Read an Excerpt
Shades of Mercy
By Dafna Linzer, Jennifer LaFleur
ProPublicaCopyright © 2012 ProPublica Inc.
All rights reserved.
Presidential Forgiveness Heavily Favors Whites
In the last decade, hundreds of African Americans sought the final recourse of the criminal justice system: a pardon from the president of the United States. All but seven of their petitions were discarded before reaching the White House. Gatekeepers at the Justice Department favored white applicants over minorities by a factor of nearly four-to-one—even when objective factors, including the crimes, sentences, criminal histories and subsequent records of conduct were held constant.
The White House never saw the petition of Denise Armstead, an African American beauty shop owner in Little Rock, Ark., who had long since paid a $3,000 fine for underreporting income on a tax return. Yet on the Justice Department's advice, the president pardoned Margaret Leggett, a white woman from the same city who attempted to defraud the IRS of more than $25,000 in tax refunds under several false names. A black, first-time drug offender—a Vietnam veteran sentenced to probation for possessing 1.1 grams of crack—went into the discard pile. A white, fourth-time convict who served prison time for selling over a kilogram of methamphetamine won a pardon. Without exception, every drug offender pardoned during the Bush administration was white.
Former Chief Justice William Rehnquist called pardons "the 'fail safe' in our criminal justice system," the final opportunity to bestow mercy or cure injustice. That is not what ProPublica found. Out of sight of the public and even the president, the Office of the Pardon Attorney has made all-but-final decisions in thousands of pardon cases without disclosing its deliberations, its idiosyncratic standards or basic statistics about its work to anyone.
For those reasons, it had not been possible to examine presidential pardons. The pardon office has guarded nearly every detail about an applicant and doesn't make public its recommendations to the president. But the results of its work are now clear.
Current and former officials at the White House and Justice Department said they were surprised and dismayed by the racial disparities in pardons.
"I'm just astounded by those numbers," said Roger Adams, who served as head of the Justice Department's pardons office from 1998 to 2008. He said he could think of nothing in the office's practices that would have skewed the recommendations. "I can recall several African Americans getting pardons."
ProPublica's review examined what happened after President George W. Bush decided at the beginning of his first term to rely almost entirely on the recommendations made by career lawyers in the pardon attorney office.
The office was given wide latitude to apply subjective standards, including judgments about the "attitude" and the marital and financial stability of applicants. No two pardon cases match up perfectly, but records reveal repeated instances in which white applicants won pardons with transgressions on their records similar to those of blacks and other minorities who were denied.
Senior aides in the Bush White House say the president had hoped to take politics out of the process and avoid a repetition of the Marc Rich scandal, in which the fugitive financier won an eleventh-hour pardon tainted by his ex-wife's donations to Democratic causes and the Clinton Presidential Library.
In a statement, the Justice Department said the pardon process takes into account many factors that cannot be statistically measured, such as an applicant's candor and level of remorse.
"Nonetheless, we take the concerns seriously," the statement said. "We will continue to evaluate the statistical analysis and, of course, are always working to improve the clemency process and ensure that every applicant gets a fair, merit-based evaluation."
Bush followed the recommendations of the pardons office in nearly every case, the aides said. The results, spread among hundreds of cases over eight years, heavily favored whites. President Obama—who has pardoned 22 people, two of them minorities—has continued the practice of relying on the pardons office.
"President Obama takes his constitutional power to grant clemency very seriously," said Matt Lehrich, a White House spokesman. "Race has no place in the evaluation of clemency evaluations, and the White House does not consider or even receive information on the race of applicants."
The president's power to pardon is enshrined in the Constitution. It is an act of forgiveness for a federal crime. It does not wipe away the conviction, but it does restore a person's full rights to vote, possess firearms and serve on federal juries. It allows individuals to obtain licensing and business permits and removes barriers to certain career opportunities and adoptions.
To assess how the pardons office selects candidates for pardons, ProPublica interviewed key officials, obtained access to thousands of pages of internal documents and used statistical tests to measure the effects of race and other factors on the outcome.
From 2001 to 2008, Bush issued decisions in 1,918 pardon cases sent to him by the Justice Department, most involving nonviolent drug or financial crimes. He pardoned 189 people—all but 13 of whom were white. Seven pardons went to blacks, four to Hispanics, one to an Asian and one to a Native American.
Fred Fielding, who served as Bush's White House counsel, said the racial disparity "is very troubling to me and will be to [Bush], because we had no idea of the race of any applicant."
"The names were colorblind to us," Fielding said, "and we assumed they would be at all levels of clemency review."
Beginning in September 2010, the Justice Department was required to make available the names of people denied pardons. Bush's pardon decisions were selected at random by ProPublica to examine the impact of the pardons office's recommendations over a president's full term and to test how well the office met the president's goal of assuring fairness in the process.
The department does not reveal race or any additional information that would identify an applicant, citing privacy grounds. To analyze pardons, ProPublica selected a random sample of nearly 500 cases decided by Bush and spent a year tracking down the age, gender, race, crime, sentence and marital status of applicants from public records and interviews.
In multiple cases, white and black pardon applicants who committed similar offenses and had comparable post-conviction records experienced opposite outcomes.
Turning over pardons to career officials has not removed money and politics from the process, the analysis found. Justice Department documents show that nearly 200 members of Congress from both parties contacted the pardons office regarding pending cases. In multiple instances, felons and their families made campaign contributions to the lawmakers supporting their pleas. Applicants with congressional support were three times as likely to be pardoned, the statistical analysis shows.
In reviewing applicants, pardon lawyers rely on their discretion in ways that favor people who are married and who have never divorced, declared bankruptcy or taken on large amounts of debt. The intent, officials say, is to reward people who demonstrated "stability" after their convictions. But the effect has been to exclude large segments of society.
The data show that applicants whose offense was older than 20 years had the best odds of a pardon. Married people, those who received probation rather than prison time, and financially stable applicants also fared better. When the effects of those factors and others were controlled using statistical methods, however, race emerged as one of the strongest predictors of a pardon.
The most striking disparity involved African Americans, who make up 38 percent of the federal prison population and have historically suffered from greater financial and marital instability. Of the nearly 500 cases in the sample, 12 percent of whites were pardoned, as were 10 percent of Hispanics.
None of the 62 African Americans in the random sample received a pardon. To assess the chances of black applicants, ProPublica used the sample to extrapolate the total number of black applicants and compare it with the seven blacks whom Bush pardoned. Allowing for a margin of error, this yielded a pardon rate of between 2 percent and 4 percent.
Adams, the head of the pardons office under Bush, said applicants were not penalized based on race. In fact, Adams went out of his way, he said, to help black applicants.
"People in general more and more feel that it is appropriate to give extra consideration to a member of a minority group," he said.
Applicants are not asked about their race. But race is listed in many of the law enforcement documents collected for the application, including pre-sentence reports, rap sheets and Federal Bureau of Prisons records.
Under Justice Department regulations, Adams said, lawyers in the pardons office conduct a rigorous review of an applicant's offense. They then examine character, reputation and post-conviction behavior—tests of what Adams termed "attitude."
"Is the person seeking a pardon for forgiveness or vindication?" Adams said. "Are they going to wave a flag around that says a pardon proves they didn't do as bad as the government said?" If so, he said, "it is counted against them."
Samuel Morison, a lawyer who worked in the pardons office for 13 years, said there is an institutional interest in preserving the convictions secured by the government's prosecutors.
"The pardon office is not a neutral arbiter, because the Justice Department was a party to every criminal case it examines," Morison said.
The yardsticks used by the office under Adams continue to be used under his successor, Ronald L. Rodgers, a former federal prosecutor and military judge.
Theodore B. Olson, a former solicitor general who has represented high-profile pardon applicants, said he has long been frustrated by the slow pace of the process and its lack of transparency. The Justice Department says the office has increased its efficiency, deciding cases in a little more than two years, an improvement since 2005, when the wait was twice that.
When a pardon is denied, the notice comes with no explanation.
"It just comes out of the blue," Olson said. "You can't explain to your client why, especially when you think you've made a strong case."
Parallel Cases, Disparate Outcomes
Denise Armstead's beauty salon sits on a busy corner in Little Rock's west side. A big sign out front beckons customers from the largely African American neighborhood.
Armstead, who is black, became a hair stylist straight out of high school and dreamed of owning her own salon. Like many small-business owners, she kept her own receipts. An accountant filled out her tax forms.
In 1994, the federal government accused Armstead, then 35, of failing to report $32,000 in income over four years. She hired a lawyer and fought the charges, ultimately getting them reduced to a single count of under-reporting her income in 1989.
Her lawyer, a former Internal Revenue Service employee, advised that a trial would cost more than the $3,000 fine, she said. In a plea bargain, she received three years' probation and paid the fine in installments.
In the same city, Margaret Leggett and her husband, who are white, were also accused of violating federal tax laws. In 1981, Leggett rented an apartment under a fictitious name and her husband created a fake bank account and fake Social Security numbers. They then filed for multiple tax refunds totaling more than $25,000.
Leggett pleaded guilty to conspiracy to defraud the government by making false claims. In her mid-30s, she was sentenced to three years in prison but was released after three months. Her husband paid a $5,000 fine and served 15 months in prison.
Years later, Armstead and Leggett each applied for a pardon. On paper, both were strong candidates. They had accepted responsibility in court and completed their sentences with good behavior.
Neither had any other criminal convictions. Both were active in their churches. Leggett and Armstead had both filled out lengthy applications in which they listed their crime, punishment and professional and personal history.
In April 2006, Bush followed the pardon attorney's recommendation and approved a pardon for Leggett. A year later, Bush again followed the attorney's advice and turned down Armstead.
Armstead had a personal reason for seeking a pardon: She had hoped to become a nurse. She was inspired to change professions while caring for her mother, who was dying of renal failure.
"I would take off work and take her to the clinic," she said.
An Arkansas nursing license requires a criminal-background check. Her felony record stood as a potential obstacle, her attorney told her. He recommended she apply for a presidential pardon. She was not aware that her 2002 request had been denied until a reporter informed her this year.
According to Justice Department memos, Armstead was denied "for a four-year course of criminal conduct for which [she] failed to take responsibility." The four years referred to the four charges of tax evasion in the original indictment against her.
Adams said that he did not remember Armstead's case but that, in general, applicants need to show remorse for any conduct they were indicted for, not just the charges to which they pleaded guilty.
"What the person did, as opposed to what they pled guilty to, is a relevant factor in judging how honest they are," Adams said. "This spills over to attitude."
A former White House lawyer said he had no idea the pardons office was considering indictments rather than only convictions in their deliberations.
"I definitely didn't know that," said Kenneth Lee, the associate White House counsel during Bush's second term who dealt with the pardons office. "If we knew these kinds of things, our decision making may have been different."
Leggett lives with her husband in Hot Springs, Ark., where they own a boat repair shop. She said she did not remember why she sought the pardon.
Kenneth Stoll prosecuted both Armstead and Leggett when he was an assistant U.S. attorney in Little Rock. Stoll said he does not recall either woman. The pardons office sought a recommendation for Leggett from the prosecutors' office after Stoll had retired. He was not asked his opinion. But, he says now, Leggett's crime was a more significant offense.
Leggett and her husband have been married for more than 30 years. They have owned or operated nearly a dozen businesses.
Though she was divorced when she applied for a pardon, Armstead would still appear to meet the "stability" test. She said her life has remained on an even keel—she continues to operate her beauty salon and does not have excessive debts.
For applicants who appear to be solid candidates, the pardons office considers the views of prosecutors and judges. Armstead's case never reached that stage.
But the pardons office solicited advice on Leggett—and received lukewarm answers.
The U.S. attorney's office in Little Rock took no position, and the judge did not object to a pardon. But Leggett's bid was opposed by a high-ranking Justice Department official. Eileen J. O'Connor, the assistant attorney general in charge of all criminal tax matters, advised the pardons office that Leggett's application should be denied because she did not "fully admit unconditional responsibility" for her crime.
O'Connor noted that Leggett omitted from her pardon application that she had rented the apartment under a false name and made a utility deposit as part of the ruse to file false returns. "It appears that she has attempted in her quest for a pardon to minimize her involvement in the crime and shift the blame to her husband," O'Connor wrote.
This could have posed a serious problem for Leggett. The Justice Department explicitly states that applicants need to take personal responsibility for their crimes and show remorse. But in this instance, pardons office lawyers appear to have accepted what Leggett's husband said at trial, which is that he had talked his wife into participating in the scheme.
Another factor in pardon applications is whether the person has a practical need for presidential mercy.
"If a person doesn't have a reason, it doesn't hurt," Adams said of the policy. "And if a person wants forgiveness, that is fine, but if the need is for licensing or business ownership or to obtain a franchise, that will help, and the office will try to push it farther."
Excerpted from Presidential Pardons by Dafna Linzer, Jennifer LaFleur. Copyright © 2012 ProPublica Inc.. Excerpted by permission of ProPublica.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.