Journalism in the Public Interest

Details Emerge on Government Study of Presidential Pardons

A review triggered by stories published by ProPublica and The Washington Post will test the effect of race on pardon decisions.


(David Goldman/AP Photo)

The U.S. Department of Justice has issued a request for proposals for its first-ever in-depth study of presidential pardons, providing fresh details on what it envisions the review will entail.

The agency said it would undertake the study in response to articles published by ProPublica and The Washington Post in December that found white applicants were nearly four times as likely to be pardoned as minorities. African American applicants fared the worst: Just 7 of 189 people pardoned by President George W. Bush were African American. So far, President Obama has pardoned 22 people, including 2 minorities.

ProPublica's analysis showed that other factors also appeared to influence petitioners' success rate. Married applicants were twice as likely to succeed and those with congressional support were three times as likely to receive a pardon.

In June, the Justice Department published an outline of the survey for contractors bidding to conduct it. The research will be expected to "test the primary hypothesis that, all other things being equal, African Americans and other minorities are less likely to progress in the pardon adjudication process than applicants of other races," the outline said.

The government's survey will focus exclusively on pardons, which convey forgiveness for federal crimes, and will not include commutations, which shorten federal prison sentences. Though presidents make the final decisions on petitions for clemency, they rely heavily upon recommendations from the Office of the Pardon Attorney, an arm of the Justice Department, to guide their determinations.

According to the study outline, the Justice Department review will be similar in many regards to the analysis ProPublica conducted for its report.

We examined pardon decisions made between 2001 and 2008. The department's study will extend from 2001 through April 2012. The government's proposal also calls for using a near-identical list of demographic variables — including age, race, ethnicity and gender — to probe what most affects the likelihood of a pardon.

One key difference between the reviews is that the government review will consider the influence of subjective categories, such as an applicant's level of remorse and acceptance of responsibility, on the success of pardon applications, the survey outlines said. It is unclear how researchers will quantify such variables.

The government study also will be overseen by a steering committee that will include staff from the pardons office, giving insiders a hand in the process.

According to the survey outline, the year-long study is expected to launch on Oct. 1, with a budget of $350,000. It is unclear whether the results will be made public, although the department has asked potential bidders to provide a "publishable quality final report."

ProPublica's methodology and findings can be found here.

Jennifer LaFleur, ProPublica's Director of Computer-Assisted Reporting, contributed to this post.

When investigating aspects as complex and deep as these, religion should also be a part of the matrix.

Stephanie Palmer

Aug. 8, 2012, 4:25 p.m.

What I remember is that George Bush the father pardoned everyone involved with Iran Contra before they had ever been charged. I remember that well.  Some of them even came back into government under George the son…........quite a gift, don’t you think?

Well, Stephanie, keep in mind what Poppy’s pre-White House job was and which organization ran Iran-Contra.  (And don’t forget which airport said organization flew its drug runs to…)

On the more general topic, it always worries me when “the government investigates.”  It seems that they always find exactly nothing, trying to apply mass analysis on a few dozen data points as if that’s sound.  The exception appears to be the GAO, but their analyses seem to find their way directly to the circular file.

Let’s hope they get this right.  To Rick’s point, probably the best way to handle this would be to take everything the census bureau asks (a known standard in demographic analysis that’s survived for at least a century, even if not necessarily the best one) and everything the OPA asks, and map the pardon requests in all those dimensions, reducing to find clusters.

For those people whose eyes glaze over when math is even referred to, what I’m saying is that they shouldn’t ask IF there’s a bias, but rather what biases there are.  If they can’t find any beyond what makes sense (the crime and the proof against the convicted petitioner), fine.  But obviously, they won’t, if they play fair.

clarence swinney

Aug. 22, 2012, 7:23 a.m.

He would have continued Clinton/Newt Fiscal Policies of Pay Your Way
We would continue to have Surpluses not Debt as far as the eye can see
Debt today would be 6000B not 12000B
2010 Budget would be 3000B not 3800
Unemployment would be 5% not the real 17%.
Food Stamp Roll would be 10% not 40%
Unemployment Insurance Roll would be low
Redistribution of Wealth back to Middle Class would have continued
The top 1% would own 20% not 43% of Total Financial Wealth
He would have continued Taxing the top 1%
Wall Street Regulators would be under experts not Partners In Crime
We would have 400 military bases not 800 worldwide
We would not have two wars
We would not have lost thousands of lives of young youths
Hundreds of thousands of Muslims would still be alive
Thousands of young Muslims would still have legs feet and minds
America would still be the most admired nation on Earth not the most despised
We would go back to behaving as a Christ-Like nation via our national policies
Millions would own homes not foreclosure notices
States would not be laying off hundreds of thousands in such important jobs as
Teaching and Policing California would have a balanced budget
John Kerry would be President.
Sound common sense policies would continue.

This article is part of an ongoing investigation:
Presidential Pardons

Presidential Pardons: Shades of Mercy

White criminals seeking presidential pardons are nearly four times as likely to succeed as people of color, a ProPublica examination has found.

Get Updates

Our Hottest Stories