1. Inicio
  2. node
  3. Written Testimony of Juan Cartagena President and General Counsel Latino Justice

Written Testimony of Juan Cartagena President and General Counsel Latino Justice

Meeting of July 26, 2011 – EEOC to Examine Arrest and Conviction Records as a Hiring Barrier

Good afternoon Madam Chair and Commissioners and thank you for inviting LatinoJustice PRLDEF to participate in this important conversation on the use of criminal background checks in employment. I will tailor my remarks towards the effects of these practices on Latino1 communities. In April 2011 I became the President & General Counsel of this important civil rights organization but throughout my career I have addressed employment discrimination issues under Title VII and state and local human rights laws. In the last ten years when I was General Counsel to the Community Service Society I began to address the collateral consequences, including employment consequences, that confront individuals with previous criminal histories. That focus was part of a larger research agenda that I undertook to explore the intersection between mass incarceration and poverty, and equally important for today's meeting, the consequences of mass incarceration on Latino communities. Inasmuch as LatinoJustice PRLDEF has concentrated many of its efforts in the states along the eastern seaboard, this statement will refer primarily to the Latino communities there.

Mass incarceration is a Latino issue.

America's increased reliance on imprisonment has clear impacts on Latino communities and their reintegration into society in ways that the general public and many Latino organizations have yet to fully understand.2 Throughout the country, and especially urban America, mass incarceration is also a Latino issue. The rate of incarceration in the United States is the highest in the world:3 reportedly one out of every 100 adults in the U.S and demographically, one in 194 for Whites, one in 29 for Blacks, and one in 64 Latinos.4 As demonstrated in Table 1 of the Appendix, in June 2009 federal, state and local prison data indicated that among men alone, the rate is 1,822 per 100,000 for Hispanics and 702 per 100,000 for non-Hispanic Whites. For women, the rate is 142 per 100,000 for Hispanics and 91 per 100,000 for non-Hispanic Whites.5 The fact is that data collection of criminal justice statistics is notoriously inconsistent as far as Latinos are concerned, while, simultaneously, the field of criminology research is skewed towards documenting the problems of racial disparities in a black/white binary. For example, data collected by federal agencies including Uniform Crime Reports and the FBI only collect data under the four federally recognized racial categories- white, black, Asian or Pacific Islander, American Indian or Native Alaskan.6 In addition, Maryland and North Carolina fail to document Latino inmates. Moreover, the EEOC has already opined in its previous guidance on this topic that an inquiry regarding the nature and gravity of the criminal conviction compared to the nature of the employment sought or held are relevant factors in employment decisions.7 Accordingly, it is important to note the differences between the types of convictions within Latino communities compared to other groups. Hispanics are twice as likely as whites to be admitted to state prison for a drug related offense.8 While in 2009, a majority of New York state-inmates were incarcerated for violent felonies, the majority of Latinos were incarcerated for drug-related offenses.9 This persists despite Latinos having one of the lowest rates of lifetime illicit drug use at 38.9%, as compared to Whites at 54% and African-Americans at 43.8%.10

Of course, a profile of America's incarcerated population provides information only for one segment in the larger pipeline that leads to the incidence of criminal justice histories within the U.S. domestic labor force. National estimates on the number of Americans with criminal histories records range from 59 million11 to 65 million.12 Close to 30% of the U.S. adult population has a criminal record on file with the states.13 While estimates are that 600,000 to 700,000 prisoners will be released annually in this decade, representing a fraction of the domestic labor force, they nonetheless equal 30% of the annual growth of the labor force.14 This is the labor pool that is directly impacted by criminal record checks. Nationally, 18% of all adults on parole are Latino15 exceeding their share of the national population in the latest census (16.3%). Conversely, 13% of all adults on probation are Latino, which reflects how Latinos are underrepresented among the ranks of persons who avoid prison altogether after conviction. The fact that states like Maryland and Virginia fail to document the number of their Latino residents on parole or probation continues to impede a full analysis of the depth of this problem in those communities.

LatinoJustice PRLDEF has long history of work in the Eastern part of US. Given the disproportionate share of Latinos involved in the criminal justice system in many of those states, problems associated with criminal background checks are likely to be found there.

LatinoJustice PRLDEF (formerly known as the Puerto Rican Legal Defense & Education Fund) is a 39 year old civil rights organization that has a long history of litigating and protecting the rights of Latino workers to be free from discrimination. We have successfully litigated challenges to the entry requirements for the New York Police Department16, challenges to unlawful terminations based on English-Only work rules,17 constitutional challenges under the 14th Amendment's right to travel clause to restore civil service credits to veterans who enlisted from Puerto Rico18 and pending challenges under Title VII to the harassment of Latina immigrant workers19 and the denial of employment by the Census Bureau for persons with arrest records.20 Our work is national but primarily on the east coast including Rhode Island, Massachusetts, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland, Washington, D.C., Virginia, North Carolina, South Carolina, Georgia and Florida.

The national data cited above documenting the number of Americans with criminal records obscures the effects of mass incarceration at the state level where Latino imprisonment is grossly disproportionate to white imprisonment. White incarceration never surpasses the national average of 751 per 100,000 residents but in some of the states highlighted above, specifically Pennsylvania, Connecticut, and Massachusetts the incarceration rates are 1,000 for every 100,000 Latino residents; and in New York the rate for Latinos exceeds the national average.21 Indeed, of the states that have been the special focus of LatinoJustice PRLDEF, New York ranks third in the country in states with the largest Latino residents and fourth in the country for its prison population. Similarly, Florida ranks third in the country in states with the highest prison populations and was fourth among states with the highest Latino population.22

Table 2, attached hereto in the Appendix, reinforces these state-level trends by using Hispanic-to-White incarceration ratios from the Bureau of Justice Statistics of the U.S. Department of Justice. Again, national figures mask the concentrated nature of regional and state incarceration rates within Latino communities. As per Table 2, the overall Hispanic-to-White incarceration ratio in 2005 for state and local prisoners was 1.8 to one. But the national ratio was exceeded in states with much higher Latino incarceration rates: Connecticut with close to 479,000 Latino residents had a ratio of 6.6 to one. Massachusetts with over 628,000 Latinos had a ratio of 6.1 to one. Pennsylvania with over 724,000 Latinos had a ratio of 5.6 to one. New Jersey with one and half million Latinos had a ratio of 3.3 to one. New York with over 3.4 million Latinos had a ratio of 4.5 to one.

Table 3 in the Appendix continues this trend with data on both probation and parole rates. As noted above, of all adults on probation, 9% were Latino and the corresponding share of Latino adults on parole is double that at 18%. While the data is unavailable for all states in Table 2, the results for Connecticut, New York and Pennsylvania are striking: each of them have rates of Latinos on probation and on parole that exceed both national averages for each indicator respectively and exceed the share of the Latino residents in their states. New Jersey and Massachusetts have similar characteristics but only for parole as per Table 3.

Table 4 in the Appendix is a compendium of statutes in each of the states highlighted in this statement which govern the permissible inquiries that can be made of prospective employees in the areas of arrests and convictions. The table covers both employment and occupational licenses and references both public and private sector employers. As noted above in this section, a number of states we have selected have major disparities between Latino and White residents in their presence among the subpopulations that compose the universe of all persons with criminal record histories. Table 3 allows the EEOC to assess how those states permit broad or narrow inquiries into such histories thus placing in context the breadth of criminal background checks that are permissible under state law. For Connecticut, Massachusetts, Pennsylvania, New Jersey and New York the state norms vary considerably. New Jersey provides no limitations on arrest or conviction inquiries and thus allows for the broadest applicability of criminal background checks. At the opposite extreme are Connecticut and New York where the broad public policy of encouraging the hiring of persons with criminal histories is manifest in their state codes. Massachusetts differentiates among the types of crimes committed and Pennsylvania effectively outlaws the use of arrests in employment decisions. I should note here that there is a big difference between sound public policy and law in states like New York and the enforcement of same on a consistent basis. New York may have a relatively compassionate view of reintegrating former prisoners into society: it is "(t)he public policy of this state … to encourage the licensure and employment of persons previously convicted of one or more criminal offenses."23 But from my perspective New York employers and state licensing agencies consistently fail to abide by this state mandate. In short, there is no pattern here between egregious Latino disparities in criminal justice outcomes and lack of state law protections for Latino employment applicants with previous criminal histories. Connecticut and New York have both incredibly skewed Latino incarceration, probation and parole rates while simultaneously providing the legal framework to best challenge discrimination at the workplace under their state laws.

Puerto Rico

Puerto Rico, an unincorporated U.S. territory whose residents are U.S. citizens since 1917,24 is another area of focus for LatinoJustice PRLDEF and merits inclusion in this discussion along the lines noted above. As the EEOC is well aware, under Title VII the term "State" is defined as including Puerto Rico.25 Indeed, the EEOC acknowledged the need for Title VII enforcement on the island of Puerto Rico when it opened its office there in 2001; previously, EEOC filings were administered in the New York field office.26

A number of law enforcement and mass incarceration indicia are also unique to Puerto Rico. The Puerto Rico Police Department has about 17,000 officers making it second only to the New York City police force.27 Moreover, the island police force often refers to Puerto Rico as America's "third border" given its strategic location in the international drug trade.28 Indeed, since 1994 Puerto Rico has been classified as a High Intensity Drug Trafficking Area by the Office of National Drug Control Policy.29

In May 2011 a total of 11,629 persons were in the custody of the Puerto Rico Department of Corrections, either sentenced to imprisonment or detained awaiting trial.30 In fact, the rate of incarceration in Puerto Rico is significantly high: 303 per 100,000.31 Additionally, persons on probation and parole now number over 10,000.32 The total population in Puerto Rico as per the 2010 census numbers 3.7 million.

Persons in Puerto Rico who were previously incarcerated do receive anti-discrimination protection in employment decisions under Puerto Rico law as per the concurring opinion issued by the Puerto Rico Supreme Court in Rosario Díaz v. Toyota de P.R., 116 D.P.R.1 (2005). The lynchpin in this analysis is the broad prohibited bases of discrimination in Article II of the Charter of Rights in the Puerto Rico constitution which lists "social status" alongside race, color, sex, etc. The decision in Rosario Díaz v. Toyota de P.R equates persons who were formerly incarcerated with those who merit protection from discrimination under law because of their "social condition." That decision is likely to be codified by the legislature. On June 24, 2011 the Puerto Rico House of Representatives approved Senate Bill 1730 which, if signed by the Governor, will provide individuals with past criminal records the same protection against employment discrimination as provided to other covered groups under Puerto Rico's General Antidiscrimination in Employment Act, Commonwealth Law No. 100 of June 30, 1959, as amended. The bill recognizes that ex-convicts carry a social stigma that is similar to other protected social conditions. It will prohibit employers from creating blanket exclusions of anyone who has been convicted of a crime.  Instead, it mandates an individualized assessment, which includes considerations of factors including: (1) the nature and seriousness of the offense (2) the relationship between the offense, the employment sought and the duties and responsibilities of the position; (3) applicant's rehabilitation, including any information the applicant or a third party may provide regarding same; (4) the circumstances under which the offense took place, including any mitigating conditions; (5) the applicant's age when the offense occurred; (6) the time transpired between the offense was committed and the job application or reinstatement request; and (7) the employer's legitimate interest in protecting its property, safety and well-being, as well as that of its employees and the general public.

Finally, migration patterns from Puerto Rico to the United States are another factor to consider in assessing the effects of criminal background checks on Latinos in the states discussed in this report. In short, the criminal history characteristics of Puerto Ricans living in Puerto Rico are a factor in Puerto Rican stateside communities because of the migration patterns that these communities exhibit over time. It is estimated that "circular migration" patterns within these communities characterized the relocation throughout the 1980s and 1990s as mass emigration from Puerto Rico resumed in large numbers while return migration continued unabated.33 Now, however, reports indicate that between 2005 and 2009 a net out-migration from Puerto Rico to the United States totaled 144,000 people, peaking in 2006 at the start of the island's official economic recession.34 Moreover, Puerto Ricans leaving the island have higher education attainment levels than those remaining and those returning35 – all of which speaks to possible entry into management, professional, sales and office occupations for the ones who seek employment stateside.

Case law sources on the effects of criminal arrest or conviction histories and discrimination on Latino employees and applicants

Title VII litigation is another potential source for documentation on the effects of arrest and/or conviction inquiries on Latino applicants and employees. Unfortunately, there are few reported decisions in the area, and fewer still that assert claims on behalf of Latino clients or organizations.

EEOC v. Carolina Freight Carriers Corp., 723 F.Supp. 734 (S.D.Fla.1989) is one of the few exceptions and perhaps the only reported opinion that analyzes the impact of these policies on Latino communities. In that case the EEOC sued on behalf of a Puerto Rican resident of Florida, Mr. Francisco Rios, who was denied employment because of his larceny conviction and brought a Title VII claim against the trucking company for its policy against hiring any driver who had been convicted of any felony or misdemeanor related to theft. After analyzing criminal justice data, the District Court did find in the relevant labor market of South Florida (Dade County in particular and to some extent Broward County) that the company's policy "adversely impacts Hispanics at a statistically significantly rate exceeding that of non-Hispanics."36 Indeed, the court accepted evidence that in Dade County in particular, from 1984 through 1988, there were statistically significant disparities between Hispanics and Non-Hispanic Whites in both felony convictions resulting in incarceration and in larceny and burglary convictions resulting in incarceration.37 However, the court then concluded that there was insufficient proof of any disparity at the trucking terminal attributable to the defendant's conviction policy.38 More importantly, the court accepted the defendant's bare-bones assertion that because drivers were in positions of trust and the opportunity to steal cargo was so great, that these jobs could not be entrusted to someone with theft-related or other criminal convictions. Indeed, the court was operating under the now-discredited Wards Cove Packing Co. Inc., v. Atonio, 490 U.S. 642 (1989) definition of business necessity which was eventually abrogated by Congress.39 The decision in Carolina Freight Carriers is not without other problems for organizations like LatinoJustice PRLDEF that now recognize the collateral consequences of criminal convictions that stem from an unfair criminal justice system. Judge Jose Gonzalez stridently noted "If Hispanics do not wish to be discriminated against because they have been convicted of theft then, they should stop stealing."40 Nonetheless, the case does underscore the evidence of disparate impact on Latinos in that section of Florida of employment policies that condition employment on the absence of criminal histories and its analysis of the evidence required to satisfy the business necessity requirement has since been criticized by other others.41

The only other reported decision that contains any discussion about the disparate effects of criminal arrest and/or conviction inquiries on Latino communities is Ganzy v. Sun Chemical Corp., No. CV 06-3424 (E.D.N.Y. 2008) and the court never reached those claims. In Ganzy a Latino plaintiff Mr. Alfredo Vidot had his case consolidated with the lead plaintiff. Both plaintiffs asserted a disparate treatment claim regarding the employer's conviction policy, but Mr. Ganzy, an African American, also asserted a disparate impact claim. The court rejected the Latino employee's Title VII disparate treatment claim when he was fired after five years because his employer discovered that he had lied about his conviction record on his employment application. The court also held that the disparate impact theory under Title VII was not applicable because they were fired because they lied on their application, not because they had been convicted of a crime.42

Finally, there are at least two pending Title VII claims in federal court that challenge the exclusion from employment of Latinos and other persons with criminal histories. E.E.O.C. v. Freeman, No. 8:09-cv-02573-RWT (D.Md. 2009) is a nationwide class action Title VII complaint against Freeman, an exhibition and corporate events marketing company, for their practice of unlawful discrimination by rejecting applicants based on credit history and criminal history. The EEOC alleged that Freeman engaged in an on-going pattern and practice of race, national origin, and sex discrimination against black, Hispanic and male job applicants. Specifically, the EEOC found that use of credit history and criminal history as hiring criteria had a disparate impact on these classes of persons, were not job-related, and that there were more appropriate, less discriminatory alternative selection procedures available. Finally, Johnson v. Locke, No. 10-cv-3105 (S.D.N.Y. Mar. 14, 2011) where LatinoJustice is one of the co-counsel on the case, a Title VII class action suit was filed challenging an unlawfully discriminatory U.S. Census Bureau hiring policy. In hiring for the 2010 Census enumerator positions, the U.S. Census Bureau requires that applicants with arrest records provide "official court documentation" of any arrest within 30 days to remain eligible for employment. This arbitrary barrier to employment has a disparate impact on Black, Latino, and Native American applicants because the arrest and conviction records for those populations disproportionately exceed those of whites. The public interest will be served by removing this discriminatory barrier because the Census Bureau will be able to expand its hiring base into historically-undercounted communities.

Labor Segmentation of Latinos

Census data for employed residents of the United States can reveal the extent of labor segmentation of various groups in the country. Detailed data of the occupational distribution of all persons above the age of sixteen exists for 2010 along with demographic differences between Latino workers and others.43 When combined with the trends in employment practices that have seen a proliferation of the criminal records check industry, additional documentation can arise that can inform the EEOC of the impact of such background checks on Latino communities.

The overall U.S. workforce, sixteen years of age and above in 2010, numbering over 139 million, is distributed among the following occupational categories:

Management, professional, and related occupations: 37%
Service occupations: 18%
Sales and office occupations : 24%
Natural resources, construction, and maintenance occupations: 9%
Production, transportation, and material moving occupations: 12%

Latino workers, close to 20 million, on the other hand, have a different distribution among these occupational categories:44

Management, professional, and related occupations: 19%
Service occupations: 26%
Sales and office occupations : 21%
Natural resources, construction, and maintenance occupations: 16%
Production, transportation, and material moving occupations: 17%

In essence, Latino workers are much more evenly distributed than the population as a whole. Whereas the total U.S. population is clearly skewed towards management and professional titles, Latinos have had less of an inroad into those titles but a much higher segmentation among the blue-collar occupations of construction, maintenance, production and transportation.

As the EEOC is aware the criminal background checks industry has exploded in the last ten years with the National Law Employment Project reporting that an incredible 92% of mostly large employers now run background checks on all or some of their employees.45

This begs the question: are Latinos likely to be subject to disproportionately more criminal background checks than the general population because of their distribution with occupational categories? A full analysis of the source data for the National Employment Law Project report is unavailable to us at this time but anecdotally, it appears that Latino workers are not more likely to be subject to these background checks than the White population – at least to the extent that marginalized labor occupations (the construction, maintenance, and production industries noted above) have yet to fully immerse themselves in the employment practices of larger employers. Nonetheless, forty percent of the close to 20 million Latino workers in the country is found in the management, professional, sales and office ranks.

More importantly, however, is the conclusion that Latinos are clearly disproportionately found within the ranks of persons with previous criminal records. Thus, given their distribution among occupations that are subject to background checks at present, a significant portion of Latino workers are subject to the inaccuracies, incompleteness and vagaries that that industry has. In short, they are more likely to subject to adverse effects of criminal background checks. It is to that area that we now turn.

Employment is a key determinant in reducing recidivism so that the consequences of a mistake-prone commercial criminal background check industry has a greater adverse effect on Latino job applicants and employees.

The EEOC, indeed the nation, is acutely aware of the importance of gainful employment to our country's families and neighborhoods. In urban America, however, where the consequences of harsh criminal justice policies and incarceration are exacerbated, the importance of job creation and job retention takes on added significance. In many urban areas the effects of the economic recession are still real and evidence that uniquely counterintuitive label, "a jobless economic recovery" that describes Wall Street today. Indeed, in New York City in 2004, 44% of Latino men were jobless46 – i.e., completely out of the labor market – and in 2009 the labor force participation rate for Latino men in the city lagged behind whites and Asians.47 When you factor in persons with previous criminal histories, the situation is dire. Most reports indicate much lower employment rates for former prisoners than the general population with some estimates marking the differential at 15 to 25 %.48 In this regard, the positive impact of employment on recidivism cannot be overstated. Employment reduces recidivism "because workers are likely to experience close and frequent contact with conventional others and because the informal social controls of the workplace encourage conformity."49 In addition, holding the same job for over a year is can predictably lower recidivism.50 For example, one report out of New York City indicates that while one-fifth to one-third of persons who initially enter prison were unemployed at the time of entry, 89% of persons who violated the terms of their parole, and hence re-incarcerated, were unemployed at the time of the violation.51

This data is illustrative. But the real stories of persons with prior conviction histories as they integrate themselves in the job market provide a compelling case for the protection of every job opportunity. I have witnessed these accounts at the Community Service Society's monthly Reentry Roundtable in New York City and at other meetings in the metropolitan area. Persons on parole, for example, take their jobs seriously – "when most people lose a job, they lose a job, when I lose a job I could lose my liberty and be back in prison" – is a common theme in these personal accounts.

Research in the New York City job market using white, black and Latino testers, also reveals ingrained stereotypes and bias against Latino and black applicants compared to white applicants who purported to have criminal histories. In one sample, when controlling just for race, testers with similar qualifications and similar criminal histories, revealed that white testers received call backs and job offers at rates higher than Latinos or blacks.52 In a second sample the same researchers found that white testers with purported criminal histories for a recent felony received similarly positive employer responses than Latino or black testers without criminal histories: "The comparison of a white felon with black and Latino applicants with clean backgrounds provides a vivid calibration of the effects of race on hiring decisions. While ex-offenders are disadvantaged in the labor market relative to applicants with no criminal background, the stigma of a felony conviction appears to be no greater than that of minority status… these findings suggest that New York employers view minority applicants as essentially equivalent to whites just out of prison."53

In this context the problems that the commercial criminal background check industry generates through faulty or incomplete reporting have a particularly adverse effect on Latino applicants and employees. The 2006 U.S. Department of Justice report on the criminal background check industry is illustrative on many of these points.54 Many of the problems associated with the industry stem from its reliance on name-based searches of court records, corrections records, and state criminal history record repositories, often through commercial databases that compile these records.55 As noted by the U.S. Attorney General, the private sector, and parts of the public sector, does not have direct access to the FBI fingerprint-based records search capacity. Accordingly, the name-based searches that dominate the practice has the drawback of producing erroneous reports by listing false positives (incorrectly tying a person's name to a criminal record) and false negatives (missing a criminal record because of a false or inaccurate name). The error rate in these systems can translate into significant numbers of individuals who may be denied employment opportunities. For example, one report extrapolated the findings in Florida of a task force consisting of the Bureau of Justice Statistics, the Florida Department of Law Enforcement, the U.S. Department of Housing and Urban Development and the FBI to estimate that if Florida's false positive rates were to apply to the nation-wide fingerprint-based checks of the FBI conducted in 1997, 346,000 false positives would have resulted.56 Adding to the list of false positives the National Employment Law Project reports that commercially prepared background checks are consistently inaccurate with multiple reports of the same incidents and uncorrected identity theft problems as well.57 It is important to note here that the fingerprint-based FBI database is no panacea as well. The U.S. Attorney General reported that while comprehensive, the FBI database was "missing final disposition information for approximately 50 percent of its records."58

In addition to these problems, in my previous work on these matters I have seen commercial records reports to include information about unresolved arrests in reports issued to employers in New York without any corresponding qualification or warning that New York law prohibits employment decisions to be based on arrests that do not lead to convictions. The regulation of these commercial databases under the Fair Credit Reporting Act59 clearly does not go far enough to ensure that the necessary and consistent state law protections60 that exist in the states highlighted in this statement are complied with – all to the detriment of Latino applicants and employees.

In short, the entrenched joblessness in urban America along with the relationship between unemployment and recidivism, coupled with the racial/ethnic characteristics of the populations most affected by criminal background checks, requires a comprehensive analysis by the EEOC of the unlawful discriminatory aspects of the industry.

The lessons of El v. SEPTA

El v. Southeastern Pennsylvania Transp. Authority (SEPTA), 479 F.3d 232 (3d Cir. 2007) is the most recent decision providing guidance on Title VII and the business necessity defense for criminal conviction bars to employment. In that case, an African-American paratransit driver was terminated based on a 40 year old homicide conviction because his employer (SEPTA) had a policy not to employ drivers or attendants with any convictions of violence or moral turpitude. Although SEPTA's policy excluded applicants with other convictions as well, this case specifically dealt with violent criminal convictions, regardless of how long ago it was committed. The District Court granted summary judgment for the defendants finding that they had submitted sufficient evidence to prove that its policy was justified by business necessity and that the plaintiff did not submit sufficient evidence of an alternative policy that would accomplish SEPTA's legitimate goal of public safety. The Third Circuit held that a reasonable juror could find that SEPTA's policy of not hiring drivers or attendants with convictions for violent crimes met the "business necessity" defense standard.61 Furthermore, it clarified the business necessity defense by establishing that under, Title VII, criminal record policies must "accurately distinguish between applicants that pose an unacceptable level of risk and those that do not."62 While the "business necessity" had traditionally been developed in the context of aptitude tests,63 El v. SEPTA incorporates this standard to the criminal conviction bar for employment. No longer will courts accept the bare-bones reasoning that hiring a person with a criminal history is presumptively risky. Employers must show that "a discriminatory hiring policy accurately- but not perfectly-ascertains an applicant's ability to perform successfully the job in question." 64 In applying this test to a bar for convictions for violent crime of any age for paratransit drivers, the court held that the defendants had established through its expert witnesses that quantitatively, former violent criminals, even if crime free for many years are at least somewhat more likely than the general population to commit a future violent act. While the court held that the defendants had established the business necessity for this criminal conviction bar, they did so only because of the plaintiff did not present any expert witnesses to rebut the expert testimony on business necessity nor did he depose the defendant's experts to establish any issues of credibility. Accordingly, the holding, while important, is a narrow one.

El v. SEPTA provides the EEOC with an opportunity to reissue its guidance on criminal convictions and arrests so as to support its clear mandate that employers must substantiate and document their reasoning behind criminal conviction bars to employment in order to avoid a Title VII violation in the face of evidence of discriminatory impact. Based on the potential impact of such employment bars on Latino communities as noted herein, I submit there are at least three areas where and EEOC guidance would assist in addressing the unlawful discriminatory impact of such policies:

  1. Policies that affirmatively address the age of the conviction;
  2. Policies that affirmatively address the nature of the conviction;
  3. Policies that allow reliance upon arrests that do not lead to convictions.

In 2010 I appeared, along with Paul Keefe, as counsel to amici in Acosta v. New York City Department of Education (New York County Clerk Index No. 400475/o7), a case pending before the New York Court of Appeals that challenges under state law the dismissal of a Latina, Madeline Acosta, as an administrative assistant in a school based on her only conviction, 16 years earlier, for robbery. The relevant portions of that amicus brief are set out below. The passage65 sets out the relevant data and reports that address recidivism concerns regarding both the age of the conviction and the nature or type of the conviction. It is important to reiterate here that Latinos who are convicted are more likely to be convicted for drug related crimes than violent crimes.66

"The majority of people who commit one crime "do not go on to lead lives of crime, but indeed age out of, or otherwise desist, from criminal activity." Megan Kurlychek et al., Enduring Risk? Old Criminal Records and Short-Term Predictions of Criminal Involvement, 53 CRIME & DELINQ. 64, 70 (2007). Among criminologists, there are many factors that predict "desistance" or "redemption"—the end of criminal activity and the reintegration into society. Chief among these is the passage of time, but other indicators include the transformation of personal identity that leads to adulthood: employment at an adequate income, a home, a spouse, children, and adult friends. John H. Laub & Robert J. Sampson, Understanding Desistance from Crime, 28 CRIME & JUST. 1, 18 (2001). These factors are consistent across types of crime and age when the crime occurred; the only difference between men and women is that women are not only less likely to commit crime, but less likely to recidivate. Id. at 24-25; see also Christopher Uggen & Candace Kruttschnitt, Crime in the Breaking: Gender Differences in Desistance, 32 L. & SOC'Y REV. 401, 421 (1998).

"Criminologists generally agree that people with criminal records desist from criminal behavior as they age. Christopher Uggen, Work as a Turning Point in the Life Course of Criminals: A Duration Model of Age, Employment, and Recidivism, 67 AM. SOC. REV. 529, 542 (2000); see also Peter B. Hoffman & James L. Beck, Burnout—Age at Release from Prison and Recidivism, 12 J. CRIM. JUST. 617, 621 (1984). While recent criminal convictions are relevant for predicting recidivism, if a person with a criminal record remains crime-free for approximately 7 years, her risk of committing a new offense is similar to a person without any record at all. Id. at 83; Shawn D. Bushway & Gary Sweeten, Abolish Lifetime Bans for Ex-Felons, 6 CRIMINOLOGY & PUB. POL'Y 697, 697 (2007). Multiple studies have shown that the risk of recidivism after release from incarceration "peaks fairly quickly and then diminishes considerably with the passage of time." Kurlycheck et al., Scarlet Letters and Recidivism: Does an Old Criminal Record Predict Future Offending?, 5 CRIMINOLOGY & PUB. POL'Y 483, 488 (2006). Specifically, the peak occurs 6 to 10 months after release; at 20 months, recidivism drops to half of the peak level; at 40 months, the level is halved again. Id.; see also, e.g., David F. Greenberg, Recidivism as Radioactive Decay, 15 J. RES. CRIME & DELINQ. 124 (1978); Pamela K. Lattimore & Joanna R. Baker, The Impact of Recidivism and Capacity on Prison Populations, 8 J. QUANTITATIVE CRIMINOLOGY 189 (1992).

"'(I)n most societies, crime rates rise in the early teen years, peak during the mid- to late teens, and decline thereafter.' Uggen, supra at 530. By their mid-20s, most individuals with juvenile convictions have a risk of recidivism that is indistinguishable from people with no criminal record, according to a 2007 study of 610 individuals from young adulthood through age 32. Kurlychek et al., Enduring Risk?, supra at 75. In fact, research shows most criminal careers are short—no more than 5 years for people who commit crimes in their youth. Laub & Sampson, Understanding Desistence, supra at 17 see also generally MICHAEL E. EZELL & LAWRENCE E. COHEN, DESISTING FROM CRIME (Oxford Univ. Press 2005) (finding declining arrests as serious youthful offenders reach their mid-20s).

"About 7 years after being convicted, [a person] has a risk of recidivism no greater than a person from the general population. Alfred Blumstein & Kiminori Nakamura, Redemption in the Presence of Widespread Criminal Background Checks, 47 CRIMINOLOGY 327, 339-40 (2009). An older study, following five hundred men from age 7 to age 70, found that someone arrested for a property crime at age 17 desisted at around 26.2 years of age. Robert J. Sampson & John H. Laub, Life-Course Desisters? Trajectories of Crime Among Delinquent Boys Followed to Age 70, 41 CRIMINOLOGY 555, 567 (2003). After age 30, the effect of youthful arrests is insignificant, making recent experiences—securing a job; starting a family—the best predictors of desistance. Laub & Sampson, Understanding Desistence, supra at 14. Therefore, predicting the risk of future crime based simply on a criminal record is "quite inadequate." Kurlychek et al., Enduring Risk?, supra at 82."

The academic literature cited above should lead the EEOC to fashion a guidance that presumptively disfavors conviction bars for convictions more than seven years old, in most cases. Similarly, the literature should convince the EEOC to issue a guidance that presumptively disfavors conviction bars for non-violent crimes, in most cases.

EL v. SEPTA should lead to a similarly empirical and documentable foundation for any employment policy that prohibits offers of employment to persons with an arrest history alone. The current EEOC guidance67 on arrest records recognizes that the use of arrests alone as an absolute bar would have a disproportionate impact on protected classes. That is clearly the case for Latinos in places like New York City where they are routinely arrested for minor infractions like trespass or possession of an open container of alcohol, or possession of marijuana (in a state that decriminalized marijuana possession in small quantities, decades ago).68 The EEOC guidance on arrest records, however, goes further to sanction an inquiry by the employer into the circumstances of the arrest that may lead to information about suitability for employment. I submit that such an inquiry rarely, if ever, occurs and the arrest alone in places that permit such inquiries, like New Jersey, will be used a pretext for denial of employment that otherwise would be unlawful.

EEOC Guidance Recommendations:

For all the reasons stated above, the EEOC should reissue both of its guidance policies regarding the use of conviction and arrest records. Specifically the EEOC should consider:

  1. Engaging in further analysis of the impact of criminal background checks and conviction/ arrest bars in the Latino community. As noted above, many of the Latino communities LatinoJustice PRLDEF is currently engaging in the states highlighted herein have disproportionate contact, arrest and convictions within the criminal justice system. The EEOC's arrest guidance appears to rely upon data from the Southwest prison populations to demonstrate adverse impact.69 Such data does not fully document the problem for Latino communities as noted herein and should be updated. Moreover the EEOC should exhort federal criminal justice agencies to conduct full analyses of prison, probation and parole populations so as to accurately capture the proportion of Latinos in each subpopulation.
  2. Restructuring its arrest record guidance along the lines suggested above, i.e., a bright line rule that the use of arrest records alone is presumptively illegal under Title VII. In short, the EEOC should follow the lead of Connecticut, New York and Massachusetts in this regard.
  3. Incorporating the direction forged by El v. SEPTA to require all employers to ensure that the use of conviction bars can only be sanctioned by a business necessity analysis that is carefully crafted and that empirically documents each element of the policy or practice so as to eliminate unlawful disparate impact.
  4. Reaffirming the current three-part elements that assess the use of convictions in employment decisions. The EEOC factors regarding the age of the offense, the nature and gravity of the offense and the nature of the job sought are reasonable standards for employers to use. They allow the proper balancing between the right to be free from unlawful discrimination, and the right of employers to secure a safe working environment. But the data cited herein also speaks to the need of the EEOC going further. It should consider and include in the EEOC guidance on conviction records that conviction bars for convictions more than seven years old, are presumptively disfavored in most cases. Similarly, the EEOC should issue a guidance that disfavors conviction bars for non-violent crimes, in most cases. The same guidance should explore the data on conviction for minor offenses (misdemeanors, violations, infractions) and their relevance to the predictability for threats to public safety or fitness to perform the job.
  5. Consider including in the EEOC guidance on convictions limits on the use of criminal background checks to the final stages of the employment process so as to eliminate the use of arrest or conviction histories as a pretext to unlawful discrimination under Title VII. The procedures set forth under Connecticut law are illustrative in this regard (see Table II in the Appendix). Under that law, an employer cannot inquire about a prospective employee's conviction history until they are otherwise deemed qualified for the job. Massachusetts also follows this path. Such a construct would eliminate the practices of unscrupulous employers who would mask their intent to violate Title VII under the pretext of relying on information gleaned from a criminal background check. It would also strengthen our workforce by ensuring that the most qualified applicants remain eligible for employment.


Criminal record checks and the use of arrest and/or conviction bars in employment, like felon disfranchisement laws have one critically important feature in common: both policies place inordinate faith on the bona fides of our broken criminal justice system. In the election context, it is our county board of elections and their equivalents who defer to the racially skewed outcomes of our criminal justice system. They defer on the most basic of questions: Who is eligible, or no longer eligible, to vote? Can the most cherished right of citizenship in a democracy, the right to vote, be terminated or suspended upon conviction of a crime even if citizenship remains intact? In the employment context what is centrally at risk is the right to be free from discrimination and to be judged on the merits. Yet without a careful scrutiny of how criminal justice racial/ethnic disparities infect employment decisions, the EEOC, charged with leading the enforcement of Title VII in the country, risks conditioning the enjoyment of this right on a false premise – that law enforcement and criminal justice systems produce racially and ethnically neutral outcomes. Without new EEOC guidelines that address the proliferation of mass incarceration and the concomitant proliferation of criminal record databases, the use of the pretext of a criminal history may shut out a sizeable part of Black and Latino employees – and all without a clear balancing of the rights of employers to have a safe environment and right of employees and applicants to be judged on the merits.

Everything that I have highlighted today before this Commission applies double to African-Americans in the U.S. They are clearly bearing the brunt of the worst aspects of the criminal justice system and its aftermath. The reforms and recommendations I highlight will address those disparities as well. But I have focused my remarks on the Latino community which is now the country's largest minority group – and not just because of that demographic phenomenon. Instead, I respectfully present these observations because the ramifications of the criminal justice system upon the Latino community have rarely received attention from our government entities. Given the inconsistency of reliable data we still do not have a comprehensive portrait of Latinos under criminal justice supervision. In that vein, I thank the EEOC for allowing me to share my viewpoints on this important discourse.

Appendix 1-3

Appendix 4


1 "Latino" and "Hispanic" are used interchangeably throughout this statement.

2 Cartagena, Juan. 2009. Lost Votes, Lost Bodies, Lost Jobs: The Effects of Mass Incarceration on Latino Civic Engagement in Behind Bars: Latino/as and Prison in the United States, Oboler, Suzanne, Ed. Palgrave Macmillan (noting the effects of incarceration for Latinos on disfranchisement, census counts and employment) (hereafter "Cartagena").

3 The rate of 751 per 100,000 in the U.S. surpasses that of every other country in the world. International Centre for Prison Studies. 2008. World prison brief. (accessed March 2008).

4 Pew Center on the States. One in 100: Behind Bars in America 2008. (accessed 19 January 2009).

5 U.S. Department of Justice, Bureau of Justice Statistics, "Prison Inmates at Mid-Year 2009- Statistical Tables." p. 2 (June 2010)

6 Morin, Jose Luis. 2009. Latino/as and U.S. Prisons. n.1, in Behind Bars: Latino/as and Prison in the United States, Oboler, Suzanne, Ed. Palgrave Macmillan.

7 U.S. Equal Employment Opportunity Commission, Notice No. N-915, EEOC Policy Statement on Issue of Conviction Records under Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. § 2000e et seq. (1982), at 1 (February 4, 1987), available at http:// (accessed July 2011) (hereafter "EEOC Convictions Guidance").

8 U.S. Department of Justice, Bureau of Justice Statistics, "Truth in Sentencing in State Prisons 1999", p.6 (January 1999).

9 New York State Department of Correctional Services, Office of Public Information, Incarceration Statistics-2011.

10 Office of National Drug Control Policy. 2003. Minorities and Drugs. Washington, D.C.: Office of National Drug Control Policy.

11 Winston, Craig. 2005. The National Crime Information Center, A Review and Evaluation. National Association of Professional Background Screeners.

12 Rodriguez, Michelle Natividad & Emsellem, Maurice. 2011. 65 Million "Need Not Apply" The Case for Reforming Criminal Background Checks for Employment. National Law Employment Project Report.

13 U.S. Dept. of Justice Office of the Attorney General, The Attorney General's Report on Criminal History Background Checks, at 3 (June 2006).

14Freeman, Richard. 2003. Can we close the revolving door? Recidivism vs. employment of ex-offenders in the U.S. The Urban Reentry Roundtable Discussion Paper, May 19. New York University

15 U.S. Department of Justice, Bureau of Justice Statistics, "Probation and Parole in the United States, 2009", p.26 (June 2010).

16 Guardians Ass'n v. Civil Service Comm'n of City of New York, 463 U.S. 582 (1983)

17 LaVigne, Puente & Berroa, EEOC charge.

18Soto-Lopez v. New York City Civil Service Comm'n, 840 F.2d 162 (2d Cir. 1988)

19 Marilyn Gonzalez, et al v. Suffolk Laundry Services, Inc., EEOC Nos. 520-2011-01130- 520-2011-01136

20 Johnson v. Locke, No. 10-cv-3105, (S.D.N.Y. Mar. 14, 2011)

21 Cartagena, supra, pp. 134-135. The chapter also notes that Idaho, Arizona, New Hampshire and Colorado also have Latino incarceration rates in excess of 1,000 per 100,000 and that North Dakota, Montana, Utah, Texas, Oklahoma, California, Iowa, Kentucky and Nebraska have rates for Latinos that, like New York, exceed the national average.

22 Id. at 139-140. With the issuance of 2010 census data, Florida now ranks third in the country among state with high Latino populations.

23 N.Y. EXEC. LAW §753(1)(a)(McKinney 2011) Indeed, violating this provision is an "unlawful discriminatory practice" under state and city Human Rights laws. N.Y. EXEC. LAW § 296(15,(16) (McKinney 2011); N.Y. CITY ADMIN. CODE § 8-107(10),(11) (2010).

24 In passing the Jones Act of 1917, Congress declared Puerto Ricans citizens of the United States. 8 U.S.C. § 1402.

25 42 U.S.C. § 2000e(i) (2010).

26 "The comparative shortage of job opportunities in Puerto Rico and the U.S. Virgin Islands compared to the U.S. mainland further bolsters the need for an EEOC presence there. A recent study showed that Puerto Rico has a 14% unemployment in a work force of 1.3 million. This situation heightens the importance of fair and non-discriminatory access to the regions limited employment opportunities." "EEOC to Open Area Office in Puerto Rico" (accessed July 2011). Unemployment rates in Puerto Rico now reach 17% in a workforce of 1.27 million. The Department of Labor and Human Resources. Employment and Unemployment Puerto Rico, p.2 (March 2011)

27 Comments by José Figueroa Sancha, Superintendent of Police, San Juan, Puerto Rico, 3 May 2011 in a meeting with the author as part of a delegation sponsored by the American Civil Liberties Union to investigate police abuse practices in Puerto Rico. Notes on file with author. In 2007 the police force was even higher at 18,262, (accessed 20 July 2011).

28 Comments by José Figueroa Sancha, Superintendent of Police, San Juan, Puerto Rico, 3 May 2011 in a meeting with the author as part of a delegation sponsored by the American Civil Liberties Union to investigate police abuse practices in Puerto Rico. Notes on file with author.

29 Office of National Drug Control Policy, (accessed 20 July 2011).

30 May 2011 population files from the Puerto Rico Department of Corrections and Rehabilitation (Oficina de Desarrollo Programático de la Secretaria Auxiliar en Administración y Gerencia del Departamento de Corrección y Rehabilitación de P.R., Informe Diario de la Población Correccional Promedio del Mes, Mayo 2011) on file with author and available at (accessed 27 July 2010).

31 International Centre for Prison Studies. World prison brief. (accessed July 2011).

32 Communication from the Puerto Rico Department of Corrections and Rehabilitation to Natalia Rivera Smith, (LatinoJustice PRLDEF) 22 July 2011. On file with author.

33 At that time the migration was characterized as "transient and bidirectional" as opposed to "irrevocable and unilateral." Duany, Jorge. 2002. Puerto Rican Nation on the Move: Identities on the Island and in the United States, p. 211. University of North Carolina Press, Chapel Hill.

34 Ferré, James, "Study confirms 'brain drain' from PR," Caribbean Business 28 February 2011, (citing a report of the Puerto Rico Statistics Institute)(accessed 20 July 2011).

35 Id.

36 Carolina Freight Carriers Corp., 723 F.Supp at 751.

37 Id. at 745-746. For example, the rate of incarceration for all felonies in Dade County in 1988 per 100,000 was 8.9 for non-Hispanic Whites and 14.5 for Hispanics; similarly in 1988, the rate of incarceration for larcenies and burglaries was 5.3 for non-Hispanic Whites and 7.2 for Hispanics.

38 Id.

39 42 U.S.C. § 2000e-2(k).

40 Carolina Freight Carriers Corp., 723 F.Supp at 753. He then followed that dicta by noting that minorities would somehow be better off if courts refused to find employers liable under Title VII for refusing to hire persons with felony histories even in the face of disparate impact, because to "hold otherwise is to stigmatize minorities by saying, in effect, your group is not as honest as other groups." Id.

41El v. Southeastern Pennsylvania Transp. Authority (SEPTA), 479 F.3d 232 244 n.11(3d Cir. 2007)

42Ganzy v. Sun Chemical Corp., No. CV 06-3424, slip op. at 10 (E.D.N.Y. 2008).

43 U.S. Department of Labor, Bureau of Labor Statistics, "Employed Persons by Occupation, Race, Hispanic or Latino Ethnicity, and sex", Table 10 (2011).

44 Numbers may tally less than 100 due to rounding.

45 Rodriguez, Michelle Natividad & Emsellem, Maurice. 2011. 65 Million "Need Not Apply" The Case for Reforming Criminal Background Checks for Employment, p. 27, n. 1. National Law Employment Project Report.

46 Levitan, Mark. 2004. A crisis of black male employment: Unemployment and joblessness in New York City 2003, Community Service Society. New York.

47 Holder, Michelle. 2010. Unemployment in New York City during the recession and early recovery: Young black men hit the hardest, Community Service Society, p. 7. New York.

48 Western B. Kling, J.R., & Weinman, D.F. 2001. Crime & Delinquency: The labor market consequences of incarceration. Paper #450. Princeton, N.J.: Princeton University Department of Economics – Industrial Relations Section.

49 Uggen, Christopher. 2000. Work as a Turning Point in the Life Course of Criminals: A Duration Model of Age, Emloyment, and Recidivism. 67 AM. SOC. REV. 529.

50 Laub, John and Sampson, Robert. 2001. Understanding Desistance from Crime. 28 CRIME & JUST. 1, 18.

51 New York City Bar Association Task Force on Employment Opportunities for the Previously Incarcerated. 2008. Legal employers taking the lead: Enhancing employment opportunities for the previously incarcerated. New York: New York City Bar Association.

52 Pager, Devah, Bruce Western, and Bart Bonikowski, Discrimination in a Low Wage Labor Market: A Field Experiment. American Sociological Review 74, 784 (2009).

53 Id. at 785-786.

54 U.S. Dept. of Justice Office of the Attorney General, The Attorney General's Report on Criminal History Background Checks (June 2006).

55 Id. pp. 38-40.

56 Winston, Craig. 2005. The National Crime Information Center, A Review and Evaluation, pp. 11-12. National Association of Professional Background Screeners

57 Rodriguez, Michelle Natividad & Emsellem, Maurice. 2011. 65 Million "Need Not Apply" The Case for Reforming Criminal Background Checks for Employment, p. 28, n. 22. National Law Employment Project Report.

58 U.S. Dept. of Justice Office of the Attorney General, The Attorney General's Report on Criminal History Background Checks (June 2006), p. 3.

59 Fair Credit Reporting Act, 15 U.S.C. § 1681, et seq.

60 The U.S. Attorney General report favorably cites another report that concludes "In general, there is no inconsistency [between FCRA and state law] if the State law is more protective of consumers." U.S. Dept. of Justice Office of the Attorney General, The Attorney General's Report on Criminal History Background Checks (June 2006), p. 45.

61El v.SEPTA, supra, at 248.

62 Id. at 245.

63 See, Albermarie Paper Co. v. Moody, 422 U.S. 405, 431 (1975)

64El v. SEPTA, supra, at 252.

65 Paul Keefe from the Community Service Society was the principal author of the amicus brief.

66 U.S. Department of Justice, Bureau of Justice Statistics, "Truth in Sentencing in State Prisons 1999", p.6 (January 1999).

67 U.S. Equal Employment Opportunity Commission, Notice No. N-915.061, EEOC Policy Guidance on the Consideration of Arrest Records in Employment Decisions under Title VII of the Civil Rights Act of 1964.

68 In 2010 in New York City, 32% of all low level marijuana arrests were of Latinos. Levine, Harry and Siegal, Loren. 2011. $75 Million: The Cost of New York City's Marijuana Possession Arrests, p. 11 (March 2011). The Latino share of the City's population in 2010 was 28%.

69 Id. at n. 5.

70 The author acknowledges the research assistance provided by: Liz Kenney, a LatinoJustice PRLDEF legal intern from CUNY School of Law for her help throughout this statement and the preparation of the appendices; and the research assistance of Natalia Rivera Smith, a LatinoJustice PRLDEF legal intern from the Catholic University of Puerto Rico – Ponce School of Law for her assistance on the Puerto Rico section in particular.