IN AUGUST THE BUSH ADMINISTRATION fired its director of the Bureau of Justice Statistics in a dispute over a press release about a report on racial profiling. Newspaper editorialists and Democrats in Congress charged the administration with suppressing painful truths. In response, Republican officials apparently spread the word to sympathetic commentators that this was a battle between the administration and the anti-Bush permanent bureaucracy. As one commentator told National Public Radio, "Bush finally clamped down on this guy."

"This guy" was President Bush's own appointee to the directorship of BJS, Lawrence A. Greenfeld, a career criminal justice statistician and longtime deputy director of the agency, a unit of the Department of Justice. Happy as the principal deputy, Greenfeld had not sought the top job, which had always gone to a political appointee serving at the pleasure of the president. Yet, when the directorship opened up after Bush's election in 2000, Greenfeld's stellar reputation within the criminal justice community brought him to the attention of the White House. President Bush nominated him to serve as director, and he was confirmed by the Senate. At his swearing-in ceremony, former Attorney General Edwin Meese praised the accomplishments of Greenfeld and the Bureau of Justice Statistics.

So what are we to make of this incident? Has the administration won a battle against a hostile bureaucracy, or has it suppressed the truth about racial profiling in the United States? A review of the facts compels four conclusions. First, the administration did not try to suppress or manipulate data, though it did seek to deny publicity to uncomfortable facts. Second, its ham-handedness backfired by attracting infinitely more attention to the sensitive racial profiling data than would otherwise have been the case. Third, it cashiered Greenfeld for doing his job in a responsible and, indeed, exemplary way. Finally, not content simply with firing a dedicated public servant, it maligned him and his agency in a way that was deeply unjust, that undermined morale at a model federal agency, and that jeopardized its good work and its reputation within the criminal justice community.

A personal note: I had the pleasure and honor of working at BJS from 1985 to 1990. For the first three and a half years I served as the deputy director for data analysis to Steven R. Schlesinger, who had been appointed by President Reagan to be the first director of the new agency. In June 1988 I began a two-year stint as acting director of BJS, after which I returned to the academy. As a political appointee, I served under, and at the pleasure of, Presidents Ronald Reagan and George H.W. Bush and Attorneys General William French Smith, Edwin Meese, and Richard Thornburgh. When I first went to BJS, Greenfeld was in charge of corrections data; when I was acting director, he served as deputy director. In both capacities I worked with him virtually every day. As I told the New York Times when asked to comment on Greenfeld's firing, "I've never met a finer public servant."

Times reporter Eric Lichtblau broke the story of Greenfeld's firing in a front-page piece on August 24. According to Lichtblau--whose account, to the best of my knowledge, has not been disputed by any of the parties involved--Greenfeld and acting Assistant Attorney General Tracy A. Henke had clashed over the contents of a press release to announce the publication of a major BJS study on traffic stops by police. While the study (which was mandated by Congress) showed that white, black, and Hispanic drivers were stopped at almost identical rates in 2002 (8.7 percent of whites, 9.1 percent of blacks, and 8.6 percent of Hispanics), once stopped, black and Hispanic drivers were two to three times more likely to suffer a negative consequence, such as being searched, handcuffed, or arrested. Henke had insisted that the information on the racial/ethnic disparities be removed from the draft press release, writing "Do we need this?" and "Make the changes" on the copy. Greenfeld refused and the press release was withdrawn. The study itself, however, was released unchanged and can be viewed in its entirety on the BJS website (Contacts between Police and the Public: Findings from the 2002 National Survey, April 2005, at Shortly thereafter, Greenfeld was brought in for questioning by the third highest ranking official in the Justice Department and then called to the White House and asked to resign.

The New York Times story attracted widespread attention. Democrats in Congress jumped on the opportunity to criticize the administration for suppressing data, and several called for an investigation by Congress's Government Accountability Office or the Department of Justice's inspector general. The Times itself carried several follow-up news stories; Maureen Dowd and Frank Rich mentioned the incident in their respective columns; and numerous major newspapers took the administration to task in their editorials.

It didn't take long for the administration to defend itself by attacking its own appointee and his agency. Just three days after the original Times story ran, a well-known conservative commentator told NPR that "the story from the people that I've talked to is that this has much more to do with a war with the permanent bureaucracy in Washington. The permanent bureaucracy in Washington, not just the Bureau of Justice Statistics but elsewhere, does not like the Bush administration very much. And for months now or years now, BJS . . . [has] been leaking to the mainstream media to embarrass Bush. This has been happening a lot, and so Bush finally clamped down on this guy." A convenient line, especially since Henke will soon appear before a Senate committee considering her nomination to a high-level position in the Department of Homeland Security, but the story is false.

THE BUREAU OF JUSTICE STATISTICS is one of the great success stories of the American bureaucracy. Created out of the Law Enforcement Assistance Administration of the 1970s, it employs about 50 people, most of whom are experts in collecting, analyzing, and disseminating statistical data on crime and justice. When I was there, a dozen of the employees, fully a third of the professional staff, held Ph.D.s. In some ways it felt more like an academic department than a government agency. The staff were, and are, extraordinarily dedicated to the agency's mission and extraordinarily successful at accomplishing it. While the agency is expected to collect, analyze, and disseminate data on criminal justice issues of importance to whatever administration is in power, it also enjoys a kind of quasi-independent status, as reflected in the fact that it has always been located in a building separate from "main Justice" and that its director is the appointee not of the attorney general but of the president himself, requiring Senate confirmation.

To say that BJS is highly regarded in the criminal justice community is an understatement. With its small staff, it publishes upwards of 50 statistical reports each year, including several thousand spreadsheets with data on every aspect of the criminal justice system: e.g., criminal victimizations, conviction and sentencing data, prisoner and jail counts, death row, recidivism, and justice expenditures. At a time when many academics are criticized for keeping tight control over their data sets, BJS makes available to the public the underlying data for all its studies, allowing researchers to replicate its findings. The BJS website, which the public accesses as many as 20,000 times each day, is a treasure trove of data on virtually every conceivable aspect of crime and justice. Despite the staggering quantity of data that BJS produces, its record of accuracy is unparalleled.

It would be hard to exaggerate the importance of BJS data, especially in documenting the leniency of the American criminal justice system and thus pointing the way to crime reduction through tougher punishment (although the agency itself steers clear of editorializing and policy recommendations). It was BJS studies that showed that a third of those convicted of felonies in state courts receive no jail time at sentencing; that nearly everyone in state prisons (93 percent) is a convicted violent offender or convicted recidivist (not the first-time nonviolent offender we hear so much about); that even those convicted of the most serious violent crimes serve what many consider unconscionably short prison terms (e.g., about five years for rape); and that two-thirds of those released from state prisons are arrested for a new crime within three years. For example, when BJS studied 270,000 offenders released from 15 state prisons in 1994, it found that these individuals had been arrested for 4.1 million crimes before their most recent imprisonment and another 744,000 crimes within three years of release. And because only a small fraction of crimes ever result in an arrest, we can reasonably conclude that this one group of offenders likely committed 10 million or more crimes through 1997. This is strong evidence indeed that tougher sentencing of recidivists will bear fruit in lower crime rates.

In recent years, BJS has expanded its data collection efforts in innovative ways to shed light on issues of growing importance. For example, its 1998 and 2001 surveys documenting huge backlogs in DNA labs throughout the nation led to a major Bush administration effort to address the problem.

No one is more responsible for BJS's record of achievement than Larry Greenfeld. He created from scratch many of the key data sets, and carefully oversaw others over the years as the agency's deputy director and director. He and his staff pioneered innovative studies on dozens of criminal justice issues and have successfully steered the agency into the Internet age, providing the public and the criminal justice community with a website that is a model of clarity, ease of use, and richness of data. It is hard to imagine what more a federal agency and its director could do to faithfully serve the public trust. And Greenfeld and his staff have done all this while maintaining the highest standards of professionalism.

PART OF THAT PROFESSIONALISM is refusing to agree to a press release that would be deceptively incomplete. BJS's study on traffic stops contained some very good news: White, black, and Hispanic drivers were stopped by the police at virtually identical rates. This is compelling evidence against the stereotype of police racial profiling. True, this encouraging finding was tempered by the fact that once stopped, blacks and Hispanics were much more likely to be searched, handcuffed, or arrested (although the vast majority of all three groups suffered nothing more than the issuance of a warning or traffic ticket). Yet this finding itself does not prove racism in police treatment: There may have been legitimate nonracial reasons the individuals were treated as they were. Indeed, if police disproportionately enforce traffic laws based on race and ethnicity, it is hard to understand why they don't pull over higher proportions of blacks and Hispanics. Serious people ought to care about and reflect upon data of this sort; and they ought not to be afraid to let the public know about it.

Ironically, virtually identical data were contained in the previous BJS study on police contacts and were properly identified in the corresponding press release. In March 2001, BJS released a comparable study of police contacts in 1999. On the second page of the two-page press release (still available on the BJS website), it was duly noted that "Black and Hispanic motorists (11 percent each) were more likely than whites (5 percent) to be physically searched or have their vehicles searched." The information entered the public domain with nary a ripple of controversy.

As Greenfeld told the New York Times in defending the administration that had done him wrong, "There's always a natural and healthy tension between the people who make the policy and the people who do the statistics. That's there every day of the week, because some days you're going to have good news and some days you're going to have bad news." Despite the tension, it is important for political appointees to understand and respect the mission of government statistics agencies. "Quasi-independence" perhaps best captures the relationship.

When I was at BJS, there was a plan afoot to consolidate all the executive branch statistics agencies into one large independent agency, along the lines of "Statistics Canada." I argued to the attorney general that this would be bad for the Department of Justice by placing too great a distance between the line agency and its statistical arm. But if statistics agencies should not be too distant from the operating agencies, neither should they be too close. In the current episode, main Justice overreached, generating an unnecessary public relations flap that ill-served the president. In the process, it wounded and weakened an outstanding public agency--one critical in the long run to the nation's battle against crime--and it mistreated a fine public servant. What seems to be lacking in those responsible for these events is an appreciation of the importance of publicly credible data for sound policymaking, an understanding of the political nuances of bridging the gap between policy and data, and, frankly, some simple human decency.

Joseph M. Bessette is Alice Tweed Tuohy professor of government and ethics at Claremont McKenna College.

Next Page