The claim

“The aggregate murder rate of Muslims is one third that of the native-born.” Bill Clinton, November 7 2017

The background

Former President Bill Clinton marked the 25th anniversary of his election this week with a speech at Georgetown University.

He bemoaned the state of gun violence in America and made a number of remarks interpreted as criticisms of the Trump administration.

President Clinton also said: “In spite of the horrible killings in San Bernardino and Fort Hood, and what happened in New York City just a few days ago, the aggregate murder rate of Muslims is one third that of the native-born. How many people know that? If all your neighbours knew it, would it make a difference?”

It’s an eye-catching statistic. But we can’t find any evidence that it’s true.

The analysis

President Clinton appears to be blurring two different concepts when he talks about a “murder rate of Muslims” compared to “the native-born”.

He is confusing religion with immigration status here, and the two groups are not mutually exclusive. Clearly, you can be a native-born American Muslim.

So it’s not absolutely clear what he means, but we can only presume from the context that he is saying Muslims – perhaps Muslim immigrants specifically – are less likely to murder people than others.

There are two official sources of national homicide statistics in the US: the FBI and the Center for Disease Control and Prevention. Neither of these organisations records the religion of the offender. So it’s not possible to construct a “murder rate of Muslims” from official figures.

The closest things we can find to what he may have had in mind are studies like this, which compare the incarceration rate for native-born Americans and people born in the mostly Muslim-majority countries President Trump is attempting to target with a travel ban.

The study shows that those born the “travel ban” countries had an incarceration rate of 0.32 per cent, compared to 1.54 per cent of the native-born.

But this is a long way from establishing a murder rate among Muslims. Not all the countries in the sights of the current administration are Muslim majority. Not everyone from one of those countries who ends up being convicted of a crime in the United States will be a Muslim.

These stats refer to people in jail, not just those who have committed murder, and the study doesn’t stand up the “one third” ratio quoted by President Clinton either.

The author of the study, Alex Nowrasteh, an immigration policy analyst at the Cato Institute, told us: “There are no official murder or crime statistics by the religion of the offender in the United States.  The US Census does not ask about religion.  There are private surveys about prisoners but I’m not aware of any by the type of offense.”

The only other source we can find for a discussion of Muslims and the murder rate is this paper published by the Triangle Center on Terrorism and Homeland Security.

The author, Professor Charles Kurzman of the University of North Carolina at Chapel Hill, told FactCheck: “I suspect that Clinton may have been trying to juxtapose these two ratios:

• The number of Muslim-Americans involved in violent extremism in a given year, as a proportion of the Muslim-American population (46 out of 3 million in 2016, or approximately 1 in 65,000).
• The number of murders in the United States in a given year, as a proportion of the entire US population (17,250 out of 323,127,513 in 2016, or approximately 1 in 18,000 ).

“These figures suggest that the rate of violent extremism among Muslim-Americans is one third the murder rate in the US population at large.”

If this is indeed what President Clinton had in mind, it’s a long way from calculating a “murder rate” just for Muslim Americans and comparing it to the same thing in the “native-born” population. Professor Kurzman’s study does neither of these things.

Killings by violent extremists (a non-academic would probably say “terrorists”) are one very small subset of total murders, so we’re not really comparing like with like.

The verdict

There are statistics which suggest that the risk of being killed by a Muslim extremist is relatively low, and that people from the mainly Muslim-majority nations targeted by the Trump administration’s plans for a travel ban are less likely to be in a US prison compared to a native-born American.

But we can’t find anything that establishes a “murder rate” either among Muslim Americans, or among immigrants from Muslim-majority countries.

Nor is it clear how such a rate could possibly be calculated from the available homicide data sets, since they do not break down offenders by religion.

We have asked President Clinton’s office for an explanation of the claim and will update if we get a response.