On January 31st, the Trump administration issued a proclamation that stopped the issuance of most green cards to citizens of Nigeria, Kyrgyzstan, Eritrea, Myanmar, Tanzania, and Sudan. If the ban applied to these countries in 2018, it would have blocked 12,313 green cards that year.

This is the second wave of ‘travel bans’ issued by the Trump administration since the initial ban of many predominately Muslim-majority countries in 2017. The stated justification for these bans is to protect the public from terrorist and criminal threats that could be committed by immigrants from those countries. Furthermore, they claim to target green card holders because it is more difficult to deport them than other migrants.

However, these claims are not supported by historical data about the threat of terrorist attacks on U.S. soil. Furthermore, because the administration fails to publicly release the criteria used to determine why these countries are banned, the public cannot truly know why these countries were chosen.

My own research shows that citizens of the new banned countries have not contributed significantly to terrorism. From 1975 through the end of 2017, 11 foreign‐​born terrorists from those countries attempted or committed attacks on U.S. soil. They murdered six people in their attacks. The annual chance of being murdered by a foreign‐​born terrorist from those six countries on US soil is approximately 1 in 1.9 billion per year.

Those six murders by terrorists, as tragic as they were, account for a mere 0.2 percent of the 3,037 people murdered by foreign‐​born terrorists on U.S. soil during that 43-year period. By those metrics, a ban on the listed countries would hardly curb terrorist attacks or do much of anything to protect Americans from the risk of attack.

The most perplexing part of the ban is that it targets green card holders, although they are even less likely to commit attacks on U.S. soil. Green cards are not the way terrorists attempt to enter the country. Only about 0.6 percent of all murders committed by foreign-born terrorists on U.S.-soil were committed by those who entered with green cards.

At most, only 1 of those six murders was possibly committed by a foreign-born terrorist who entered on a green card: Emanuel Kidega Samson. And there is plenty of evidence that he might not have even come in on a green card, but rather, entered as a refugee. The new immigration ban, if applied retroactively, would have prevented exactly 1 murder in a terrorist attack at most from 1975 through the end of 2017.

But Samson is the exception. The other five murders were committed by the Boston Marathon bombers, who entered as asylum-seekers from Kyrgyzstan as children. The other 8 foreign-born terrorists didn’t murder anyone and they entered on other visas that weren’t green cards.

Crime is another justification for the immigration ban, but it too falls short. According to research by Michelangelo Landgrave and me, the incarceration rate for legal immigrants from the six banned countries is about 186 per 100,000. In other words, for every 100,000 legal immigrants from those countries living in the United States in 2017, 186 of them would be incarcerated.

In contrast, the incarceration rate for all legal immigrants, a population known for low criminality, was twice as high as for those from the banned countries. And the incarceration rate for native-born Americans was about 8‑times higher.

If another justification for the ban was that it is harder to deport green card holders who commit acts of terrorism, that does not pass muster either.

Paul Herzog, a Los Angeles-based immigration attorney, says the “[p]roclamation states that it is easier to deport non-immigrants than permanent residents [green card holders] but that’s not correct. It’s the same legal procedures and standards.”

So, if the ban is not based on calculated threats of terrorism, crime, or the ability to deport criminals and terrorists after they arrive, what is it based on? The administration supposedly uses a government mathematical model that estimates the risk of terrorism and crime from immigrants from these countries and balances that against security measures abroad as well as the sharing of information about green card applicants with U.S. authorities.

I write “supposedly,” because that model has not actually been released, and our own analysis shows that those factors haven’t predicted bans in the past.

Given what we know, the immigration bans are inappropriate and arbitrary. There is no good public safety rationale for banning green card applicants from those six countries. Worse, the public doesn’t even know how the government determined which countries to ban because they haven’t made their mathematical model public.

For the safety of Americans, the United States should continue to ban the immigration of criminals and terrorists, but they should do so based on facts, evidence, and publicly defensible justifications.