OAKLAND, California – The two Honduran women told nearly identical stories to the immigration courts: Fear for their lives and for the lives of their children drove them to seek asylum in the United States.

They were elected in 2013 to the board of the parent-teacher association at their children’s school in the Honduran capital, Tegucigalpa. They hoped that mothers working together could oust the violent gangs that plagued the campus.

Instead, they became targets. Weeks apart, in the spring of 2014, each of the women was confronted by armed gang members who vowed to kill them and their children if they didn’t meet the thugs’ demands.

Unaware of each other’s plight, both fled with their children, making the dangerous trek across Mexico. Both were taken into custody near Hidalgo, Texas, and ended up finding each other in the same U.S. Immigration and Customs Enforcement (ICE) detention center in Artesia, New Mexico. There, they applied for asylum.

That’s when their fates diverged.

Sandra Gutierrez joined her husband in California, where her case was heard by a San Francisco immigration court judge. At the end of her asylum hearing in September 2016, she received a one-page form, with an “X” in the box next to “granted.” She was free to settle into life with her family in the United States.

The other woman, Ana, joined her daughter’s father in the southeastern United States, and her case was assigned to an immigration court in Charlotte, North Carolina. The judge denied her petition and ordered her deported. She is now awaiting a court date after new lawyers got her case reopened.

Ana declined to be interviewed for this article. Through her lawyers, she asked that her full name not be used because of her uncertain status and her fear that Honduran gangs could find her.

The women’s lawyers framed their respective cases with some important differences. However, the women said their reasons for seeking asylum were the same: Gangs had targeted them because of their involvement in the parent-teacher association, and for that, they and their families had been threatened.

Taken together, the two cases – nearly indistinguishable in their outlines but with opposite outcomes – illustrate a troubling fact: An immigrant’s chance of being allowed to stay in the United States depends largely on who hears the case and where it is heard.

Judge Stuart Couch, who heard Ana’s case in Charlotte, orders immigrants deported 89 percent of the time, according to a Reuters analysis of more than 370,000 cases heard in all 58 U.S. immigration courts over the past 10 years. Judge Dalin Holyoak, who heard Gutierrez’s case in San Francisco, orders deportation in 43 percent of cases.

In Charlotte, immigrants are ordered deported in 84 percent of cases, more than twice the rate in San Francisco, where 36 percent of cases end in deportation.

Couch and Holyoak and their courts are not rare outliers, the analysis found. Variations among judges and courts are broad.

Judge Olivia Cassin in New York City allows immigrants to remain in the country in 93 percent of cases she hears. Judge Monique Harris in Houston allows immigrants to stay in just four percent of cases. In Atlanta, 89 percent of cases result in a deportation order. In New York City, 24 percent do.

The Reuters analysis used data from the Executive Office for Immigration Review (EOIR), the U.S. Justice Department unit that oversees immigration courts. The count of deportations included cases in which judges allowed immigrants to leave the country voluntarily.

The analysis excluded immigrants who were in detention when their cases were heard because such cases are handled differently. It also excluded cases in which the immigrant did not appear in court, which nearly always end in a deportation order, and cases terminated without a decision or closed at the request of a prosecutor.

About half the cases in the analysis were filed by asylum seekers like the two Honduran women. The rest were requests for cancellation of deportation orders or other adjustments to immigration status.

“GROSS DISPARITIES”

Of course, other factors influence outcomes in immigration court.  For example, U.S. government policy is more lenient toward people from some countries, less so for others.

Also, immigration judges are bound by precedents established in the federal appeals court that covers their location. Immigration courts in California and the Pacific Northwest fall under the 9th U.S. Circuit Court of Appeals, and they rule in favor of immigrants far more often than courts in the 4th Circuit, which includes North and South Carolina, Maryland and Virginia, Reuters found.

Even so, the Reuters analysis determined that after controlling for such factors, who hears a case and where it is heard remain reliable predictors of how a case will be decided. An immigrant was still four times as likely to be granted asylum by Holyoak in San Francisco as by Couch in Charlotte.

The Reuters analysis also found that an immigration judge’s particular characteristics and situation can affect outcomes. Men are more likely than women to order deportation, as are judges who have worked as ICE prosecutors.  The longer a judge has been serving, the more likely that judge is to grant asylum.

“These are life or death matters. … Whether you win or whether you lose shouldn’t depend on the roll of the dice of which judge gets your case.”

The findings underscore what academics and government watchdogs have long complained about U.S. immigration courts: Differences among judges and courts can render the system unfair and even inhumane.

“It is clearly troubling when you have these kinds of gross disparities,” said Karen Musalo, director of the Center for Gender & Refugee Studies at the University of California Hastings School of the Law in San Francisco. “These are life or death matters. … Whether you win or whether you lose shouldn’t depend on the roll of the dice of which judge gets your case.”

EOIR spokeswoman Kathryn Mattingly said the agency does not comment on external analyses of its data.

Devin O’Malley, a Department of Justice spokesman, challenged the Reuters analysis, citing “numerous conflicting statements, miscalculations, and other data errors,” but declined to elaborate further.

Immigration judges, appointed by the U.S. attorney general, are not authorized to speak on the record about cases.

Dana Marks, president of the National Association of Immigration Judges, said each case is like “a 1,000 piece puzzle.” While two cases might look identical on the surface, she said, each judge has to weigh the nuances of immigration law to allow someone to stay in the country, which could lead to different outcomes.

The question of equality of treatment among judges has gained urgency as the number of cases in immigration court has ballooned to record highs. Under President Barack Obama, the courts began efforts to hire more immigration judges to reduce the system’s burgeoning backlog, which now stands at more than 620,000 cases, nearly 100,000 of them added since last December.

The administration of President Donald Trump is continuing the effort. Attorney General Jeff Sessions said in April that the Justice Department planned to hire more than 50 judges this year and 75 in 2018, which would put the total number of sitting judges above 400.

Of the 28 immigration judges Sessions has appointed so far, 16 are former ICE prosecutors. That experience, the Reuters analysis found, makes them 23 percent more likely to order deportation. (Neither Holyoak nor Couch worked as an ICE prosecutor, according to their EOIR biographies.)

In a wish list of immigration proposals sent to Congress on Oct. 8, the White House said that “lax legal standards” had led to the immigration court backlog and that “misguided judicial decisions have prevented the removal of numerous criminal aliens, while also rendering those aliens eligible to apply for asylum.” Among the proposals offered in exchange for a deal with Congress on the roughly 800,000 “dreamers” – children brought to the country illegally by their parents – the Trump administration said it wanted to hire even more immigration judges and 1,000 ICE attorneys, while “establishing performance metrics for Immigration Judges.”

CRISIS AT THE BORDER

In 2014, an unprecedented 68,000 parents and children, most of them fleeing violence and lawlessness in El Salvador, Guatemala and Honduras, crossed into the United States from Mexico – a refugee crisis that has contributed to the bloated backlog of asylum petitions. Many of the migrants, including Gutierrez and Ana, convinced initial interviewers that they had a “credible fear” of returning home, the first step in filing an asylum claim.

Having come from a country with one of the highest murder rates in the world may have helped establish “credible fear.” But the two women were already at a disadvantage – precisely because they came from Honduras.

Country of origin is a big factor in determining who gets to stay in the United States because immigrants from some countries are afforded special protections. For example, courts ruled in favor of Chinese immigrants 75 percent of the time, the Reuters analysis found. A 1996 law expanded the definition of political refugees to include people who are forced to abort a child or undergo sterilization, allowing Chinese women to claim persecution under Beijing’s coercive birth-control policies.

Hondurans enjoy no special considerations. They were allowed to stay in the United States in just 16 percent of cases, the Reuters analysis found.

The mass exodus from Central America was under way when Gutierrez and Ana were elected to the board of the parent-teacher association at their children’s school in spring 2013.

Two rival gangs – the Barrio 18 and the Mara Salvatrucha, also known as MS-13 – were operating brazenly in the neighborhood. The year before, according to police records in Honduras, gang members killed a school security guard. Now, they were extorting teachers, selling drugs openly and assaulting or killing anyone who confronted them.

The new six-member association board set about trying to improve security at the school, which sits on a dirt road behind a high wall topped with razor wire.

“Before, no one wanted to say anything about the gangs,” Gutierrez said. “We were the brave ones. The previous president was a man, so we thought, ‘We are women, they won’t do anything to us.’ ”

The school’s principal, who asked that he and the school not be identified out of fear of retaliation, worked with the board. They had early success, he said, when they persuaded police to provide officers to guard the school. But the patrols left after a few weeks, probably intimidated by the gangs.

One evening in April 2014, Gutierrez was watching television at home with her two sons, ages 5 and 11, when she heard banging at the front door. Her older boy recognized the three armed and heavily tattooed young men on the stoop as the same ones who had thrown him to the ground earlier that day, telling him, not for the first time, that they wanted him to join their ranks. Now they had come to deliver a message to Gutierrez.

“They said they knew I was involved in the parents’ association,” Gutierrez said. “They said they would kill me and my children.

“I began to panic and shake,” she said. “I thought, ‘I have to go now. I am not going to risk my child’s life.’ ”

She quickly packed some backpacks for her and her children and called the only friend she knew who had a car. They drove all night to her friend’s mother’s house in another town.

“NO POLICE HERE”

Two months later, according to court documents, Ana was walking her 7-year-old daughter home from school when three members of a rival gang confronted them. Two of them grabbed Ana and her daughter, pinned their wrists behind their backs, and pointed a gun at the child’s head. The third pointed a gun at Ana’s head. They demanded that a payment of more than $5,000 be delivered in 24 hours, a huge sum for a woman who sold tortillas for a living.

Ana testified in her asylum hearing that she knew they were gang members “because they were dressed in baggy clothing and they also had ugly tattoos … all over their bodies and faces.”

Ana and her daughter ran home and then, fearing the gang would come after them, fled out the back door. “We had to jump over a wall, and I hurt my foot doing so,” she said in an affidavit. “I was desperate and knew that I had to leave – my daughter’s life and mine were in danger.”

The school principal said he understands why Gutierrez and Ana left Honduras. “Because there were no police here, (the gangs) did what they wanted,” he said. “They said, ‘We’re going to kill the members of the parent-teacher association to get them out of here.’ So the women fled.”

Gutierrez hid for two months at her friend’s mother’s house outside Tegucigalpa. She joined another woman and, with their children, they set out to cross Mexico. On the journey, they were kidnapped – common for Central American migrants – and held for a $3,500 ransom. Gutierrez contacted relatives who wired the money. The kidnappers released her and her two sons near the U.S. border.

There they piled with another group of migrants into an inflatable raft and crossed the Rio Grande, the border between Mexico and the United States. They landed near Hidalgo, Texas.

After walking for an hour and a half, lost and desperate, Gutierrez and her sons sat down in the middle of a dirt road and waited for someone to pass. Two officials in uniforms picked them up. They were eventually transferred to the ICE detention center in Artesia.

Ana fled with her daughter the night the gang members threatened them on the street. “We bought a bus pass to go to Guatemala and from Guatemala to Mexico and to the U.S.-Mexico border,” according to her court testimony. The journey took three weeks. In Mexico, she hired a coyote – a smuggler – to help them cross into the United States and then turned herself in to Border Patrol agents near Hidalgo. She arrived at the Artesia detention center just weeks after Gutierrez.

“The other women in the center told me that there was someone else from Honduras who I might know, but I wasn’t sure who they were talking about,” Gutierrez said. “And then one day we went to lunch, and there they were.”

Gutierrez said that was when she first learned that her fellow parent-teacher association board member had been threatened and had fled from home.

Volunteer lawyers helped the women prepare and submit their applications for asylum.

In late 2014, the two women were released on bond. Gutierrez moved with her boys to Oakland, California, to join her husband, and petitioned to have her case moved to San Francisco. Ana moved with her daughter to live with her daughter’s father and petitioned to have her case moved to Charlotte.

“ASYLUM FREE ZONES”

Many immigrants released on bond before their cases are heard have no idea that where they settle could make the difference between obtaining legal status and deportation.

People familiar with the system are well aware of the difference. When Theodore Murphy, a former ICE prosecutor who now represents immigrants, has a client in a jurisdiction with a high deportation rate but near one with a lower rate, “I tell them to move,” he said.

The Charlotte court that would hear Ana’s case was one of five jurisdictions labeled “asylum free zones” by a group of immigrant advocates in written testimony last December before the Inter-American Commission on Human Rights. The courts in Dallas, Houston, Las Vegas and Atlanta also received the designation.

The advocates testified that, while asylum is granted in nearly half of cases nationwide, Charlotte judges granted asylum in just 13 percent of cases in 2015. The Charlotte court was singled out for displaying a particular “bias against Central American gang and gender-related asylum claims.”

Couch is the toughest of Charlotte’s three immigration judges, according to the Reuters analysis.

The Transactional Records Access Clearinghouse, a research organization at Syracuse University in New York, first sounded the alarm about disparities in immigration court decisions in 2006. The next year, researchers at Temple University and Georgetown Law School concluded in a study titled “Refugee Roulette” that “in many cases, the most important moment in an asylum case is the instant in which a clerk randomly assigns an application to a particular asylum officer or immigration judge.” In 2008, the U.S. Government Accountability Office (GAO) found similar disparities in its own study.

In response to the rising criticism, the Executive Office for Immigration Review began tracking decisions to identify judges with unusually high or low rates of granting asylum. Mattingly, the EOIR spokeswoman, said the agency held training sessions for judges to address the disparities in 2008 and 2009. It then created a system for the public to file complaints against immigration judges.

In a 2016 report, the GAO found that little had changed. EOIR held a two-day training session last year. There is no training on the 2017 calendar.

From 2012 to 2016, EOIR received 624 complaints against judges. The 138 complaints lodged in 2016 alone included allegations of bias, as well as concerns about due process and judges’ conduct within the courtroom. Of the 102 complaints that had been resolved when the data were published, only three resulted in discipline, defined as “reprimand” or “suspension” of the judge. “Corrective actions” such as counseling or training were taken in 39 cases. Close to half the complaints were dismissed.

The agency does not identify judges who were the subjects of complaints.

Mattingly, the EOIR spokeswoman, said the agency “takes seriously any claims of unjustified and significant anomalies in immigration judge decision-making and takes steps to evaluate disparities in immigration adjudications.”

DAY IN COURT

Asylum applicants cannot gain legal U.S. residency because they fled their countries in mortal fear of civil strife or rampant crime or a natural disaster. They must convince the court that they have well-founded fears of persecution in their country because of their race, religion, nationality, political opinions or membership in a particular social group. The definition of a “particular social group” has been subject to conflicting interpretations in the courts, but in general, such a group comprises people who share basic beliefs or traits that can’t or shouldn’t have to be changed.

In the San Francisco court, Gutierrez’s lawyers argued that she qualified for asylum because as a leader of the parent-teacher association, she was at risk for her political opinion – her stand against gangs – and for belonging to a particular social group of Hondurans opposed to gang violence and recruitment in schools. The lawyers also argued that she was part of another particular social group as the family member of someone under threat, since the gangs had terrorized her son in trying to recruit him.

Holyoak was convinced. Gutierrez told Reuters that during her final hearing, the judge apologized for asking so many questions about what had been a painful time in her life, explaining that he had needed to establish her credibility.

In the Charlotte court, Ana’s lawyer focused more narrowly on her political opinion, arguing that she was at risk of persecution for her opposition to gangs in her position on the parent-teacher association board.

After hearing Ana’s case, Couch concluded in his written opinion that Ana was not eligible for asylum because she had “not demonstrated a well-founded fear of future persecution on account of a statutorily protected ground.” He wasn’t convinced that she risked persecution in Honduras because of her political opinion.

Well-established law recognizes family as a protected social group, according to the Center for Gender & Refugee Studies. Cases that claim opposition to gangs as a protected political opinion, the center says, have generated fewer precedent-setting decisions, making that argument a more difficult one to win in court, though it has prevailed in some cases.

Ana’s response to Couch’s extensive questioning played a part in the decision. In immigration court, the asylum seeker is typically the only witness.  As a result, “credibility is really the key factor. Persecutors don’t give affidavits,” said Andrew Arthur, a former immigration judge who now works at the Center for Immigration Studies, a nonprofit organization that supports lower levels of immigration.

Couch wrote in his opinion that Ana’s difficulty recounting the names of the women on the association board weighed against her credibility. He noted that she testified about her fears of the gang “with a flat affect and little emotion,” displaying a “poor demeanor” that “did not support her credibility.”

The judge also questioned why, in an early interview with an asylum officer, Ana never mentioned threats to the parent-teacher association, and instead said she thought the gangs were targeting her for the money her daughter’s father was sending from the United States to build a house in Honduras.

Ana’s assertion that she learned from Gutierrez in detention about gang threats to the parent-teacher association was not “persuasive,” Couch wrote. “The evidence indicates this is a case of criminal extortion that the respondent attempts to fashion into an imputed political opinion claim.”

“SOMEONE WANTS TO KILL THEM”

Gutierrez said Ana told her in one of their occasional phone conversations that she felt intimidated by the intense questioning of the ICE attorney. Gutierrez also said her friend “is very forgetful. … It’s not that she is lying. It’s just that she forgets things.”

Lisa Knox, the lawyer who represented Gutierrez, said judges where she practices tend to give applicants the benefit of the doubt. “They have more understanding of trauma survivors and the difficulty they might have in recounting certain details and little discrepancies,” she said.

Further, Knox said, asylum seekers aren’t thinking about the finer points of U.S. asylum law when they are fleeing persecution. “People show up in our office (and) they have no idea why someone wants to kill them. They just know someone wants to kill them.”

Ana’s lawyer appealed her case to the Board of Immigration Appeals (BIA), the first step in the appellate process. This time, her lawyer included arguments about her membership in a particular social group. She lost. In a three-page ruling, one board member said Ana’s lawyer could not introduce a new argument on appeal and agreed with Couch that Ana hadn’t proved a political motive behind the gang members’ attack.

Ana missed the deadline to appeal the BIA decision to the 4th U.S. Circuit Court of Appeals because her lawyer confused the deadline. She petitioned the BIA through new lawyers to reopen her case and send it back to the immigration court to allow her to present new evidence of her persecution. The new lawyers argued that her previous representation had been ineffective.

In July, the BIA granted Ana the right to a rehearing in immigration court, sending her case back to Charlotte, where it could be heard again by Couch.

Gutierrez can live and work legally in the United States and will ultimately be able to apply for citizenship. The 43-year-old, who worked as a nurse in Honduras, lives in a small one-bedroom apartment with her husband, her two sons – now 15 and 8 – her adult daughter and her grandson. She works as an office janitor and is taking English classes. Her boys are in school. The older one, once threatened by gangs in Honduras, likes studying history and math and is learning to play the cello.

Ana, 31, has had a baby since arriving in the United States and has been granted work authorization while she awaits a final decision on her case. She and her lawyers declined to share more detailed information about her situation because she remains fearful of the gangs in Honduras.

“I am very worried about her,” Gutierrez said. “The situation in our country is getting worse and worse.”

Last February, a 50-year-old woman and her 29-year-old son who were selling food at the school Gutierrez and Ana’s children attended were kidnapped from their home and decapitated, according to police records.

The head of the son was placed on the body of the mother and the head of the mother was placed on the body of the son. The murders, like more than 93 percent of crimes in Honduras, remain unsolved.