Commission Meeting of November 20, 2008 - on Employment Discrimination Faced by Individuals with Arrest and Conviction Records

 

PRESENT:

NAOMI C. EARP Chair
STUART J. ISHIMARU Commissioner
CHRISTINE M. GRIFFIN Commissioner
CONSTANCE S. BARKER Commissioner

ALSO PRESENT:

RONALD COOPER General Counsel
REED RUSSELL Legal Counsel
BERNADETTE B. WILSON Program Analyst

This transcript was produced from a video tape provided by the Equal Employment Opportunity Commission.

TABLE OF CONTENTS

Announcement of Notation Votes

Motion to Close a Portion of Next
Commission Meeting

• Panel 1: Barriers Experienced by People with Criminal Records
  • Devah Pager
  • Diane Williams
• Panel 2: Stakeholder Perspectives and Litigation Issues [Video failure caused incomplete transcript]
  • Michael Foreman
  • Donald Livingston
  • Janet Ginzberg
• Panel 3: New Research Developments [Video failure caused incomplete transcript]
  • Shawn Bushway
• Panel 4: Employer Practices
  • Rae Vann
  • Laura Moskowitz

Adjourn

(9:38 a.m.)

P R O C E E D I N G S

CHAIR EARP: Good morning everyone. The meeting will now come to order. Thank you all for being here today. In accordance with the Sunshine Act, today's meeting is open to public observation of the Commission's deliberations and there's no voting today. Oh, there is voting, deliberations and voting. At this time, I'm going to ask Bernadette Wilson to announce any notation votes that have taken place since the last Commission meeting, Ms. Wilson?

MS. WILSON: Good morning, Madam Chair, Commissioners. I'm Bernadette Wilson from the Executive Secretariat. We'd like to remind our audience that questions and comments from the audience are not permitted during the meeting and we ask that you carry on any conversations outside the meeting room, departing and re-entering as quietly as possible. Also, please take this opportunity to turn your cell phones off or to vibrate mode. I would also like to remind the audience that in addition to the elevators, in case of emergency, there are stairways down the halls to the right and left as you exit this room. Additionally, the restrooms are down the hall to the right.

During the period October 23rd, 2008 through November 19th, 2008, the Commission acted on one item by notation vote:

Approved the obligation of funds for a non-competitive labor economist contract.

Madam Chair, it is appropriate at this time to have a motion to close a portion of the next Commission meeting in case there are any closed meeting agenda items.

CHAIR EARP: Thank you, Ms. Wilson. Do I hear a motion?

COMMISSIONER ISHIMARU: So moved.

CHAIR EARP: Is there a second?

COMMISSIONER BARKER: Second.

CHAIR EARP: Any discussion? Hearing none, all those in favor, please say Aye.

ALL: Aye

CHAIR EARP: Opposed?

(No response.)

CHAIR EARP: The ayes have it and the motion is carried. Thank you again, Ms. Wilson.

I want to extend a warm welcome to everyone in attendance today. I anticipate that this will be a very informative meeting. One housekeeping note before we begin. To afford each panelist their fully allotted time to speak and to allow the Commissioners time to ask questions, I've asked the Legal Counsel to make use of the timing lights. The light will turn yellow, giving each panelist and the Commissioner a one-minute warning. When the light turns red, it is the traditional meaning, it means stop, your time has expired. You may finish the thought but please respect the time limits so that later panelists are not rushed.

Two years ago, Commissioner Ishimaru worked with me to roll out the E-RACE Initiative, Eradicating Racism and Colorism from Employment. E-RACE is basically a 21st Century framework for looking at some old and persistent problems of race and color. We wanted to especially look at those things that may constitute proxies for race, color or ethnicity. Today's Commission meeting on employment discrimination against individuals with arrest and conviction records is an issue that has long been with us but which in recent years has re-emerged as an important civil rights issue.

Of course, the concern about arrest and convictions is also a business issue, a security issue, a safety issue and a tort liability issue. This is also an area where facts and reason can easily be overwhelmed by fears, stereotypes, and myths. So the need to balance so many competing interests including whether or not criminal records are a proxy for race discrimination means that we all have our work cut out for us.

With the help of EEOC's Office of Legal Counsel, Reed Russell and the very, very thoughtful OLC Staff, we have been actively reviewing our existing enforcement guidance on arrest and conviction records. We are working desperately and trying hard to think through these issues in order to provide updated guidance. We need to get guidance to the staff as well as to our stakeholders.

I'm sure that this meeting today and the information that we learn will help inform the process we are engaged in. We have several very distinguished panelists that will share their expertise with us and Commissioner Ishimaru will introduce the first panel in just a minute.

I'm sure most of you know that EEOC is in the process of moving and today is effectively the last Commission meeting that we will have at 1801 L Street. So for some of us, it's a little bit nostalgic. But I want to thank Commissioner Ishimaru for suggesting this meeting and I especially want to thank Jacinta Ma of Commissioner Ishimaru's staff and Corbett Anderson in the Office of the Chair for working so hard to pull this meeting together. It seems fitting to me that the last Commission meeting at 1801 L Street be devoted to a topic as important and as timely as this. Commissioner?

COMMISSIONER ISHIMARU: Madam Chair, thank you and thank you for holding this meeting, for letting my office work on helping you pull this together. We really appreciate the opportunity. I, too, want to thank the staff because that's where the work gets done. Jacinta Ma and Corbett Anderson have done a fabulous job in pulling this together.

And we come to this in the context of the Chair's E-RACE Initiative and we have talked about trying to address a number of issues that are new issues and issues for this new century and I know with all the responsibilities that the Chair has and has to juggle, I was delighted that we were able to find time today to do this, and I'm very thankful to you, Madam Chair, for allowing this to happen.

And as we'll hear during the course of the day, many, many people have interaction with the criminal justice system and that number is growing and it's grown in recent years by substantial numbers. And the question for all of us is, how do we deal with that issue and its aftermath? And this is not an issue that comes on one side of the political aisle or the other. These are difficult questions of how do you get people back into society when they've been excluded from it for a variety of reasons.

And you know, I wanted to note the bipartisan nature of this issue and I know that this year President Bush signed into law the Second Chance Act, that he supported and the Second Chance Act will assist prisoners' transition back into society in hopes of reducing recidivism. It authorizes funding for states and non-profit organizations to provide job training and placement services, housing, substance abuse treatment, mental health assistance, and other services to help ex-offenders re-enter our communities.

And in signing the Second Chance Act, President Bush stated, "This country was built on the belief that each human being has limitless potential and worth. Everybody matters. We believe that even those who have struggled with a dark past can find brighter days ahead. One way we can act on that belief is by helping former prisoners who have paid for their crimes. We help them build new lives as productive members of our society."

I fully agree with the words of the President. It's something that our country should do and should be proud of doing. And studies have shown that having a job helps keep people from becoming recidivist. As President Bush said, "A high recidivism rate places a huge financial burden on taxpayers. It deprives our labor force of productive workers and it deprives families of their daughters and sons, husbands and wives and moms and dads.

Fears, myths and such stereotypes and biases against those with criminal records continue to be part of the -- part of a decision making for many employers. Business and industry suffers as a result because it is not able to benefit fully from the skills of every potential worker. For our economy to be successful, we cannot afford to waste any available talent. And the EEOC has a long role in addressing this issue. One of the reasons why we're here is that as we'll hear during the course of the day, the Courts have questioned some of our guidance that was issued many years ago.

The first came during the tenure of Chairman Thomas. I don't know if you were here at the time when this was issued but there was 1987 guidance under Chairman Thomas during the Reagan Administration and there was other guidance issued in 1990 under Chairman Kemp, during the George H.W. Bush Administration. But, you know, much has changed. Some of that guidance has been out of date and this discrimination continues to arise in our work at the EEOC even today.

Just this past September, the Commission unanimously approved the filing of a case in the Western District of Michigan against People Mark, alleging that a class of African Americans were discriminated against due to its policy that denies the hiring or employment of any person with a criminal record.

We have been reviewing our guidance as the Chair noted and it would be useful to update it. Having updated guidance would be an important step in achieving the E-RACE goal and objective to develop strategies for addressing 21st Century manifestations of discrimination, such as discrimination based on arrest and convictions.

So I'm looking forward to today's discussions with our invited panelists and I know their remarks will help guide us as we review our guidance. It's my pleasure to introduce the first panel, including Dr. Devah Pager, Professor of Sociology from Princeton University, who will discuss her research which looked at the success of job applicants with criminal records compared to those without them, and Ms. Diane Williams, CEO and President of the Safer Foundation that works to help formerly incarcerated individuals to find jobs.

So, welcome, both of you. We're delighted to have you and Madam Chair; I turn the floor back to you.

CHAIR EARP: Why don't we start with Ms. Pager?

DR. PAGER: Great, thanks, Madam Chair and Commissioners. Thank you very much for inviting me to participate in this meeting today. The issue of the employment of individuals with criminal records has clearly become a pressing policy priority today. There are now nearly 700,000 individuals being released from jail and prison each year, roughly 12 million ex-felons in the United States.

Further, given high rates of racial disparities in the United States, I'm sorry, given high racial disparities in the rates of incarceration, any impact of a criminal record on employment will disproportionately affect the outcomes of African Americans. And on this point, I just want to show you some basic patterns. For those of you that can't see the PowerPoint, I will give you the main punch line. At any given time, young Black men are roughly seven to eight times more likely to be incarcerated than Whites of their age group, but when we look at the lifetime chances of incarceration, those numbers reach quite staggering proportions, so the lifetime chances, the chance that at some point during his lifetime a young Black man will spend time in prison is roughly one in three.

And when we look at young Black male high school dropouts, those figures rise to nearly 60 percent. So nearly 60 percent of young Black male high school drop-outs will expect to spend some time in prison during their lifetime. And I think these figures emphasize the degree to which the issue of employment for individuals with criminal records is also very much an issue of Black men's employment.

I want to spend my short time here today telling you about a series of field experiments that I've conducted, investigating the effects of race and criminal background on employment in low wage labor markets. Why did I conduct a field experiment? Well, we know that those individuals who wind up in prison or wind up with criminal records are not a random sample of the population and they may have characteristics that make them unappealing to employers irrespective of whether or not they have a criminal record. And so I wanted to be sure to construct a study in which I was comparing identically qualified applicants who differed only in whether or not they reported a criminal record. So in this field experiment, I hired a group of young men to pose as job applicants. I carefully selected and matched these young men so that they had very similar kinds of physical characteristics, interpersonal styles of self-presentation. I assigned them fictitious matched resumes that reflected identical levels of education and work experience. So these applicant pairs presented identical candidates except for the fact that one member of each team presented evidence of having a criminal record, a drug felony conviction.

I sent these young men all over the city applying for hundreds of entry level jobs. These were jobs like restaurant workers, retail sales, couriers, customer service, laborers, warehouse workers, telemarketers, a wide range of entry level positions.

Two applicants went to each employer. This is a basic design where I have two White applicants, two Black applicants, one with a criminal record, one without. Two applicants went to each employer, one with criminal record, one without and I wanted to see how race and a criminal background affected their prospects of finding work.

So what did I find? Well, there are three main results that come out of the study. I'll show you the Milwaukee results and then I'll show you the almost identical results from New York City.

The first effect is that the impact of having a criminal record was large. So here when we look at White applicants, we see that even for these otherwise very well qualified young men, having a criminal record reduced their employment opportunities by roughly 50 percent. So an individual with a criminal record was about half as likely to receive a call-back or a job offer relative to an equally qualified non-offender.

Second, we see evidence that the effect of a criminal record was even larger for Blacks, so there's about a two to one ratio for Whites, whereas for Blacks, the 14 versus 5 percent is a nearly three to one ratio. So in Milwaukee, this effect was larger. In New York we see the disparities between Black offenders and non-offenders even larger than we do in Milwaukee. So Blacks pay, in addition to Blacks being more likely to have a criminal record, they also pay a larger penalty for that criminal record when seeking employment.

And then finally, the results of the study point to the persistence of direct racial discrimination in employment. So I find that Black men with clean backgrounds, that 14 percent, actually faired no better in their search for work than a White man just out of prison, which is the 17 percent. So essentially, these results suggest that being Black in America today is essentially equivalent to having a felony conviction at least in one's chances of finding low wage work.

This is a startling result and I think it points to some of the ways in which high incarceration rates among African Americans may have strengthened the association between race and crime in the minds of many Americans and many employers. So in some ways all Black men pay a penalty for their membership in a group with such high levels of incarceration.

Okay, so those are the main conclusions from my research and in the remaining time I'd like to flesh out some of the details of this study. One of the things that came out, especially in the New York study where we observe a lot of the interaction between applicants and employers, is the degree to which rapport building really matters in finding these jobs. So in these low wage jobs, employers are not looking for any specialized skills or much experience and they're not willing to spend a lot of time evaluating or screening applicants for these positions. So for the most part, employers, seem to be basing their hiring decisions on very short interactions with job applicants and what they refer to as a gut feeling about the candidate.

As it turns out, the race and criminal background of an applicant seem to have a really significant effect on opportunities for this rapport building. Both these characteristics seem to make employers wary and uncomfortable and often make it difficult for the employer to see past these characteristics in their review of the applicant.

For example, African American applicants were significantly less likely to be granted the opportunity to interview. They were roughly 30 percent less likely to be offered an interview relative to their White counterparts. This means that Blacks have fewer opportunities to establish a personal rapport with the employer and for Black ex-offenders, fewer opportunities to explain the context of their conviction or to present evidence of rehabilitation.

Likewise, when the topic of a conviction did come up during an interaction, employers were often very uncomfortable with the issue, didn't know how to ask about it or express their concerns. In many cases they simply preferred to avoid the issue all together or to terminate the interview quickly in order to avoid the uncomfortable topic.

I want to show you just a couple of excerpts from our interactions between our testers and employers. So in one case, Kevin, one of our White testers, reports his interaction with an employer for a sales job. Here Kevin says, "I noticed the sign on the door which read, help wanted, part time, some experience needed. The employer told me he had great part time positions."

So the two of them talked for a while as the employer looked over Kevin's application. "The employer pointed at my response for reasons for leaving past jobs on the application right next to the correctional facility and said, `Why did you write parole?' I said that I was currently on parole. The employer then looked me in the eye and said, `Did you commit a crime'? I said yes. He then looked down at the sheet and said he really wasn't hiring right now".

In another case, Tony one of our African American testers, reports his experience applying for an administrative position. Here Anthony reports, "She told me to hold on a second while she went to get someone to interview me. As she walked away, she looked over my resume. She must have walked no more than 20 feet before she stopped, turned around and said, `Thank you, I'll pass this along'. I can only assume that it was a result of my conviction."

So we see these interactions sort of abruptly terminated when the employer sees evidence of the criminal conviction and it's difficult for the applicant to get past this in presenting his otherwise very appealing qualifications. And as you can see, these experiences were reported by both the Black and the White applicants and both groups reported seeing the issue of the criminal record break the flow of conversation with an employer or starkly affect the rapport. But beyond this, we actually see evidence that the consequences of an employer's reluctance to discuss the issue with applicants seemed more consequential for Black applicants. Employers seemed to have concerns about Black ex-offenders but if they weren't willing to talk about them in an interview or weren't willing to find out more about the applicant to help them contextualize the information, these Black applicants were extremely unlikely to get the job and that's part of where we see the huge disparities in the effect of a criminal record where it's larger for Blacks than for Whites.

This issue, I think, raises some important policy questions about how and when information about criminal records should be introduced to maximize the chances that individuals with criminal records would be given a fair shot at employment. If employers were more comfortable with the issue, if they knew what to ask about and what to look out for, it might be productive for the information about criminal records to be discussed in the initial interview so that employers can actually voice their concerns to applicants or have a chance to see how the candidate presents himself and explains his trajectory since the conviction.

On the other hand, I think it's more likely that for many employers, this information is simply too scary to ignore and it makes it hard for them to look beyond that information when evaluating the applicant. In this case, then, I wonder if it might be more desirable to encourage employers to wait until a later stage of the employment process before soliciting information about the applicant's criminal background. Wait until applicants have had a chance to build rapport, to show their personal qualities and qualifications and then to introduce the criminal record information and let employers decide whether or not it seems relevant for their evaluation.

This approach has been adopted now in several city governments, in Oakland, San Francisco, Boston, I think, Chicago, following what I refer to as these "ban the box" initiatives, where city employers are now allowed to ask about criminal records only after they have selected a potential hire. So this allows applicants to go through the hiring process without the shroud of the stigma hovering overhead and allows the information about the criminal record to be introduced in the context of a more personalized review so that it's not the only characteristic that employers are thinking about as they interview and evaluate the applicant.

I know I'm out of time, so I'll stop there and I'll look forward to your questions and comments, thank you.

CHAIR EARP: Thank you. Ms. Williams?

MS. WILLIAMS: Good morning, and thank you for the opportunity to participate in this discussion. I smile because as anybody who knows me would expect, I have prepared far more information than I can share with you today in a matter of 10 minutes. And so I'm going to just kind of put this aside for just a second, if I can and talk a little bit about the organization that I work for and to talk a little bit about the experience of the people that we serve and maybe make a few recommendations at the end of that.

COMMISSIONER ISHIMARU: Madam Chair, all this information will be put in the record. The written statements will be put in the record and put on the website so we don't lose all the wonderful information that's been prepared for us.

CHAIR EARP: Yes.

MS. WILLIAMS: Thank you very much. Safer Foundation is a not for profit organization headquartered in Chicago, Illinois but operating throughout the State of Illinois and part of Iowa. And we've been working exclusively with people with criminal records in their search for employment for 38 years. We have, if you will, started very small and now are placing last year as many as 2700 people in employment.

Unfortunately, that 2700 people does not represent half of the number of people who came to see us and asked us for help in finding employment. In fact, over 9,000 people came to us last year asking us to help them find employment. I would tell you that their plight as you might surmise from what Devah has presented, is a very difficult plight.

In many cases, they are people who don't have work histories or don't have significant work histories, are not well educated and so even without a criminal record, may not be appropriate for lots of opportunities that are out there. But even when they're only applying for minimum wage, low skilled to no skilled kinds of opportunities, they tend to get turned down far more often than they are accepted.

If I talk to the sector managers of the Safer Foundation who have responsibility for identifying employers and convincing those employers to hire people with criminal records, I find that in the best of cases, one out of five of those employers that they contact will hire people with criminal records. This is fairly consistent with the research that Harry Holzer out of Georgetown University has done, where he talks about fewer than 40 percent of those employers that he surveyed actually being willing to accept people with criminal records. What we find, of course, in today's environment where the economy is getting worse, is that the number of people who will hire people with criminal records goes down even further because they don't have to hire people with criminal records. They can hire anybody to do a minimum wage job even though those people may not stay and they'll have to start all over again at some point in time.

I will tell you, though, that we have better success with the small to medium business than we have with the larger business. By and large, small and medium businesses don't have very sophisticated human resources functions and we can serve a purpose for them that they don't have to fund, quite frankly, but gets them candidates that are pre-screened quickly, that will help them to work with those candidates once they've hired them and if it does not work out after all that can be done has been done, we will replace that candidate in a relatively short period of time. So because there is an intermediary in that process, those clients that we serve have a better chance at least of getting some opportunity for employment.

But again, those are minimum wage jobs, quite frankly, with limited futures and what we know is that we all have a desire and a will to take care of ourselves, to feed ourselves, to shelter ourselves, to clothe ourselves and also to do that for our families. You cannot do that with minimum wage for any prolonged period of time. So if there is no opportunity for movement up on the wage category or in the benefit category, then people fall prey to what they fell prey to in the first place, and that's crime.

We are working hard, obviously, to make those opportunities available. We have limited positive responses from large corporations that have those kinds of opportunities. And I won't name this particular cable company, but we actually have on our Employer Advisory Board a cable company representative, but that company will not hire our clients. So we keep that person onboard because we are always hopeful that we will find a way to convince them that they should, in fact, hire our clients.

There was, in fact, a smaller cable company that we had significant numbers of conversations with, found that they were struggling in the area of unemployment compensation costs. They were hiring lots of people, took them through a fairly extensive training program and the people did not make it all the way through. These were not our clients, these were just people that they hired off the street. So after months of conversation, we convinced them that they should allow us to do the training, that we would pay those folks a stipend as opposed to hiring them. They could interview them at the end of the process and only hire those that they chose to hire based on their performance in that program.

As we got close to those sort of closure on that deal, venture capitalists decided that it was time to take that company over. It disbanded the company and we did not have that opportunity. But my point being there are real advantages for working with organizations like ours and real advantages for considering our clients, because they will do that job and they will have more appreciation and more loyalty because they got that job than folks who don't have to struggle as hard for it.

So I would tell you that people who come home and knock on our door are not obligated to do that. They do it because they want to make a difference in their lives. They want to change their lives. They want to go to work, and they want to make a legal legitimate income to support themselves and their families. If we give them that opportunity as is seen in our recidivism rates, people do better and crime goes down. In fact, if we look at the Safer Foundation recidivism rates, and we do this every three years, we have Loyola University come in, take a look at everybody that has come through our programs and our recidivism rate for everybody that comes through is 28 percent.

The state's recidivism rate for the same period of time is 52 percent. If we are able to give them both supportive services and assist them on the employment side, that rate drops to 13 percent. So we know that if people have a real opportunity to go to work, they have -- we have an opportunity to have taxpaying, law abiding citizens rather than to have crime going up on a regular basis.

With all of that said and with my concern for time, I'm going to go to my recommendations. And there really are lots of recommendations that we could make but we'll focus, if you will, on just a few. Devah started to make one of those recommendations and that was the "ban the box" recommendation where we take that whole process of asking the question up front away. We don't take it out of the process all together, but just move it to the tail end.

We did get legislation passed in the House in Illinois. We did get legislation passed in the Senate in Illinois. We did not get the Governor to sign our legislation unfortunately last year. So the session coming up in the spring, we will be back at it, pushing that again, trying to make sure we understand what the issue was at the Governor's office. But I would tell you that the City of Chicago has adopted the "ban the box" concept in their hiring process and I was just yesterday with a representative from the City who indicated that he had looked at their most recent hiring numbers and 20 people, as it turned out with criminal records, had applied for jobs. All 20 were being hired for those jobs.

They didn't find out about their jobs or about their criminal records, I'm sorry, until the end of the process though. They looked at their qualifications. They looked at their willingness to work. They looked at all of those kinds of things that they would look at any candidate for and based on their responses to those, they were desirable candidates, and in fact, were being hired. To date, in fact, they quote hundreds of hires of people with criminal records based on having employed this process versus the process of having the box on the first page of the application. And so we have a small example of success when that sort of policy is in fact, employed.

I would also say that we need to look at prohibiting employers and other non-law enforcement agencies from inquiring about or using information about arrests that did not lead to conviction or missing dispositions on criminal record reports issued by the Federal Bureau of Investigation. What we know is that when people get that information as they are looking at their candidates, they don't always understand the nuances associated with it. And so I think it has to be incumbent upon those who do know the potential impact to make sure that we're getting the right information and it's going to be used properly.

We talk often about how this affects our individual clients. I would tell you that at one point at the county level in our city, even when a record was expunged, there was a statement left on that record that said, "Record expunged". So without having any information at all about that arrest, that conviction, or lack of conviction, employers made decisions not to hire people and so we know again that we have lots of records that need to be cleaned. We have lots of records that need to be treated properly in order for success to occur.

We would suggest that we enact a federal standard based on recommendations outlined in the Equal Employment Opportunity Commission guidance on the use of background checks for employment purposes when screening applicants with arrest and conviction records. I need to make a note that the recommendations that I'm making today some come exclusively from the Safer Foundation but many are recommendations that were formed in conjunction with other people working in this field and concerned about this sort of blockage, if you will, or barrier to employment for people with criminal records.

I've got a yellow light, so I want to ask one other thing that's actually not embedded in my recommendations but I think after having a brief conversation this morning over coffee, that I should put on the table. And that is to go back to what I talked about earlier and that's large employers that have many opportunities and opportunities, in fact, with career ladders embedded in them and if there is any way that those employers can be convened; if we can look at how the federal government vendors are, if you will, evaluated, and considered for contracts, there should be an aspect in that contract about agreement that asks them the question about hiring people with criminal records and their policies around doing that. Thank you very much.

CHAIR EARP: Thank you. Commissioner Ishimaru?

COMMISSIONER ISHIMARU: Thank you, Madam Chair. Thank you for both of those statements. I know we just sort of scratched the surface and within my five minutes I'll try to get to a number of issues. I think that's a fascinating possibility though of looking at big employers, looking at the federal government power when it contracts, as we've done in the race context for many years back.

You know, one of the sort of conundrums here, I think, is that when we look at a lot of the other work that we do dealing with big employers versus small employers, quite often I'm told that the big employers understand the other civil rights issues that we deal with and have infrastructures in place to deal with it. Here is sounds like it's flipped on its head and you know, it was fascinating reading Professor Pager's work of talking about the rapport that develops and how complicated that is and how hard that is to get to from a study standpoint, and it was fascinating reading that.

And the other thing from Professor Pager's work that sort of links with other issues we've talked about is the whole use of match-pair testing and how powerful that is as a starting point to begin to look at this from a very quantitative point of view and how powerful those numbers are. When you did those match-pair studies were the employers -- was the race of the employer factored in as well, because I noticed in some of the narrative that went along, there was this bonding over a shared ethnic heritage.

DR. PAGER: Uh-huh.

COMMISSIONER ISHIMARU: And I didn't know whether that -- you know, how that played into it.

DR. PAGER: Right, the vast majority of the employers were White and so in most cases, those are the results that I'm presenting. And we can't really make statistically reliable generalizations about the small sample of minority employers, but there is some evidence that they were more open both to African American candidates and ex-offenders. I can just think of one example where an employer was sitting down with our tester, noticed the parole officer listed on the resume and said, you know, "So you have a good relationship with your parole officer". He said, "Yes", and the employer said, "You know, mine was great, really helped me get out of a bind", and there was sort of this bonding around a shared experience that helped to create more empathy, but in terms of just race and employer effects, I think the numbers are just too small.

COMMISSIONER ISHIMARU: And I would also assume that given the changing demographics around the country where people from different groups are in various different positions, that it would be interesting to see what the studies would show. I can't even fathom to guess what it might look like because it gets fairly complicated fairly fast, but it would be fascinating work for other studies. Did you find a difference between your study in Milwaukee versus your study in New York or was it basically along the same lines?

DR. PAGER: So the core results were very similar. I only presented one set of teams from New York but we had a number of teams. Most of the results very much replicate Milwaukee. The effect of the criminal record overall was actually a little bit smaller in New York, so employers seemed a little bit more open and a little bit more willing to be sympathetic to the plight of a young man who wants to get back on track but the racial disparities and the effect were actually larger in New York than they were in Milwaukee.

COMMISSIONER ISHIMARU: Yes, and the other, given the timing, probably my last question of this round, is the use of intermediary organizations. I know when Commissioner Griffin has talked about the employment of people with disabilities, I remember at one hearing we did having an intermediary organization talking about how they pre-screen people as well and it really brought to a much higher degree of success.

I guess, you know, the trick for us as the federal enforcement agency, enforcing civil rights laws, is you know, obviously, you need intermediary organizations. Obviously, they help to do a better job in this area. But for us at the enforcement agency, you know, the larger question is what can the enforcement agency, what can the government be doing as an institution to try to get over some of these barriers, and I think some of the issues you raised at the end, some of the suggestions you raised are very intriguing.

MS. WILLIAMS: I would tell you, and again, I go back to the criminal records and the information that's shared, how that information is shared, you know, what they can do with the information, and just how accurate that information is, is something that could absolutely use your support. As we think about the kinds of records that we find in the FBI files, I smile because as a child of the `60s, I'm terrified to think of what pictures or anything else might be in there from my college, my undergraduate days in school and all those other kinds of things.

You know, as we walked across campus, really with nothing else on our minds but the next class, sometimes, you could find yourself amongst a crowd of people that were voicing very strong opinions about different kinds of things. And so there are all kinds of data, there's all kinds of information embedded in those files. I think we need to make sure we clean those files up.

I think there needs to be a conscious effort to do that and that we should not send that information out with anything that's inaccurate. And I think that the people who are discussed in that data should have access to that same information to have a clear sense of what's being presented.

And then the tail end of that, of course, is how that information then can be treated and I believe that employers should be held accountable for looking at the crime that was committed, the time that has lapsed between when that crime was committed and the application and what's happened in that person's life in between rather than being able to have a blanket policy that says, "We will not hire people with criminal records".

COMMISSIONER ISHIMARU: Could I follow up with one last question to Ms. Williams? Are there other entities like the Safer Foundation in other metropolitan areas and I guess the flip side of that is how much of the country doesn't have similar programs?

MS. WILLIAMS: I would tell you that there are definitely other programs that are similar to Safer. They may not mirror all the different kinds of things that we do and they may do things we don't do. What comes to mind is a CEO in New York, Pioneer Human Services, in Seattle, Washington. And so there are agencies like ours and we do work together to improve how we approach our work. But I would tell you there still is a lot of territory, if you will, that's uncovered.

COMMISSIONER ISHIMARU: Thank you. Thank you, Madam Chair.

CHAIR EARP: A related question; has your foundation or any other to your knowledge ever approached the Department of Labor, OFCCP or those organizations that have responsibility for federal contractors compliance issues?

MS. WILLIAMS: We work with the Department of Labor on a regular basis. We are part of the President's Re-entry Initiative. We have done a lot of different projects in conjunction with the Department of Labor. I will tell you, we probably have never had a meaningful conversation around the issue that you're putting on the table now.

CHAIR EARP: Okay, thank you. Commissioner Griffin?

COMMISSIONER GRIFFIN: I want to thank the Chair for holding this meeting. This is a topic that I think since you rolled out the E-RACE Initiative that we've been talking about, and so it's great that we get to actually have a meeting on this and I want to thank Commissioner Ishimaru and his staff, especially Jacinta and Steve Zanowic on my staff for putting these materials together and preparing us for this meeting.

It is unfortunate that it's our last meeting here in this building but I'm sure we'll have beautiful new digs in our next place. This is -- you know, the statistics are pretty startling. As you look at more than two-thirds of employers are using this information, are looking for this information to base decisions on and the number of people being incarcerated is increasing, it's certainly a huge issue. We should be very concerned about recidivism and as President Bush said, you know, we need to really use the talent pools that we have and we shouldn't be segregating these folks out.

******************************************** [Video failure caused transcript gap. Statements of other panelists are available at http://www.eeoc.gov/abouteeoc/meetings/11-20-08/index.html] ********************************************

COMMISSIONER GRIFFIN: ... on the cost issue, but why would you want to do all that? Why wouldn't you wait until you got to the point where you were serious about someone to actually spend that money? That's --

MR. LIVINGSTON: Right, I'm not disagreeing.

COMMISSIONER GRIFFIN: No, I just wanted to add that on so you can think about that.

MR. LIVINGSTON: I don't know what the individual cost from location to location it is. It's probably less than a sandwich, you know, you have at lunch, but I'm talking about you know, the circumstances that would confront employers that may have hundreds of thousands of applicants in any one year and the cost of evaluating applicants, you know, it can be significant.

There's one thing I would like to add to the discussion is, is that across the panels the terminology tends to differ from speaker to speaker and even by each speaker, point to point. Criminal histories generally includes criminal convictions, it would include arrest where there's been no criminal conviction and it might also include expunged or sealed records.

In my experience, the large employers do not rely upon arrests. They don't rely upon expunged records nor do they rely upon records that have been sealed. They rely upon -- they rely only upon criminal convictions.

COMMISSIONER ISHIMARU: So from your point of view, if we came out with a blanket statement that said employers shouldn't rely on arrest records in making an evaluation, that would not be troubling?

MR. LIVINGSTON: Well, I'm not an employer and I'm not a politician or a social scientist, I'm just a lawyer.

COMMISSIONER ISHIMARU: Just a lawyer.

MR. LIVINGSTON: But I would –

COMMISSIONER BARKER: Just a simple little lawyer!

(Laughter)

MR. LIVINGSTON: But I would suggest that before you do that, you know, you might want to look at what the recidivism study from 1994 says about the predictive value of arrest which is, you know, a bit unsettling.

CHAIR EARP: I get the feeling if we did that, there would be a lot of billable hours for just lawyers. Are there any final questions or comments for this panel?

COMMISSIONER GRIFFIN: I just want to ask, does anyone know of a large employer that actually is doing the post-offer scenario? Is there any large employer out there that's actually said, you know, even for a cost issue, "I'm not going to do a background check on everybody, you know, that arguably is qualified for this job if I'm, you know, hiring a lot of people. You know, I am going to wait because of the cost and because, you know, frankly, it turns out it's not a huge issue a lot of times unless it's, you know, the conviction is, connected to the position.

MS. GINSBERG: I will say that other than the municipalities that were mentioned, I don't know of any where it's an explicit policy.

COMMISSIONER GRIFFIN: Yes.

MS. GINSBERG: It's very common in my practice to see employees who are hired pending their background checks. So employers are certainly hiring people. That's often why they come to us is that four weeks down the line, ...

COMMISSIONER GRIFFIN: Okay.

MS. GINSBERG: ...they get fired when their criminal background check comes back, so there's not an explicit policy.

COMMISSIONER GRIFFIN: Right, right.

MS. GINSBERG: This is one of the big problems, that there's a huge time lag in getting the criminal background check.

COMMISSIONER GRIFFIN: Right. Thank you.

CHAIR EARP: If there are no further questions for this panel with the concurrence of my fellow Commissioners, I'd like to take a 10-minute break. We'll reconvene in 10 minutes.

(Whereupon, a short recess was taken.)

[Video failure caused transcript gap. Statement of panelist is available at http://www.eeoc.gov/abouteeoc/meetings/11-20-08/index.html]

MR. BUSHWAY: -- social scientist to be in this kind of policy setting, particularly because when I did my dissertation in 1996 on the topic of the use of criminal history records for employment purposes and the problems that might have caused for offenders, I was counseled to get out of that area and study something else because it was largely uninteresting and not relevant. So I made sure that I told my advisor that I was presenting in front of the EEOC.

And it turns out it's just over a variety of interests I've stayed in this area over the last 12 years. I've done research on recidivism. I've looked at the process by which individuals desist. I've studied the use and the different types of criminal history records that are used for employment purposes in terms of the types of issues that have already come up here, in terms of criminal history records from public repositories versus court records provided through consumer reporting agencies and then looking at the different quality issues there. And then more recently, I'm trying to answer questions like how long does it take before an ex-offender looks like a non-offender on recidivism and that's the question that was specifically, that type of question, recidivism, was the question I was asked to address today and so that's what I've prepared my comments on, but I have done research on other areas and I'd be open to questions on those topics.

When you talk about assessing risk for ex-offenders, it's important, I think, to sort of think about the initial state, what's currently going on by employers and I think the appropriate term is clinical risk (problems with audio) -- making a decision of the human resource person about the level of risk that a person poses based on their own good judgment, their expertise, their knowledge. And they're not, in most cases, using statistical risk assessment, that is risk assessment pools that have been validated to actually predict risk.

And the best practice that's been typically shown that often times in many -- across many settings and counseling and social work and including criminal histories in the criminal justice system that often clinical risk assessment does no better than chance. And so the best practice is some mix of clinical judgment and expertise and statistical research tools.

However, I'm aware of no case in employment law where a statistical risk assessment is being used. The best you get is rules that are promulgated by employers that say, if you have this particular criminal history record, you can't get hired, but those rules are typically not based on statistical assessments of risk but they're based on legal and clinical judgments of the relative problems associated with different criminal history records. They're not based on risk assessments.

So then the question is, well, while we don't have really the right risk assessment that we'd like to know so what can we have and the issue is what we do have is recidivism studies in the criminal justice system where people do use statistical risk assessment and so that's what I'm going to talk about, but I'd like to sort of throw a caveat is why is recidivism, for re-arrest or re-conviction, or however you want to define recidivism, the relevant measure here when it seems that the relevant measure should be job performance? And so I think that's an important caveat, but there is a large literature on the use of statistical rules to conduct risk assessment of ex-offenders both in terms of just the social science literature on this and also actual use of rules both at sentencing. Virginia, for example, now uses a risk assessment tool to decide who should be incarcerated or not, it's a mix. It's a clinical combined with a statistical treatment rule, in prison management in terms of who gets which level of supervision or what programs in prison, and then finally at parole, who gets paroled and what level of supervision are they provided at parole.

Over half the states now use some type of risk assessment tool at parole. So there is evidence here that this is being used – that this type of thing is being used and it speaks to two questions that come up in SEPTA. In SEPTA as I read it, and I'm not a lawyer, but as I read it from a social science perspective, there are two questions that got asked. One is, is it possible to differentiate between ex-offenders in terms of levels of risk and is it possible to differentiate between ex-offenders and non-offenders? In other words, once an ex-offender, are you always at higher risk? And I think the answer to those two questions are actually pretty clear in social science literature. Yes, it is possible to differentiate between ex-offenders in terms of levels of risk. You can't do it perfectly, but you can do it. You can do better than if you were guessing. And then the second element is, you know, do ex-offenders eventually look like non-offenders and the answer is, yes.

There isn't the case that once an ex-offender always higher risk in terms of recidivism. So, if you look at these studies, and there's meta analyses out there. This is just compiled studies of what the best predictors are, and it's unambiguous. The two best predictors of risk are age and criminal history. Okay, so those two things carry the vast amount of the load. If you had two things that you could ask about, those would be the two that you'd ask and you'd get 80 percent of your predictive power right there.

And so there is some real power in the use of the criminal history record that predicts risk. It's a valid tool for thinking about risk in terms of recidivism. But here when we report to the employment contacts I think we have to throw some caveats out. This kind of study that's done with criminal history records by social scientists with ex-offenders is using repository data, rap sheets, not court records. And there are two important distinctions there, one is the quality of the match so fingerprint quality match as opposed to name and date of birth matches, which are -- name and date of birth matches are not nearly as good as fingerprint matches, and second, disposition data, public repositories from which the FBI gets its data work hard to get dispositions but admit that they have a lot of trouble with that.

Court record data is much worse with respect to disposition data than are repository data. And there's a lot of focus in the federal government about -- and there's a BJS program that intends to make the repository date better but that's not the same thing as the court record data which is significantly worse.

So, I'm going to talk about research that's based on repository data with fingerprint matches which is a much higher quality source, although not perfect, than the typical source of information used by employers, and I think that's a really important distinction.

The quality of the prediction is modest. We shouldn't pretend that we can predict perfectly who's going to offend and who isn't among a group of ex-offenders, but it isn't the case that we can't predict. Okay, so, you know, it isn't that we don't know anything but it isn't the case that we can predict perfectly.

And then when you think about prediction, model fit is not the only issue. In other words, the ability to sort of explain is not the only issue because in the end, you're going to have to put a cutoff line. Anybody above a certain line gets excluded, anybody below a certain line gets included. The problem with that, especially when you're trying to predict very rare events, particularly workplace violence or something like that, you're going to include a lot of people who shouldn't be included when you restrict people.

So for example, if you say, I'm going to restrict anybody that has a greater than 10 percent chance of offending in the next two years, it seems like a reasonable rule. Ninety of the 100 people that you exclude would have been perfectly fine. Okay, so you're going to have, when you have restrictions that are in the very, you know, rare events, you're going to end up having a lot of false positives and that's a real issue with what in criminology we call selective incapacitation, trying to incarcerate only the highest risk people, but it's also the case here as well. It's the same basic issue. You're going to have lots of false positives. So let's talk about the criminal history record. There are three components of a criminal history record that tend to get a lot of focus, the type of offense, time since last offense, and the number of offenses, okay, or seriousness of offenses or as criminologists like to think of it, rate of offending. So this is the first thing that's always surprising when a criminologist talks to a lawyer or someone who's involved in this area is the type of offense is largely not relevant, okay, when you're trying to predict recidivism risk.

Okay, that's immediately is sort of a bit of a head shaker. If you think about - there is a large literature in criminology on specialization. To what extent do people seem to specialize in certain types of offenses as opposed to just being generalists. And the overall consensus is that the primary answer to the question is, they're generalists. There are very few types of specialists, sex offenders as an example of a group that seems to show higher levels of specialization. In general, there isn't a lot of evidence for specialization. There's some, but not very much.

When we think about specialization, and you can see this in the BJS report that many of you are familiar with. Typically the way we approach is, "Okay, given your last offense, what's the probability that your next offense is going to be the same thing"? And there is some slight -- there is some correlation in that dimension. Robbers are more likely to commit a robbery next. Okay, it's important to point out that that's not what employers are doing. They're typically looking at the most severe offense, not so you get excluded after your most severe offense is a certain type of thing, like violence. They're not looking at specialization the way this is typically recorded.

The second point is that if you start to control for other things like time since last offense, length of the criminal history record; that correlation starts to go down a lot. Okay, so I don't want to overstate this. There is some evidence for specialization, but the big story here is people are generalists.

Last comment, in particular if you think you're trying to get at recidivism or probability of re-arrest, those people who are most likely to be categorized as property offenders have higher recidivism rates than do violent offenders. Violent offenders are less likely to re-offend and this is a fairly significant -- substantial finding. You can see that in basically any study you want.

In general, you know, property offenders are higher rate offenders, part just the nature of the offense and the level of commitment you have to have to be a property offender. Okay, so time since last arrest, I've done a number -- the reason why I'm here primarily is because I've done a study on this. I was asked by someone who said, "You know, how long does it take before an ex-offender looks like a non-offender"? And I said, "Gee, I don't know but I can find out because that, for those of you who made it through college level statistics, that's a hazard rate and that's a natural statistic you can calculate pretty easily, and so I did.

I went ahead and calculated with two different data sets follow-up data for between seven and 15 years for two groups of people, those who offended at a certain period of time say at age 20 and those who didn't. And the question is, in each year after that, so when they're 21, what's the probability that the guy who offended at 20 offended at 21, and what's the probability that the person who didn't offend at 20 and has no criminal history record, offended at 21? And then for those who made it past that one year, what's the probability that you offend. For those who survived two years, what's the probability you offend, et cetera.

After I published these two studies with my colleagues, mainly Curley Jack and Robert Brand, there have been a number of replications using bigger data sets, perhaps, more appropriate data sets, but all three of them were presented at the American Society of Criminology meetings last week. One is using New York State repository data, the police repository data. Others are in Europe where they have better data sets for this kind of thing. And everyone is sort of finding the same thing, which is, initially the rates of offending for the people who've just recently offended are much higher than non-offenders, but that declines and eventually the two lines cross. In our study, it was six to eight years. In other words, after eight years, the ex-offenders looked like non-offenders. Depending on how you describe the non-offenders, that different people are doing that differently, the line takes longer to cross but the predominant message everyone is finding is that the line drops, drops considerably and it goes to approach the line of what you would consider to be a non-offender.

So ex-offenders, I think it's pretty unambiguous from the study so far, ex-offenders do, in fact, eventually look like non-offenders. The criminal history record farther out tells you something about the risk. If someone stays clean for a number of years, this is informative. A couple really important ways -- things about the one -- a couple of important things I want to point out with respect to that phrasing of the question; the first thing, this is not when do ex-offenders have zero risk. It's when do ex-offenders look like non-offenders, and non-offenders do not have zero risk.

Okay, second of all, I think non-offenders need to be defined by the pool of people that you may be thinking about in terms of employment. When you're talking about the risk of an individual that's being hired for a particular low wage job, the relevant pool here is the group of people who don't have criminal history records that are also applying for the job or have gotten the job and for whom may have fairly high risk. The other part of that question, I think, is important is, the type of criminal history records that you use for this assessment. If name and date of birth searches from CRAs miss half the people with criminal history records, the pool of people without criminal history records that you hired actually has a large number of people with criminal history records in them and therefore, have higher levels of risk. So I think you need to think about that.

A couple caveats, time since conviction is not the same thing as time since release from prison. Typically, criminal history records only have time since conviction, but it really this is -- the clock starts when you get out of prison. Risk assessment, this was brought up earlier as being done in an environment where the criminal record history has costs. It's possible it would look differently if people weren't being discriminated against on the basis of their criminal history.

Finally, the biggest thing that carries the weight in terms of risk prediction, is the length of criminal history record. That's a very powerful tool. I don't see a lot of people using that in employment that I can tell. The interesting question is whether the number of offenses or the criminal history rate is relevant as you -- for these hazard rates that I've tried to describe. You know, the people with higher rate offenders are going to fail first, so the question is, six, eight, 10 years out, is it still relevant that you've had five offenses, five convictions versus two? And the answer is, I don't know. There's a bunch of people trying to get at that. I see a red light.

CHAIR EARP: Wow. After the break someone representing an employer who is still sitting in the audience asked me a question and I'll ask you, Mr. Bushway. I don't know if you're the appropriate person or not, but I'll ask you anyway just to start. In terms of predicting success on the job, this is an employer that hires thousands and thousands of workers at various seasons throughout the year, does it make a difference whether the person is paroled and has no further contact with the judicial system or is on probation in terms of probability of success on the job or risk on the job?

MR. BUSHWAY: So in terms of being able to answer that, what I can share with you is some results that people have looked at with respect to -- for prison, coming out of prison, looking at the official employment, so there's some controversy about what the right measure of employment is, but if you think it's formal work as measured by the unemployment insurance programs in the different states, you can actually get employment information before someone goes into prison, and after. And you can sort of compare what happens and what people have now shown in three different states is that in the three years immediately prior a release from prison, I am sorry, immediately after a release from prison, the employment actually goes up, relative to what had happened before.

Now this may not be the relevant comparison for looking at the impact of incarceration on employment because you'd like to know what would happen if you hadn't been in prison, so you need a control group. But if you just sort of thing about that comparison, it does go up. And the obvious potential explanation is parole, that this is a requirement of supervision and then people are more motivated to work.

To the extent to which that's been tested, that hasn't been proven to be true, but it's hard to say with any clear proof, but that's the only evidence. So there is some evidence of, you know, sort of two, three year blip that eventually disappears. Some people think that's parole. I don't know for sure.

CHAIR EARP: Good, thank you. Commissioner Ishimaru?

COMMISSIONER ISHIMARU: In the work you mentioned, recidivism goes to another crime being committed, it doesn't -- I assume it doesn't go to where the crime was committed. So when the concerns were raised earlier about workplace violence and the employer having a legitimate interest to make sure that they provide a safe workplace, that isn't necessarily linked to whether a person will commit a future crime or not. That it may not -- where the future crime may be committed isn't necessarily in the workplace?

MR. BUSHWAY: Right. That's right. I mean, this is an attempt to answer the question that a lot of you want answered in terms of relative risk. The only available data that I'm aware of and anyone's aware of is, this type of recidivism study where it looks at re-arrest, which is, as I mentioned, maybe not the right question to be asking. This is the best you've got.

COMMISSIONER ISHIMARU: Right, right.

MR. BUSHWAY: And the only caveat I would throw up is workforce criminal behavior is extremely -- is fairly, you know, it's pretty rare and to try to start to predict very rare events, it gets harder.

COMMISSIONER GRIFFIN: Has anyone -- just to piggy back, has anyone actually looked at in the workplace -- when a crime is committed whether it's theft or you know, or violence, do you know of anyone that's looked at the -- you know, who's committing that, are they ex-offenders or non-offenders? What's the percentage?

MR. BUSHWAY: Okay, so that is the obvious next study to do and I wish I could do it, the problem is getting employers to give you access to data or finding other ways in like through licenses. So the federal government, as well as the state government do background checks all the time and have also some information about the employment performance of people that have criminal history records that are hired. And if you could put the right agencies together, you could do a study that sort of looks at that. The best agency that I'm aware of that has done something similar is the Department of Defense with respect to their hiring -- enlisting people with criminal history records and they've tried to look -- they've tried to do some followup of the job performance of the individuals with and without criminal history records.

Some of you may have seen the article that started out in the Sacramento Bee about the court-martialing. They looked at the court-martials of people in Iraq and looked at whether they had criminal history records and tried to determine -- so that was a crime, a workplace crime in a sense, and tried to make the argument that that -- those people that were court marshaled had much higher rates of criminal history records than in the general population. They don't have the evidence to actually state that, so they didn't prove that, but that's where -- you know, that's the example of the type of thing that you could do.

The Department of Defense has shown that sort of job performance measures of the individuals from which the criminal history records seem to be worse. This is like whether you make it through boot camp. There's some evidence that they don't perform as well. But it's not published in any kind of, you know, there's -- it's funded by the Department of Defense by and individual and he filed the report kind of thing. It's not something that's in the academic literature.

But you can also look up the article in the Sacramento Bee. It was syndicated, so it hit a fair number of newspapers this summer.

COMMISSIONER GRIFFIN: Thank you. Sorry I jumped in on your time, Stuart.

COMMISSIONER ISHIMARU: No, no, nothing else for now.

CHAIR EARP: Commissioner Griffin?

COMMISSIONER GRIFFIN: Well, yeah. If employers, you know, they seem to be biased against you know, maybe hiring especially very newly released offenders or people that have been arrested, you know, very recently for something. Has anyone looked at if because of that you can't get a job, is now your rate of recidivism is now higher because economically you're in a tough spot? Is that factored into the study or --

MR. BUSHWAY: Okay, so that's a huge literature which is somewhat controversial, I've also written reviews on that and so we can talk about that a little bit. One way to phrase that is, if you do employment programs for ex-offenders, do you seem to reduce recidivism? Okay, so one is -- so that's -- yeah, and there's a large number of attempts to do that, for example, Supported Work, if, you might be familiar with that, that's a federal program in the `70s. And the evidence there was that for older offenders, there did seem to be an impact on recidivism, so people over the age of 27, but under the age of 27, there was no impact. The effects are modest but reasonable given the level of funding. I mean, in other words, you don't expect -- when you spend $1,000.00 on something, you don't expect the world, you know. Rate of return was reasonable, 10, 20 percent reductions.

If you look at employment programs in prison, vocational training programs, that estimate appears to be about 20 percent reduction in recidivism, although controls for selection, which is the big issue here, you know, is -- you know, aren't great but that seems to be a reasonable estimate. The young offenders are particularly hard to serve. If you look at job corps studies, job corps has shown some impact on criminal behavior, but it's not huge for work programs. And I think that you know, Diane was talking -- Diane, I think, you guys have the experiment going now and the folks at CEO which she mentioned have a study going on now. So CEO is a study in New York where they work with people coming out of prison and they offer them transitional employment and then movement to work.

And they were finding something kind of interesting one year out. What they found is the employment outcomes of the people -- so they have a treatment group that get transitional employment like they actually give them a day job. And then they help them find additional employment. One year out the employment outcomes of the treatment group versus the control group is almost exactly the same. So they don't have a long-term impact on employment outcomes but for those people who kind of hook up with the program fairly soon after release from prison, there's a dramatic impact on the recidivism rate. It's about a reduction of half.

Now, that's one of the biggest effects I've seen ever seen. And I think in part what's going on here is the period right after prison is particularly vulnerable and it's not clear that it's necessarily just employment because remember, you're involved with this group and there's a lot of other support services and things, but it seems like it isn't a long-term impact on employment so economists would say, this doesn't make any sense because there's no long-term impact on employment, but, in the short term it appears that something having a structured program, having some direction, having some support seems to have an impact and it's at least marginally employment related, because the program is an employment program.

So I think the overall statement is that there's a modest relationship between work and crime and I think that, you know, I don't think you want to overstate it, because I think if you do, you're going to find that there's some evidence against you. But I think it's reasonable to conclude that employment does reduce recidivism.

COMMISSIONER GRIFFIN: Thank you.

CHAIR EARP: Commissioner Barker?

COMMISSIONER BARKER: I really don't have any questions. A really interesting presentation, Mr. Bushway, really appreciate it.

MR. BUSHWAY: Thank you.

CHAIR EARP: Commissioner Ishimaru?

COMMISSIONER ISHIMARU: Well, now it was a fascinating presentation and I guess my question would be looking forward of unanswered research questions and you knowing the literature out there, what -- you know, what are the next three big research type questions that are out there and things, perhaps we should be focusing on here if we were, you know, either doing the research or assisting in providing funding or something, you know, what are the open questions that are out there that need to be addressed?

MR. BUSHWAY: Okay, I can think of two big ones and that come right to mind. The first one is, I think you really need to address more forcefully the nature of the criminal history record information that's being discussed here. Do not treat official -- repository information and court record information that's obtained through private sources the same thing. They are not.

I did a study where we compared -- we had 120 people that we knew we had criminal history records. We did a background check by the FBI. We did a background check through a private court record search and we got 86 out of 120 with the FBI and we got 56 out of 120 with the court record search. You know, so they're not the same thing and I think we need to really seriously consider the source of information here, it matters both in terms of the quality of the match and the quality of the dispositions. And I think that that needs to be part of the conversation in the way I just don't see it being part of the conversation. And there's a way to study this. There are people involved with BJS and there has been some interest by the consumer reporting agencies to sort of get at this, which would involve a large scale study of background checks and subsequent, by two different sources. What would happen if you go through the FBI? What would happen if you go through a CRA? And I know from what I've understood from talking to Gerald Ramker at BJS that there has been some interest by the private agencies to do that.

Coupled with that, you could potentially start to think about creating risk assessment tools. You know, what predicts risk and in that context I think the second study is to really get this out of the conversation of recidivism. The language that you use isn't really about recidivism, it's about employment performance and I think it's not unreasonable to think that employment performance is related to criminal history records and I think Devah's work sort of shows that. And I think most employers are actually just worried about trustworthiness of the workers and I think the big guys -- I think the distinction has been made very clearly, the big firms which is generally not what Devah was working with, are worried about negligent hiring lawsuits, but the little guys are just worried about the quality of workforce. They're trying to get people that are good.

And I think in that context, just trying to understand the relationship between criminal history record and the performance on the job, and there are ways to get at that through government agencies records, license performance. For example, in New York State they have -- the Health Department does the background checks for the nursing homes, and so they also license and have to -- the nursing home has to let the person -- the Health Department know when the person is no longer working there. So you have that information. The question is being able to get someone to let you study it.

And so I think you know, looking more closely at the link between criminal history records and job performance, I think is a really important step. But I think the first step is figuring out what you mean by criminal history record and really sort of being more clear in understanding the difference in the qualities between the sources, because I think the differences are quite large.

And then, of course, it would be nice to know -- you know, you could do experiments about employment programs for ex-offenders and fund people like Diane to do -- you know, who are trying really hard in this area to see what kind of works, what makes -- if you're trying to service, particularly the very high risk people coming right out of prison or right after conviction, where the risk is very high, you know, what are the kind of things that seem to work in that area. And the funding for programs in that area is just very small.

So I think you know, there's a lot to be done. We know that's a high risk time but the amount of money and time to do that is hard.

CHAIR EARP: So an employer listening to your testimony would say despite the public policy concerns regarding criminal history, possible disparate impact, if there is no positive correlation between work and recidivism, setting job performance aside, why should I as the employer, spend the time and the money and take the risk to hire this group of workers that it may take six to eight years before they return the investment to me?

MR. BUSHWAY: Okay, so be really careful on that and you didn't get to the last part of my comments which is that there are mitigators which is what the last thing that I was asked to address which is are there other things in addition to the criminal history records that sort of tell you something about an individual that in some sense counterbalance and age is a big one. So people don't have, you know, 45 year old with -- who's been out a couple of years is no where near the same level of risk as a 22-year old.

Employment, stable employment is another one. The -- if you -- you know, Diane will tell you, it's hard to get some of these folks coming right out of prison to have a stable job for a year, that's a major accomplishment. And if they do, whether it's causal or not, the reality is, if you have a stable job for a year, you are at very low risk for re-offending. And so I always find it slightly amusing, if not frustrating, to hear about these folks that get fired have retroactive criminal history records after they've been working for five years, because that's saying that the information about the criminal history record trumps what the employer knows from hiring this person for five years and I think -- I find that on the face implausible.

So there are other things that affect, so I don't think it's the case that you have to -- I think there are other pieces of information in a puzzle. And any responsible statistical risk assessment tool would not focus only on a criminal history record, would consider these things as -- I don't know why criminal history record has been elevated as the trump card. I think I know why, it's negligent hiring, but it's not clear to me why it's -- you know, it trumps everything else. I think the package of things tells you something about an individual and I want to repeat, I did not say that employment did not have an impact on crime and recidivism. I said the effect was modest, that's not the same thing.

Tell me something that's -- you know, 20 percent which is what people are finding for these types of programs, 20 percent is not -- nothing. And it's -- you know, show me another program that reduces recidivism rate 20 percent.

CHAIR EARP: Okay, are there other questions or comments? Anything Commissioner? Thank you, Mr. Bushway, fascinating.

MR. BUSHWAY: Thank you.

CHAIR EARP: The final panel, employer practices. Rae T. Vann, General Counsel at the Equal Employment Advisory Council and Laura Moskowitz, Staff Attorney at Second Chance Labor Project at NE -- National Employment Law Project. Thank you, welcome. Thank you for your patience. Rae, why don't we start with you?

MS. VANN: Sure. Thank you, Madam Chair, Commissioner Ishimaru, Commissioner Griffin and Commissioner Barker and colleagues. Thank you for -- on behalf of EEAC for having me participate in this very fascinating meeting today on behalf of the EEAC.

I've been asked, as you mentioned, to discuss employer practices, what employers, at least in our experience, are doing with respect to the use of criminal arrest and conviction records. I commend the Commission for taking on this issue which is extremely important, particularly to the large employer base that EEAC represents. I would say at the outset that my comments are not going to speak very much at all to the issue of arrest records because what we have found is that in our experience, EEAC member companies don't take into account or consider in the employment selection process, arrest records that don't ultimately lead to convictions.

As -- I'll describe it more fully in a few moments, EEAC member companies, by and large, are pretty heavily regulated, given what they do and who they are and many operate in many jurisdictions and as has been discussed throughout the course of the day, there are a number of state laws that restrict or even prohibit the use of arrest records or the circumstances under which that information can be considered. So by and large, in our experience, most EEAC member companies don't take into account arrest records, except in very minor and extraordinary circumstances, for instance where there might be a need for a national security clearance and there's some overarching obligation to conduct more stringent reviews that might include consideration of arrest records.

As all of you know, and as others have discussed today, over the last several years there have been a number of new laws and regulations that have been put into place that require more and more employers to conduct criminal background investigations or background investigations that also include criminal background checks. EEAC, I should say, is comprised of large employers. It's an association of over 300 Fortune 300 corporations who, by and large, are federal government contractors. Probably over 90 percent of EEAC's membership is comprised of federal government contractors who are subject not only to Title VII and the state and federal employment non-discrimination laws, but they're also subject to federal government affirmative action program requirements that are administered by the Office of Federal Contract Compliance Programs.

In addition, many EEAC members are regulated by other laws. They're in heavily regulated industries such as health care, insurance, financial services, banking and that sort of thing. And so there are other laws and regulations that require these companies to take into consideration an individual's criminal conviction history with respect to certain convictions for hiring into certain jobs.

Having said that, EEAC certainly recognizes the danger in having arbitrary policies that categorically exclude as was discussed earlier, all individuals with criminal conviction histories and records. Member companies, by and large, take a balancing approach consistent with the laws that they're subject to, state and federal laws, and the guidance that EEOC has on the books currently that look at various factors in taking into account whether or not a criminal conviction will serve as a disqualifying factor to employment.

I also want to say, and this is somewhat off topic, but there have been a number of issues that have been raised this morning that I'd like to at least attempt to address. With respect to the -- and I don't recall what it was referred to earlier, but banning an employer from being able to ask the question up front, as a practical matter, based on conversations that I have had with EEAC member companies; a good majority of our members conduct the inquiry at the post-offer, pre-employment stage of the process.

Having said that, though, as I was sitting there listening to folks talk about this, it occurred to me, particularly when we start talking about the ADA, and start talking about age being a factor that ought to be thought about in considering whether or not this information should be considered, especially when you talk about the ADA, it's my understanding that the reason why we have the rule, the post-offer, pre-employment rule is to be able to segregate disability related information from everything else. So I would just ask you -- and it's not part of my formal recommendation because it just occurred to me as we were discussing it, to take that into consideration if to the extent that the agency decides that it's going to consider imposing a post-offer, pre-employment rule, because it kind of muddies the waters a little bit with respect to ADA inquiries. And again, I have not thought it through beyond this morning to the extent that it was raised here but I'm happy to go back and talk to my members and think about it more to the extent you think it may be helpful.

My testimony today is going to focus on some of the reasons why employers are conducting background investigations that may include a review of criminal conviction histories and some of the procedures that companies are using to insure that their legitimate business interests are served in a way that minimizes discrimination, particularly against protected groups, people of color, who may be more likely to suffer adverse impact as a result of the use of this information.

I'll conclude my remarks with some recommendations for ways that we think the EEOC can clarify its existing regulations or, enforcement guidance I should say, in order to take into account some of these 21st Century national security related obligations that many more employers now are faced with, with respect to the question of first conducting criminal background investigations and then to the extent that some employers are required to consider certain offenses in making employment selection decisions.

Well, why do employers conduct criminal background investigations? As I said, and as Mr. Livingston pointed out, I believe, many employers, large employers, perform background investigations that include criminal records checks because simply they're required to based on other laws and regulations that they're required to comply with. For instance, Section 19 of the Federal Deposit Insurance Act bars financial institutions from hiring anyone who has been convicted of any criminal offense involving dishonesty, breach of trust or money laundering unless and until they are able to obtain a written consent letter from the FDIC essentially.

Insurance companies are subject to similar rules. They are prohibited from willfully permitting any person who has been convicted of insurance fraud or similar crimes involving dishonesty, to participate in the business unless, as is the case with banks, they receive a written consent from the regulatory body, the applicable regulatory body. It's a little more difficult or sticky for insurance companies because the term "participate in the business", has been interpreted quite broadly and basically can be read to apply to anyone who works for an insurance company. So that imposes -- and I should say there are pretty hefty fines associated with violating some of these laws, including one particular provision that imposes a million dollar per day fine potentially. So employers who are subject to these legal requirements try to be very careful in who they consider for employment because failure to conduct the right investigations or inquiries could lead to significant legal penalties under these rules.

CHAIR EARP: Rae, can I interrupt you for just one second?

MS. VANN: Sure.

CHAIR EARP: This scenario probably doesn't exist in today's environment, but if there were a bank somewhere that hired its own janitorial staff as opposed to contracting out that staff, are you suggesting that the any person would include not just the tellers and the bank officers, but the janitor?

MS. VANN: I don't know if that's a requirement for banks. It certainly is for individuals within the insurance industry. The body that regulates insurance companies has construed that term "participating in the business", participating in the business more broadly than I think it's been construed in the context of financial institutions, banks and things like that. I think my understanding is that in the banking sector, you're talking about people who are dealing with people's money --

CHAIR EARP: Right.

MS. VANN: -- and that sort of thing, so there's that job-relatedness component. That is not necessarily applicable in the insurance context based on how the regulations have been interpreted.

CHAIR EARP: But in insurance, it could be construed so broadly to include the janitor if such a non-contract janitor actually existed somewhere.

MS. VANN: It could be.

COMMISSIONER ISHIMARU: Yeah, but has it ever been?

MS. VANN: I don't --

COMMISSIONER ISHIMARU: And have fines that you mentioned, you know, the million dollar fines, which obviously is a lot of money per day, has that ever happened?

MS. VANN: I don't know. I don't know the answer to that. And as I'll discuss later on, beyond those legal requirements, those two specific laws that I mentioned, by and large, large employers look at the relationship of the specific criminal conviction or criminal record information to successful, likely successful performance in the job. There may be instances in which a criminal conviction, depending on how recent it is, depending on the severity of the offense, could serve as an automatic disqualifier with respect to a particular position. But it is really as a practical matter, a case by case assessment that's undertaken in many instances. I'm not going to suggest that there are not large companies that have -- that don't have blanket disqualification rules. Some do but typically, again, it goes to the question of how this particular serious offense affects successful performance in the job or may potentially impact upon the safety of customers and other employees.

Does that answer your question, Madam Chair?

CHAIR EARP: Yes, yes.

MS. VANN: All right, well, outside of these over-arching legal requirements that some employers are obligated to operate under, what might motivate employers who are not required by some law to perform criminal background investigations, and I think we've touched upon those issues throughout the course of the morning. For one thing, depending on the requirements of a particular job, certain criminal conviction information may be very relevant in assessing the individual's ability to perform the job in a safe and acceptable manner. For instance, but again, it's a case by case assessment. A company that employs drivers to transport merchandise from Point A to Point B may legitimately disqualify someone who has a history of criminally reckless driving or of driving under the influence of alcohol or controlled substances. At the same time, that criminal record may very well be irrelevant to that applicant's consideration for another job that doesn't involve driving specifically.

I have the red light, so I'll stop.

CHAIR EARP: I'll give you a minute. I'm sorry, I interrupted you.

MS. VANN: Let me just go to my recommendations. Again, as I said, EEAC recognizes the importance of insuring that criminal conviction records do not inappropriately and arbitrarily exclude qualified candidates from consideration for employment and obviously, we recognize that there is proven disparate impact against certain protected groups, people of color, in particular.

However, as we've discussed, many employers, especially federal government contractors and those in heavily regulated industries such as insurance, healthcare and financial services, now are required to perform detailed criminal background investigations and to automatically disqualify certain applicants based on certain criminal offenses -- convictions, I should say. So we would strongly encourage the Commission, if it decides to update its current enforcement guidance, to make clear that an employer's reliance on those laws is sufficient to demonstrate business necessity in cases where adverse impact is shown.

Furthermore, we would ask the Commission to emphasize the categorical bars on the employment of persons who have been convicted of serious violent crimes will not violate Title VII as long as the prohibition is demonstrated by the employer to be job related and consistent with business necessity, which I think is consistent with what you've heard from others. Thank you.

CHAIR EARP: Okay, thank you. Ms. Moskowitz.

MS. MOSKOWITZ: Madam Chair, Commissioners, thank you for the opportunity to participate in this important meeting. I'm a Staff Attorney with the National Employment Law Project and our program has a project called the Second Chance Labor Project which works to reduce unfair barriers to employment for people with criminal records.

We've been asked to comment on employer practices today as well and I'm going to start with an overview of the current landscape of criminal background checks which at this point will just a recap since we've heard a lot already today and then discuss our view on best practices and how the Commission can provide additional guidance for employers in these areas.

I also want to highlight the "ban the box" movement which has gotten significant air time today as well, where public employers are leading the way towards less discriminatory hiring practices and I also want to note some assistance available for employers in hiring people with criminal records.

So as we've heard a lot about today, the context in which workers and employers are grappling with issues about hiring workers with criminal records has changed dramatically in the decade since the Commission issued its existing guidances on employer consideration of arrests and conviction history and this is due, as we've heard primarily today, the exponential growth in the number of people who have had contact with the criminal justice system and also the number of employers conducting criminal background checks, particularly since 9/11, and Rae mentioned some of those particular areas.

So I won't go a lot into the statistics but one in five adults now in the United States has a criminal record that will show up on a routine employment background check. These often include arrests and often arrests that don't lead to conviction and they're overwhelmingly for non-violent offenses.

Just between 1987, the year of the Commission's conviction guidance and 2007, the incarceration rate nearly tripled and now approximately one in 100 American adults are incarcerated, as we heard earlier, close to 700,000 individuals returning each year to re-enter their communities after release from incarceration.

Again, we know that most large employers are conducting criminal background checks and they are typically using the private screening firms that we talked about like Choice Point that generate millions of background checks a year and as we discussed, much of this information is now available publicly on the internet as well. And although there's a lot of information out there on the potential for negligent hiring liability, there's not a lot of information for employers on how to comply with Title VII when considering arrest and convictions.

Advocates on behalf of workers with criminal records typically find that employers are unaware, maybe not some of the larger employers, are unaware of the EEOC's existing policies on considering arrest and convictions and we know that close to 60 percent of employers have told researchers that they would not or probably would not hire someone with a conviction history.

As we heard earlier from Devah Pager, the mark of a criminal record is significant at least to dramatically reduce employment outcomes particularly for people of color who are already experiencing race discrimination in employment.

So why does the EEOC play such a critical role here? As the Commission has already recognized and as we've all acknowledged today, the racial disparities in our criminal justice system mean that hiring policies that screen for arrest and conviction histories are going to have a disparate racial impact. The current racial disparity statistics in the criminal justice system reflect the ongoing nature of this problem and again, I know we've heard a lot about the statistics and they're in the written statement. I'm not going to go into all of them but suffice it to say that African Americans and Latinos continue to be arrested and convicted at rates disproportionate to their overall representation in the population as do Native Americans.

And these racial disparities are largely rooted in law enforcement practices that disproportionately target people of color such as the war on drugs, which has led to higher arrest and conviction rates in some minority communities. So it's the combination of these factors, increased numbers of Americans coming into contact with the criminal justice system, the racial disparities in the criminal justice system and the growing numbers of employers conducting background checks that leads to this increased employment discrimination and exclusion of minorities from employment.

So what are our suggestions or the practices we recommend for employer consideration of arrests and convictions? First, with respect to arrests that have not led to convictions, we echo what a lot of the panelists have said and what Commissioner Griffin mentioned earlier about arrests. We encourage the Commission to adopt a bright line rule that arrests that have not led to convictions should not be a permissible line of inquiry. The Commission has already noted in the current guidance that arrest records, unlike conviction records, are nothing more than an allegation, and as mentioned earlier, arrests often do not lead to conviction and they also present significant challenges for employers in interpreting this information which we've talked to some degree about today.

Often the disposition information is missing, such as the fact that the charges were dropped or that the person was convicted of a much lesser offense that looks less serious than the initial charge. And as other people have mentioned and the Commission has noted in the current guidance, that a substantial number of the states, 20 states, have prohibited or advised against pre-employment arrest inquiries because of the possible misuse of this information. This has created a disjointed approach right now in states like California, for example, and Pennsylvania, where under state law employers are not allowed to consider arrest but the EEOC guidance would allow for that consideration.

And while we respect that the current arrest guidance attempted to craft a fair policy where in addition to requiring employers to look at the job relatedness and how recent the arrest is, to also make that added inquiry, the person actually engaged in the conduct alleged. In theory that might be okay, but we know in practice most employers are not conducting that kind of inquiry and, in fact, they're more often misled by arrest records and often unfairly denied employment as a result. So again, we encourage the EEOC to adopt a policy for no consideration of arrest records except perhaps in certain security sensitive situations that are mentioned in the current guidance.

Now, second, with respect to convictions, we believe that the Commission's current policy retains -- should be retained in terms of some of the essential elements. First the presumption that employers' policy of excluding people based on criminal records has a disparate impact on African Americans and Latinos.

Second, the rule that blanket policies barring employment based on criminal records is impermissible, absent a justifying business necessity.

And third, the requirement that the employer justify its hiring decisions by connecting an individualized assessment of each applicant taking into account some of the factors listed in the current guidance, the nature and severity of the offenses, the nature of the job, how they're sought and also whether the conviction is still determinative based on the time that's passed.

In addition, we encourage the Commission to adopt some additional guidelines for employers, the first one being that convictions should not be considered if the person has not re-offended within seven years after release from incarceration and this goes back to Shawn Bushway's research which you've heard about, which is demonstrated that permanent disqualifications are unwarranted and unfair barriers to employment because after six or seven years, or six to eight years, without re-offending, a person with a criminal record is no more likely to re-offend than someone who has never offended.

Second, we believe that minor offenses for which prosecution is highly discretionary should not be considered absent compelling circumstances. And we believe that employers should -- in any situation where an employer is considering a criminal record, the employer should weigh additional factors, including the age at the time of the offense, the work history and rehabilitation. Rehabilitation and work history were in the prior guidance and we think those should be retained.

We also suggest that the Commission consider some background check models. One of the models that we believe is promising is the federal post-9/11 background check for port workers which limits the types and ages of disqualifying offenses and also gives workers a really important opportunity to seek a waiver of any disqualification upon a showing of rehabilitation.

Another model that we've talked a lot about today, again, is the "ban the box" policies where employers postpone the background check until a final stage of the hiring process. The initial data that's tracking the outcomes of these "ban the box" policies around the country is favorable. For example, in Minneapolis, fewer applicants are being rejected based on the criminal record. More applicants with criminal records that they would call of concern are being hired and it is saving staff time. And I was really glad to hear from Diane about the positive outcomes in Chicago as well.

And finally, just a couple quick notes on what employers can do to make it easier to hire people with criminal records; there's the Department of Labor's free bonding program which insures employers against losses due to employee's dishonesty or theft and also the Federal Work Opportunity Tax Credit, which allows a company to claim a tax credit for hiring an employee with a felony conviction within one year of the date of his or her conviction or release from incarceration.

So again, we think this is an incredibly important area for the Commission to continue to play its role here and we encourage you to put out an updated guidance and establish some sound and fair screening policies for employers when considering arrest and conviction histories. Thank you.

CHAIR EARP: Thank you. Commissioner Ishimaru?

COMMISSIONER ISHIMARU: No questions for now, Madam Chair.

CHAIR EARP: Commissioner Griffin?

COMMISSIONER GRIFFIN: I'm interested in the bonding program. I know they say that they have a high success rate. Do you have any sense of who takes advantage of that?

MS. MOSKOWITZ: You know, from my experience in practice, there's not many employers that know about it or take advantage of it. I saw that the number --

COMMISSIONER GRIFFIN: Not heard about it --

MS. MOSKOWITZ: -- it's not a huge number, yeah, since it was established. So I personally don't, I'm sorry, I don't know that many employers taking advantage of it and one of the reasons I've heard recently for that is because it used to be offered solely through the Department of Labor and now it's been sort of devolved to the state, sort of employment offices and unemployment offices, the one-stop centers and that's created some difficulties with you know, how do employers find out about it, how do they get it, all that kind of stuff, how is it administered. So I think that's part of the problem there.

COMMISSIONER GRIFFIN: Diane, is that something that you talk to employers about at all?

MS. WILLIAMS: We do. In fact, at one point we were the only users of that in the Chicago metropolitan area. And more than anything, quite frankly, it's sort of like a deal closer, so you don't necessarily lead with that, but when you have somebody that's close to --

COMMISSIONER GRIFFIN: That are on the fence, yeah.

MS. WILLIAMS: -- you know, you give them that added incentive and that's pretty much how we've used it. We are fortunate that our state continues to allow us to offer those. Many states will now make you pay for them. And so all the guidelines have changed since they were initially implemented.

COMMISSIONER GRIFFIN: Okay, that's interesting. And then I just want to clarify, Rae, because I think I heard you say lots of the employers that are your members have these federal rules that they have to live with and yet, at the same time, they're doing -- they're doing this post-offer. They're doing this check post-offer, so --

MS. VANN: Many are. I can't say that all are but many are simply because -- well, for various reasons, one being, and I think someone else recognized it, large employers get thousands or if not hundreds of thousands of expressions of interest per job and it would be -- it could be prohibitively expensive to do these types of background investigations on everyone who comes through the door. So I think probably one of the reasons why some companies perform the investigation at the pre-offer -- I mean, post-offer, pre-employment stage is because it whittles down the number of people who need to have that review conducted. Furthermore, it makes better sense.

You're looking at someone who appears to be minimally qualified for the position so you've kind of ruled out other things. So taking into consideration a potentially disqualifying criminal conviction may make better sense at that stage, particularly when you can have -- you have more time to do the case-by-case assessment and have a conversation with people about whether or not there may be mitigating circumstances that should be taken into account.

COMMISSIONER GRIFFIN: Just the whole issue with the ADA, it was really the process. I don't think we were really comparing people with disabilities with, you know, people who have arrests or have conviction records. But what we're just saying is that the way the process works is, you know, you just don't get to that issue. So you get to build that rapport and look at someone's skills and do all the things. So I think the effect is, you know, somewhat the same.

MS. VANN: Well, yes, understood, and my only point was that the Commission ought to maybe consider what impact that would have on consideration of disability related information and enforcement, because again, if you want to segregate disability related information from everything else to determine whether or not disability discrimination is occurring, making a rule whereby you only consider criminal conviction records at the post-offer, pre-employment stage might muddy that. I don't know.

COMMISSIONER GRIFFIN: Yeah, right.

MS. VANN: And as far as age is concerned, I have to say that my -- EEAC's member companies would be horrified if they were told that they would have -- that they have to -- and I understand why, but, as you know, many state laws aren't like the ADEA where it's a 40 and over protection. New York State law protects anyone of any age and if you get into the business of considering age, I'm telling you that my member -- EEAC member companies would be completely horrified at the prospect of being told that they have to consider age in making decisions about --

COMMISSIONER GRIFFIN: When, that came up, I was actually sitting here trying to figure out how does that all fit in, but it's -- I'm not sure, but I'm thinking, well, now you're really -- if it was post-offer and you're thinking that you're really talking about this specific issue about when you offended and you know, I don't know. I don't know how we'd get around it but I thought about it as well that, you know, you can't be saying, "Well, I'm going to ask age now because I've got to consider criminal record".

CHAIR EARP: Commissioner Barker?

COMMISSIONER BARKER: You know, I just have one thought and that is, we've had -- all the testimony today has been so informative and a lot of work has gone into it, but I guess one thing as a Commission, I think, we can't lose sight of is that what we're talking about today is disparate impact against African Americans and apparently Hispanics. You mentioned minorities, you used the term "minorities", but what I heard you say specifically was African Americans and Hispanics.

So I think when it comes to arrest and conviction records, there may be some practices that are unfair to people who have conviction records, but those practices are not necessarily a violation of Title VII. So I think we have to be careful to separate what is within the parameters of the EEOC and what is just you know, it's unfair. And if there are -- if there are practices that have a disparate impact on Blacks and Hispanics, that we address.

If it has a disparate impact on unprotected groups, it may not be a good practice, it may not be a fair practice, but we don't have the jurisdiction to do anything about it.

Another thing, you know, I'm sitting here thinking is, suppose you had a situation where you're in a state like maybe Utah, I don't know, that is like very few African American or Hispanic applicants or within that state a particular industry and within that industry a particular location and whatever so that you get down to the point that a particular company for a particular job description has only White male applicants. Are we saying that that company in that situation cannot use arrest and conviction records, or is it only where you're dealing with White and Black, which is most cases, and because of the disparate impact on Blacks it cannot be used?

In other words, if we take those protected groups out of the equation, is the practice in and of itself illegal? I think the answer is no. I don't think that occurs often but I think the answer is no.

MS. MOSKOWITZ: I would just comment, I know there's an unpublished case out of Pennsylvania that dealt with this issue, that a White applicant had highlighted a blanket hiring policy where the employer wasn't hiring anyone who had a criminal record and the court took a look at EEOC guidance and said, you know, that type of policy is not permissible, but I think in general response to your question, what's within the purview of the EEOC is disparate racial impact on behalf of the protected classes under Title VII.

And while there's a lot of things that are unfair or aren't good public policy for the country overall given the rate of incarceration and the number of people with criminal records, there's no other federal statute that deals with that type of unfairness, but because of the stark racial disparities in the criminal justice system, it does come at the feet of the EEOC to develop some sound policies here.

COMMISSIONER BARKER: To the extent that it addresses disparate impact on certain protected groups.

MS. MOSKOWITZ: Exactly, right.

MS. VANN: Yeah, and that's a point that I would encourage the Commission to emphasize in its deliberations. I don't know that it would be appropriate for the EEOC in its enforcement guidance to create a new protected category of criminal offenders outside of what already is prohibited by the law, which is the law that the EEOC enforces, which is the disparate impact against protected groups. It's not -- respectfully, it's not the EEOC's charge currently to carve out a new protected group.

MS. MOSKOWITZ: And I don't think that's what any of us were suggesting today but I understand it's the concern about general unfairness in terms of hiring people with criminal records.

CHAIR EARP: Any other questions?

COMMISSIONER ISHIMARU: I'm interested in trying to get a feel for how companies actually use criminal background checks, pre-offer, you know, as a screening device. You had mentioned, Ms. Vann, that many of your member companies get so many inquiries, so many applications, that they use various screening techniques to get down to a more workable number.

When -- as a general matter, when does it kick in for your member companies, when might they use a pre-offer or pre-interview as a screening tool?

MS. VANN: Well, I heard from some companies who have developed, for instance, these matrices that help to assist either their own HR personnel or their third party vendors in evaluating the relevance of this information and it could come into play at the pre-offer stage depending on when it kicks in but, if for example, you have somebody who has disclosed on an application a criminal felony conviction for embezzlement and they're applying for a job and maybe it's not with a company that's in one of these heavily regulated industries, but they're applying for a job where that recent conviction would be relevant, a company might use the matrices that they've established to determine whether or not a red flag should go up.

And from the folks who I spoke to, typically the process is except for that very narrow category of, you know, offenses that are always barred, typically there's an opportunity to have a dialogue with the applicant about what the offense was and whether or not there's some mitigating circumstance or information that ought to be considered with respect to this particular job for which the applicant has applied.

COMMISSIONER ISHIMARU: Do your members have a preference of whether they should be allowed to ask the question versus running a check?

MS. VANN: I don't know. I didn't ask the question and frankly, I hadn't heard the "no box ban" terminology until just this morning. I don't know if employers really understand that this is an issue that's being discussed. I can certainly take it back and I'd be happy to ask folks about it, but frankly, I have not given it much thought and I certainly didn't ask the folks who are members of our association about it.

COMMISSIONER ISHIMARU: Ms. Moskowitz, a similar question; do you have any feel for how often employers use criminal background checks as a screening tool pre-offer?

MS. MOSKOWITZ: I would say in my experience, I tend to agree that most of it's done by self-disclosure, that the pre-offer is usually when it's disclosed on an application and someone is using either a matrix or they have a blanket policy but given how easy it is to locate information now on the internet, for example, there's one agency that has free 50-state criminal background on the internet, it's very easy for an employer to plug that in if they're just looking at someone's application.

COMMISSIONER ISHIMARU: Free?

MS. MOSKOWITZ: Free. And there's been a lot of complaints about the inaccuracy of that service. There was a New York Times article on it that I cited in my statement. So, it's very easy for employers to do that even if they're not paying and often it's a minimal amount. As we said, it could be a little as $20.00 for this background check. So I would just say that I think a lot of employers are informally just checking things out on the internet or in Pennsylvania, as Janet was saying, it's very easy to get the court information.

So while it may be that they're waiting to pay to use the private screening company for the report until after they're ready to make -- or they've made a conditional offer, it's very easy to check out someone's criminal record beforehand, too.

COMMISSIONER ISHIMARU: Thank you very much.

COMMISSIONER GRIFFIN: And Stuart, I would assume actually based on your study, Professor, that it's the lower level service industry people who frankly, you know, where I think a lot of these folks are trying to get jobs, that have these applications that ask this question and you know, that's it. I mean, you check that box and, you know, they just -- I mean, that's -- that, I think is what happens and where the real issues probably are, you know, based on what you're saying about the larger employers.

MS. VANN: Well, and I have to say, too, that I do disagree with the suggestion that small and mid-sized companies are doing a better job by and large than large companies are in considering this information in a responsible way. It may very well be the case that it's a question of education and outreach. Maybe lower level hiring managers or supervisors need to be better educated as to what company policies are, what the law requires and what the EEOC's current enforcement guidance requires with respect to considering this information but I do have to disagree with some of what was said earlier about large employers generally being bad about being responsible --

COMMISSIONER GRIFFIN: Who said that?

[Laughter]

MS. VANN: Okay, I'm --

COMMISSIONER GRIFFIN: No one said that.

MS. VANN: -- reaching a little bit.

CHAIR EARP: Let me ask a question about arrest. Ms. Moskowitz, you would recommend to EEOC that we have a very bright line that says arrest information cannot be used in any employment decision or am I over-stating?

MS. MOSKOWITZ: I mean, we've said we recognize there are some very limited security sensitive positions. For example, in the current arrest guidance there is discussion of, you know, can a police officer give testimony in certain situations if he or she has a certain background, but in general, I do think the bright line is the best, and again, this is what -- almost a majority of the states have laws about this, so you have this conflict where the EEOC policy would allow for more consideration of an arrest than a state law, and given the issues we talked about today with respect to arrest that don't lead to conviction that it's just not definitive proof that someone is engaged in that conduct. And so few employers -- really I have yet to see one that's doing the kind of inquiry about, you know, did the person actually engage in that conduct, according to the current guidance, that it's just unworkable.

CHAIR EARP: And Ms. Vann, you would have us recognize that there are some national security interest and perhaps, stipulate that that is a consideration that the review of an arrest record would be allowable.

MS. VANN: Yes, to the extent that there are obligations that employers may be under to take a look at arrest history, I would ask the Commission to take that into consideration, but again, as a practical matter, as I said before, for many of the reasons Ms. Moskowitz outlined, employers that we deal with typically don't go there with respect to arrest records that do not lead to criminal convictions because it's just too slippery a slope, if you will.

CHAIR EARP: So my final question is, my 16-year old that I told you about earlier assuming I don't kill him myself and get a criminal record, if he applies for a job, a low level sales job, and he has 15 arrests on his record, but no conviction, are you --

COMMISSIONER BARKER: Wait a minute, let me ask you about that. As a 16-year old, doesn't every state have a -- like a youthful offender law where they are not anything that you -- any crimes committed under 18 are not criminal to --

CHAIR EARP: Well, let's change it. He's 22 and he managed to get out of college some kind of way and he's still alive, notwithstanding his past. He has these arrests. They're numerous but there are no convictions.

COMMISSIONER GRIFFIN: His mother is a lawyer.

(Laughter)

CHAIR EARP: Are both of you suggesting that an employer who's contemplating hiring him we should bar depending on the job from looking at the arrests?

MS. VANN: No, that's not what I'm suggesting in all instances. I think your --

CHAIR EARP: This is other than a national --

MS. VANN: -- that depends on a categorical rule. There are many state laws that prohibit employers from looking at that information, but there are also laws I believe that allow you to take certain information into account, the totality of the circumstances. So to the extent that an arrest record, chronic arrest record, might impact upon that person's ability to perform a specific job, I think it may be appropriate for an employer to look at that information, but what I'm saying is that as a practical matter, particularly given the fact that many states, most states that regulate in this area, bar consideration of arrest histories that don't lead to criminal conviction, as a practical matter, employers just aren't regularly doing it.

CHAIR EARP: Okay, it might be that arrest records are so unreliable and issues of racial and other kinds of profiling that on balance, they shouldn't be reviewed. Commissioners, any other questions, comments? I want to thank you, thank you so much to all of our panelists, all of our experts. There being no further business, do I hear a motion to adjourn?

COMMISSIONER GRIFFIN: So moved.

CHAIR EARP: Is there a second?

COMMISSIONER ISHIMARU: Second.

CHAIR EARP: All in favor?

ALL: Aye

CHAIR EARP: The ayes have it. The meeting is adjourned, thank you.

(Whereupon, the above-entitled matter concluded at 1:14 p.m.)