Data vs. Anecdotes in a Fight Over New Mexico’s Bail Reform

New Mexico legal leaders and court officials have praised the reforms as an objective success, but a groundswell of opponents have tried to claim that people on pretrial release were still committing too many crimes, and the new rules were too lenient. In response, conservative legislators introduced bills to roll back reforms. Among the changes were proposals to shift the burden of proof from the government to the accused at pretrial proceedings to show that they won’t be a danger to the public if released.

The bills have yet to gain traction in the Legislature even as reform opponents claim that bail reform isn’t working in New Mexico.

Research by University of New Mexico researchers shows otherwise.

In How Accurate are Rebuttable Presumptions of Pretrial Dangerousness?: A Natural Experiment from New Mexico, which was released in July, researchers analyzed more than 15,000 felony cases in Bernalillo County across four years. They found that just 3% of people in these cases were charged with a new violent felony. If the state were to repeal its reforms and presume that every defendant should be held pretrial, there is no evidence that crime would be reduced. At the same time, thousands of people who would’ve otherwise simply returned to their lives would be forced to stay behind bars. Jail populations would increase, exacerbating long-standing safety issues at the Bernalillo County jail.

“You know, when people discuss this in public, they’re often unmoved by these statistics because they say ​‘What about those crimes that would have been prevented?’” said Cristopher Moore, a researcher at the Santa Fe Institute. ​“But what we typically don’t hear are the many stories of people who are detained unnecessarily and who lose their jobs, their housing, custody of their children, and so on.”

Replacing a ​‘Less Honest Method’

Prior to New Mexico’s bail reform, judges were required to set the least restrictive bail option and order release conditions that ​“will reasonably assure appearance of the person as required” and ​“the safety of any other person and the community.” In setting these release conditions, judges were supposed to evaluate factors like the strength evidence against the individual, their family ties, and criminal history.

However, the New Mexico Supreme Court found judges were not following those rules. This problem came to public attention during the high court’s review of the case of Walter Brown.

Brown was charged with first-degree felony murder and second-degree murder for fatally stabbing someone with a pocketknife in a scuffle. He had been held for nearly three years on a $250,000 bond, which Brown could not afford. Though a state psychologist and pretrial services coordinator had testified that Brown was a suitable candidate for release under GPS monitoring because he had a place to live and work, and no prior history of violence, the judge had held him on two separate occasions because of the severity of his charge.

“Intentionally setting bail so high as to be unattainable is simply a less honest method of unlawfully denying bail altogether,” wrote Justice Charles Daniels in his 2014 majority opinion.

He continued, ​“We understand that this case may not be an isolated instance and that other judges may be imposing bonds based solely on the nature of the charged offense without regard to individual determinations of flight risk or continued danger to the community.”

Once voters agreed that New Mexico’s bail system needed to be transformed, the state’s highest court determined new rules. These included a presumption of release for individuals who came before the court, with prosecutors bearing the burden of convincing a judge that they should be jailed before trial.

Judges would also reference a risk assessment, an empirically formulated evaluation that’s designed to help inform decisions about whether an individual will succeed —avoid new charges and make their court dates — during the pretrial period. Judges use the score to help determine release conditions. For example, in Bernalillo County, an individual may be required to regularly check in with their pretrial supervision officer or agree to drug use treatment in order to be released before trial.

‘It’s Pretty Jarring’

Judges have sided with prosecutors roughly half of the time with regards to ordering someone to be jailed before trial, according to University of New Mexico researchers. Opponents of bail reform have viewed that data through a different lens, basing their opinions on a handful of high-profile cases in which individuals released pretrial committed serious crimes. Those cases have been held up by critics as supposed proof that the new system is not working even though under the previous system of cash bail there would have been nothing to prevent these individuals from simply paying their way out of jail.

University of New Mexico researchers say that the data shows that reform opponents’ approach would not make communities safer while also needlessly detaining too many people. To start, over 97% of people released pretrial do not commit a new felony.

“Our research has also shown that large portions of crime are not committed by people who are out pretrial. It’s the wrong target population,” said Paul Guerin, a researcher with the University of New Mexico’s Center for Applied Research and Analysis.

“We would have to detain 20 or 30 people for every violent felony that we would prevent,” said Moore. ​“It’s pretty jarring.”

Researchers also found that prosecutors who oppose release aren’t accurately predicting when someone will commit a new crime. More than 75% of the time, prosecutors filed a motion to try to detain someone based on their threat of dangerousness but, once released, 85% of individuals did not commit a new crime, according to the study.

Legislation set forward as supposed fixes wouldn’t be effective either. A proposed House Bill 5, for example, would have created a presumption of detention if the individual committed a serious violent offense or discharged a gun. Researchers say that, based on the existing data, 85% of the people who would have been detained under the legislation would not go on to commit a new offense.

Meanwhile, the Bernalillo County jail, where defendants are detained pretrial, is in crisis. The jail’s former chief declared a state of emergency in June, citing staffing shortages that had led to unsafe and unsanitary conditions. Though overcrowding has largely been alleviated for the time being, problems remain.

“We’ve got more beds, more footage and roughly the same number of people that we had before,” Peter Cubra, a lawyer who represented prisoners in a 1995 lawsuit over jail conditions, told Searchlight New Mexico. ​“But the staffing is grossly inadequate. The safety, medical and mental health care are not significantly better than 1995.”

This summer, for instance, the jail’s air conditioning broke down for a week. Outside, temperatures soared to 102 degrees.

“I don’t know how you prove a negative”

State Representative Moe Meastas, who serves parts of Albuquerque, said that he thinks it’s unlikely that any of the bills to push back reforms will be signed into law. But it’s difficult to convince constituents who want to enforce harsher bail conditions that the data shows they won’t be effective.

“People associate crime prevention with arrests and jail as opposed to successful prosecutions,” Meastas said. ​“So issues with bail reform are eighth or ninth on the list [of things concerning public safety]. But it’s number one in everyone’s minds because it’s so easy to associate with the current crime problems. So that’s where the political pressure is applied.”

Meanwhile, defense lawyers have urged lawmakers to rely on the available data on bail reform rather than emotion and criticized the idea of shifting the burden to people accused of a crime.

“I don’t think that people who are presumed innocent should have to prove that they should get out of jail,” Jonathan Ibarra, vice president of the New Mexico Criminal Defense Lawyers Association, told Source NM in January. ​“Shifting the burden onto primarily poor people, primarily people of color, to somehow prove a negative, to prove that they’re not going to do something bad — I don’t know how you prove a negative.”