Last week, ProPublica launched a new major investigation that found white criminals are four times as likely as minorities to receive a presidential pardon, even when factors such as type of crime and sentence are considered.

ProPublica's director of computer-assisted reporting, Jennifer LaFleur, joins the podcast to explain the mysterious process of awarding pardons, the data and how it was gathered for this investigation, and what this correlation between race and presidential pardons ultimately means.

"We found this racial disparity that jumped out of the data. We don't know exactly why that happens, but there's an opportunity here to re-examine the pardons process, and for the president to do something that might make the process more fair," LaFleur said. "Even if there are subjective things going on as pardons are decided, they really shouldn't fall along racial lines. So hopefully, this is just a start to provide a little bit of insight into a problem that can be fixed."

Read the full transcript below. You can also subscribe to all of ProPublica's podcasts on iTunes.

Transcript

Mike Webb: Hi, I'm Mike Webb and welcome to the ProPublica Podcast. This past weekend, ProPublica launched our presidential pardons investigation. The pieces showed how whites were four times as likely to get a presidential pardon than most minorities. And we also explored how members of Congress can exert influence in the process. The investigation was produced by Dafna Linzer, Jennifer LaFleur, Krista Kjellman Schmidt, and a host of researchers and interns. Joining us in the storage closet studio is Jennifer LaFleur. Jen is our director of computer-assisted reporting and she did the statistical analysis of pardons during the Bush presidency. Welcome to the podcast, Jen.

Jennifer LaFleur: Thanks for having me, Mike.

Mike: OK. How does a person go about getting a presidential pardon, and who is eligible and what are the steps that make it happen?

Jennifer: A pardon is the one unfettered power that the president has. They don't have to get a review or help with that decision, a president can pardon anyone they want. But the process has been for many years, since in 1893 Grover Cleveland issued an executive order that basically put the paperwork processing for pardons into the Justice Department. That office, the Office of the Pardon Attorney, processes pardon requests. So someone who would like to apply for a pardon, there is a very lengthy application on the Pardons office website they need to fill out and various pieces of information that have to be provided. That's the first step.

Mike: OK. Does the Justice Department's Office of Pardons have specific criteria or guidelines that they follow?

Jennifer: Some of it is subjective. On their website the office lists a five point test for applicants. The first test is straightforward they must have completed their sentence five years previous before applying. Then lawyers look at other things such as their conduct and character. Whether they're doing community service, whether they've committed other crimes since the crime they are seeking a pardon for. Whether or not they have a stable family life seems to be important. Some of those factors are objective and we can actually gather data on those. Things like financial stability. So we looked at things like bankruptcies and liens, marital status. Other objective data we could find were things like whether they had committed other crimes, how long ago their crime was, what it was, what their sentence was. There are other factors the office says are more subjective. Whether or not the person has really atoned for their crime and really given back to their community, those are a little harder to gather data on.

Mike: Things that are really up to the Justice Department lawyers that make the decision.

Jennifer: Yes.

Mike: OK. How did you start your analysis and where did you get the data?

Jennifer: We started with a Freedom of Information Act request to the Department of Justice for the list of people who were denied pardons during the eight years of the Bush administration. We also used the list that's already on their website of everyone who has been granted a pardon during that same time period. From that combined list we pulled a random sample of 500 applicants so that we could do further research on those individuals. Keep in mind, we had to pare down the list somewhat so it was still sound statistically, but so that it wouldn't take us a decade to do all of the research on all of those people. So I really want to point out the amazing amount of work everyone who did research on this did. Doing background research on hundreds of people is a monumental task, which was a giant mountain when we began this process. But the folks who worked on that, coordinated by our research director, Liz Day, was just a phenomenal effort: digging through public records, combing through court records, contacting people directly. I think every ProPublica intern that has been here this year has had an opportunity to participate in this process and I'm just really grateful to all the great work they did.

Mike: And all of that was really just to make sure that what we had was correct and to hear people's stories?

Jennifer: Well, it was to find all those pieces of information in our database. We started with a name. They found offense information, sentencing information, the court cases, whether people were married, where they lived so that we could contact them, whether they had bankruptcies or other crimes – all of that was something that didn't exist until we did all this research to actually build the database.

Mike: Now is that something that the Justice Department had? They assign a lawyer to investigate each case, right?

Jennifer: Yes.

Mike: So, they have all of this, but it's not necessarily included in the information that you obtained.

Jennifer: We obtained a list of names. This is the first time that that had ever been provided by piggybacking onto another lawsuit under which the names alone were released.

Mike: All right. How did you go about breaking the data down?

Jennifer: This is a sort of analysis I've done on similar projects. I've done projects looking at jury selection. This is a technique that's often used in stories on traffic stops. It's a statistical analysis where you look at what factors tend to correlate strongly with whatever your outcome is. In this case our outcome was whether someone was granted a pardon or not. I'm not going to get into the statistical nitty gritty, but basically, as a researcher my goal should be to figure out what are the factors that most strongly correlate with someone getting a pardon. So I plugged all of those pieces of data we got in, looked at various permutations of those data. When something jumps out of your analysis, particularly something like race, the first thing I need to do is try to figure out is there something else that will explain away race.

Mike: Is that what happened to you when you were looking through the data, race stood out as a factor?

Jennifer: Yes. Race did stand out as a factor. But especially when you talk to experts and folks familiar with the topic, they suggested that perhaps there were other things that were accounting for that. Perhaps all of the minorities had done much worse crimes or had longer sentences or something else about them made it so it would be natural that all of them would have a lesser chance of getting a pardon. We could not make that race factor go away when we added all these variables that we had. In our original sample, looking at raw numbers, 12 percent of the whites got pardons. None of the blacks in our sample actually got a pardon. We extrapolated that to the population of a whole of everyone applying for pardons and we estimate that somewhere between two and four percent of blacks probably got a pardon from the population, the larger population of applicants. But that's still much lower than the rate for whites. Even after we controlled for the other factors we still saw that disparity between whites and minorities.

Mike: What did you notice about the people who were denied pardons?

Jennifer: The numbers are one thing, and we have online data and all the details of our analysis online, but what really brings this forward is when you start reading some of the stories of the individual people who sought pardons, who had gone back to college, really gotten their lives in order and still were denied pardons. And some of the people who maybe didn't have such remarkable lives who did get pardons with the difference in many of those cases simply being race. So that, I think, really draws a human face on some of these numbers.

Mike: Did the Justice Department respond to your analysis?

Jennifer: We went to the Justice Department early on, when we were still working on our analysis, and told them all of the variables that we looked at and what our preliminary findings were. At that time and up until when we published the stories they did not provide other variables they thought that we were missing. In fact, they said that they didn't think there were variables that were missing in our analysis but that there was subjectivity in the process that might not be accountable by data.

Mike: If President Obama wanted to take action to fix this, is there something that he can do?

Jennifer: Well this is his one power that he can do anything he wants with. He absolutely could. He doesn't have to go through the Justice Department to process pardons. He could bring that process back to the White House.

Mike: An executive order…?

Jennifer: I'm not even sure that he would need an executive order to do that. I think it's pretty much a guaranteed power that's his thing that he can do.

Mike: Jen, why is this a ProPublica investigation?

Jennifer: Well, here at ProPublica we do stories where we can actually make a difference. Where reforms can be made and we can right wrongs. We found this racial disparity that jumped out of the data. We don't know exactly why that happens, but there's an opportunity here to reexamine the pardons process, and for the President to do something that might make the process more fair. Even if there are subjective things going on as pardons are decided, they really shouldn't fall along racial lines. So hopefully, this is just a start to provide a little bit of insight into a problem that can be fixed.

Mike: Ultimately it's an accountability problem, it's an accountability story.

Jennifer: Yes.

Mike: OK. Well, Jen, fantastic job on this. It's a big project and I know it took a lot of work. Thanks for joining us to talk about it.

Jennifer: Thanks for having me.

Mike: Be sure to check out all of the other elements of the investigation, like the editor's note, a sidebar on Marc Rich's continuing influence, the Document Dive that has letters that politicians wrote to support pardon requests, there's the timeline of all pardons and there's a lot more. And you can see all of this at ProPublica.org/pardons. Now, for our Officials Say the Darndest Things Tumblr quote of the week, "I'd like to think of myself as the flavor of the decade." Who said it? Republican presidential candidate Ron Paul, explaining his recent jump in an Iowa poll.

OK. Thanks for joining us for another podcast. As always, thanks to Minhee Cho for producing. Thanks to you for listening. For ProPublica, I'm Mike Webb. We'll catch you next time.

Transcription by CastingWords