: New Directions for Methodology in 1990.

Other CNSTAT reports include further assessments of the decennial census in both its planning as well as its implementation. There have also been examinations of the American Community Survey to help respond to congressional objections. The committee has also provided NSF’s Science and Engineering Statistics center blueprints for improving its operations. With support from the National Institute on Aging, CNSTAT has assessed the state of data on the elderly both here and abroad. It also has helped the Department of Agriculture refocus its data collection activities.

During Citro’s tenure the committee examined the Survey of Income and Program Participation, which led some to call for its elimination. It survived, albeit in a much improved form. CNSTAT has also taken on the issue of revising the way the nation measures poverty, which has provided the impetus for the development of alternative data measurements. Its most recent report, Innovations in Federal Statistics: Combining Data Sources While Protecting Privacy looks at ways to improve federal statistics for social science research and policy making.

Perhaps most importantly, Connie has edited and now directed the preparation of the Principles and Practices for a Federal Statistical Agency, soon to have its sixth edition published. This document, for which CNSTAT works closely with OMB’s SSP, serves as guidepost for how our decentralized statistical system operates and should operate. It is indispensable, particularly for any new administration. Like Kathy Wallman, Connie Citro has made public service a distinguished enterprise. She too will be missed.


Although their replacements — Nancy Potok, a former Census Bureau deputy director for Mrs. Wallman at SSP, and Brian Harris-Kojetin, who worked at SSP, for Citro as head of CNSTAT — are good people, we should not underestimate the importance of losing these two excellent public servants at this particular time. The new administration has so far, mostly through its budget proposals, suggested that nurturing the statistical agencies is not among its priorities.

For example, the FY 2018 budget blueprint produced by OMB Director Mick Mulvaney makes a big deal about how the Census Bureau would receive a $100 million boost over its current budget to fund the 2020 count. As anyone, who knows anything about the ramp up to the decennial, this puny boost in year eight is totally inadequate to complete the preparations, including a significant marketing campaign necessary to accomplish a fair and adequate enumeration. In President George W. Bush’s FY 2008 budget, the bureau received a 54 percent increase for the decennial compared to the proposed 7 percent for FY 2018. Although other statistical agencies’ budgets have not been addressed yet, given the administration’s priorities one does not hold out much hope for advancements.

In addition, like many agencies across the federal government, most of the statistical agencies are led by acting-directors or commissioners. Erica Groshen’s term as Commissioner of Labor Statistics is done and she is gone. One wonders about a new head of BLS when the chief executive can’t make up his mind whether the unemployment figures are real. The Energy Information Administration has also lost its leader. One suspects their replacements will not come for a while. John Thompson, who leads the Census Bureau, was given a one-year extension past his term. Does this mean we will have a new director just as the count begins?Will the administration’s appointees be distinguished scientists or statisticians, or will they be ideologues committed to the administration’s goal of destroying government as we know it? Stay tuned!

Howard J. Silver

Howard J. Silver served as the executive director of the Consortium of Social Science Associations (COSSA) in Washington, DC, from 1988 to 2013. He has testified before Congress, spoken on federal funding of science at many professional meetings, and written extensively on executive-legislative relations, the federal budget process, and science policy as it affects the social and behavioral sciences.

Skip to toolbar