911 Emergency Response System Significantly Reduced U.S. Murder Rate, Says Study

More than a 34 percent to 56 percent decrease in homicides attributed to the 911 systems


Monkey Business Images/Dreamstime

The U.S. homicide rate peaked in 1980 at 10.2 per 100,000 Americans and the number of annual homicides rose to 24,703 in 1991. Since then U.S. homicide rates have been falling and reached 4.5 per 100,000 in 2014 (a rate not seen since the 1950s) and an annual toll of 14,249. Both the rate and number of murders ticked up in 2015 to 5 per 100,000 and 15,809 respectively. Lots of research has been devoted to trying to figure out why homicide rates fell over the past couple of decades. Some researchers focused on higher incarceration rates; others on more effective policing; still others cited the aging population; and some attributed lower homicide rates to better emergency medicine. Given the fact that the data on criminally inflicted gunshot injuries is not collected comprehensively, this this latter claim has been challenged.

A fascinating new working paper, "Dial 911 for Murder," by George Mason University economists Thomas Stratmann and David Chandler Thomas argue that advances in the 911 emergency response system over time, combined with the advent of cell phones have contributed significantly to the lower U.S. murder rate because victims are receiving emergency medical care ever more speedily. The argument is that victims of violent gun firearms attacks are increasingly likely to be saved from death thus converting what would have been homicides into aggravated assaults. In addition, they find the establishment of 911 systems initially increased reporting of aggravated assaults and that the subsequent proliferation of cell phones seems to have had a deterrent effect on such assaults.

The initial 911 system devised in 1968 connected to an operator who then transferred the call to the police, fire, and ambulance services as appropriate. In the late 1970s, dedicated 911 call centers were established with monitors that displayed the home number and address of callers and from which emergency services were dispatched by trained personnel. Now GPS 911 is rolling out enabling dispatchers to pinpoint callers and dispatch services to people using cell phones. The researchers suggest that "when potential violent street offenders know that cell phone users have a quick way to reach the police, via GPS 911, their incentive to commit a crime decreases."

Stratmann & Thomas

The researchers account for the lethality of weapons over time, socioeconomic changes, improvements in trauma care, and so forth. Once they've parsed the data, they report:

While the level of violent crimes, measured as aggravated assaults, is 219 percentage points higher in 2014 than in 1964, homicides in 2014 are eight percentage points lower than in 1964. In this paper, we find support for the hypothesis that the introduction of 911 services explains much of the decrease in homicide rates. Moreover, the introduction of 911 provides an explanation for the divergence between aggravated assault and homicide rates that started in the early 1970s. …The empirical results in this paper indicate that reductions in emergency response times played a significant role in reducing U.S. homicides over the past 45 years. ….

The reported 911 effects are quantitatively important, suggesting more than a 34 percent to 56 percent decrease in homicides that can be attributed to the 911 systems and associated reductions in response times. One simple approach to illustrate the benefits of 911 is to use our estimates to quantify the lives saved by 911 innovations. Such an exercise shows that in 2014, without 911 emergency services, the number of homicides in our sample cities would have been more than 13,000 homicides, instead of the reported 5,872.

Citing the trends in Mobile, Alabama they report that the introduction of basic 911 coincided with an immediate 29 percent drop in that city's homicide rate which continued to fall in the following years. However, after the introduction of enhanced 911 in the late 1980s, the city's homicide and assault rates continued to rise. They attribute that increase to the crack cocaine epidemic that dramatically boosted violent crime throughout the U.S. Mobile's murder rate did increase slightly between 1988 and 1991, but was much lower than the 160 percent national increase. The 2009 roll out of GPS 911 was followed by an immediate 42 percent decline in homicides in the city.

Violent crime is indeed is down in the U.S., but Stratman and Thomas make an interesting case that the speedy dispatch of emergency services made possible by the improving 911 system may make homicide statistics much better than they would otherwise be.

NEXT: Jeff Sessions Provides Slippery Answers at Confirmation Hearings, Thanks to Senatorial Decorum

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Good news? We don’t want it.

  2. See? What do you need a gun for?

    1. Exactly, an armed security drone is much more reasonable for self defense.

      1. are you referring the our police or an actual robot?

        I am assuming a robot….but….cops are drones but they aren’t good or reasonable….. :/

  3. The phone is mightier than the sword.

  4. Watching CNN’s Pravda-like panel respond to Trump’s press conference. It’s gold. Larry Noble was just blatantly shot down over conflicts of interest.

    1. What exactly was “Pravda”-like about it? I saw the panel, was Larry Noble the business ethics guy they had on? He was wrong about what Trump was doing was a violation of the law – it isn’t – but he was right that Trump’s proposed remedy for the conflicts of interest doesn’t really solve the problem. Transferring ownership to his kids and then forbidding the kids from talking to their dad about what their businesses are doing? Yeah like that’s really going to work. Plus, he is right, if foreign businesses want to influence Trump, they can just influence his family instead, and it will have the same net effect.

      I do hope Congress takes a look at the conflict-of-interest rules and revises them to deal with situations like Trump and his vast company.

      1. “I do hope Congress takes a look at the conflict-of-interest rules and revises them to deal with situations like Trump and his vast company.”

        Why would having a company leave him any more open to a conflict of interest than someone who doesn’t?

      2. Federal acquisition rules have pretty tight controls on conflict of interest. Cabinet officials have to be confirmed by Congress and can be removed by Congress. I’m not sure what more needs to be done, except for Congress to keep a watchful eye like they always should be doing.

        1. Certainly didn’t keep the hag from profiting handsomely while she dealt with various countries.

        2. It occurs to me that I must have had a temporary lapse of imagination if I couldn’t fathom ways that the President could work things out to his favor while still technically complying with the law.

      3. When Congress subjects itself to real conflict of interest rules, and related laws like insider trading, then I’ll give a fuck what their opinions are on other people’s conflicts of interest.

        1. I do hope Congress takes a look at the conflict-of-interest rules and revises them to deal with situations like Trump and his vast company.

          Considering ‘Password’ is sufficient to bring half the government to it’s knees entirely without for-profit businesses explicitly involved, fuck conflict-of-interest hearings.

          I mean, the Clinton Foundation was rather explicitly/openly established for political favors and only acquires a conflictless legitimacy by not obeying rules handed down by a POTUS who’s on the same team.

          I’d hope Congress craters with conflict-of-interest hearings and are forced to rein in executive power rather than divining thought crimes based on whether a sitting President maliciously intended to keep an email server in their closet or not.

        2. Whoops, meant in reply to chemjeff above.

  5. “While the level of violent crimes, measured as aggravated assaults, is 219 percentage points higher in 2014 than in 1964…”

    No, that is not what the graph shows. If it is a percentage change from 1964, how can 1964 be at 100%???? And the percentage change from 1964 is 8% not 92%

    Either state that in 2014, aggravated assaults were 219% of 1964 while murders where 92% of 1964
    That in 2014 aggravated assaults where 119% percentage points higher than in 1964 while murders where 8% lower.

    1. And the percentage change from 1964 is 8% not 92%

      They got that part right, oddly enough. It’s a mess the way they’ve worded it. It would be both clearer and more accurate to say “the homicide right declined slightly since 1964 while the assault rate has more than doubled since then”.

      1. RATE not right

      2. It is funny, in the narrative they got that one right. However, the title and the description of the graph are wrong.

        “The vertical axis measures the percentage change in homicide and aggravated assault rates relative to 1964…”

        Clever trick that 1964 had 100% CHANGE compared to 1964.

          1. Sorry, misread your comment.

    2. BO: You are entirely right that their wording is confusing. I think that kb gets what they are trying to say correct. They’ve set both murders and assaults at 100 in 1964 as an index number, so an increase to 219 would imply more than doubling in assaults, not a tripling in assaults.

      1. Ron: I appreciate your comment, however, looking at that graph, I am more convinced that it isn’t just that the wording is confusing. The rates of increase of both crimes were constant and about the same from 1964 to 1974. Then for about 2 years, the homicide rate actually dropped, but the aggravated assault rate kept increasing at the same rate. Then in 1976 or so, the changes in the rates of both crimes tracked fairly together. The murder rate is a little smoother, but the trends look quite similar.

        Granted, without parsing the data, it is difficult to make serious conclusions, but that graph is not even remotely indicating what the authors are trying to say.

  6. What the shit was going on in the early 90s that caused such a spike in the assault rate?

    And why can’t we get it back to where it was in 1964? I blame desegregation.

    1. Crack. No, seriously: crack.

      1. Pooky hardest hit.

    2. Cheers ran from 1982 to 1993. That spike in assaults in 1986 corresponds to the season leading up to Shelly Winters leaving and being replaced by Kirstie Alley. coincidence?!

      1. Shelly Long.

        1. “Sorry Mr. Peterson, but my chiropractor says I can’t carry you home anymore.”

        2. Shelly Winters wide not long. Extra wide

      2. Then The Fresh Prince of Bel-Air aired from 1990-1996 and millions of young Americans had Uncle Phil to look up to, thus contributing the decline in violent crime.

    3. *bounces Jim on my knee*

      There was the thing called Crack back in the olden days. Entire neighborhoods became shadows of their former husks where literally every 2nd or 3rd house was boarded up.

    4. What the shit was going on in the early 90s that caused such a spike in the assault rate?

      Talk Shows. EVERYONE had one. Almost drove me to commit act of depraved, violent evil.

    5. It’s interesting to watch action movies from the 80s and early 90s that depict the then near-future as a violent hellscape of urban blight. People back then were genuinely convinced violent crime would continue to escalate out of control with gangs and drugs and such.

      So movies like Robocop or The Death Wish series were just popular reflections of cultural anxieties along with all the other low-budget knock-offs that had the same themes.

      1. To be fair, Robocop was kind of right about Detroit.

        1. That’s why they put up a statue.

      2. Heinlein caught the trend earlier & wrote I Will Fear No Evil.

        1. Lesson – always wait for your bodyguards to escort you all the way to your door.

    6. The public had just found out that 911 was a joke in their respective towns.

      1. Glad I refreshed before commenting.

        1. Glad I searched first, I was gonna make that joke.

          1. Everyone who is anyone was going to make that joke.

  7. Given the fact that the data on criminally inflicted gunshot injuries is not collected comprehensively

    How conVEENient!

    /Church Lady

  8. Nah nah nah, it was the elimination of leaded gas.

    1. The Internet became mainstream around 1994 didn’t it? That’s when the graph really started trending down.

      1. Most people probably didn’t have Internet access in the home until around 2000. And the Internet in 1994 was drastically different than today, even for the people who did have home access to it.

        1. You’ve got to be kidding. I’ve been online from home since ’85. Granted it was a lot slower … I could read faster than the data came across on a 300 baud acoustic coupler connection. 🙂

  9. If what you report — a drop in murder rate due to quicker response and care — is correct, we would expect a rise in aggravated assaults to accompany a downturn in the murder rate. Is that what has been found?

      1. “People committing same amount of violence, just fewer dying from it” isn’t as catchy.

      2. Am I completely losing my mind or something? I ask because the graph in the embedded image right smack in the middle of your piece doesn’t show that. At all.

        That graph pretty clearly shows what looks to me like a correlating (yet even stronger and sharper) decline in the aggravated assault rate compared to the homicide rate over the time frame.

        1. To be a little more specific and accurate, what I see in that graph is an almost perfectly correlated increase in homicides and aggravated assaults from roughly 1964 to 1980. Then, from roughly 1980 to 1993, the aggaravated assault rate continues to rise sharply while the homicide rate levels off. Then from roughly 1993 to 2014, we have an almost perfectly correlated drop in aggravated assault and homicide, except that the aggravated assault rate is even sharper and faster.

        2. I think the case can be made up until 1990 but after then other factors are playing as large or larger of a role.

          Also, it’s easier to hide an assault than a dead body.

        3. Am I completely losing my mind or something?

          Can’t tell if serious.

          1. I’m 100% serious. Why don’t you try actually looking at the fucking graph? If you do, even a complete idiot like you should be able to notice that during the period of time where the homicide rate sharply drops (roughly 1993 to 2014), the aggravated assault rate drops in a correlating, but even sharper, manner.

            1. I agree with you 100%. From 1964 until about 1974, the rate of increase of homicide and agg. assault rates were almost exactly the same. Then for about 2 years, the homicide rate dropped off, while agg assault continued to increase. Then from 1976 on, the 2 rates change in a very similar (not exact) pattern.

              I am not going to waste my time parsing all of their data. But the more I look at this graph, the more aware I am that the author’s are full of shit. At least based on the graph that Ron presented.

              For those mathematically challenged, the points represent the rates of homicide or agg. assault (i.e. the numbers of crimes per 100,000 people). The slope of the graph represents the time rate of change of those rates. So if the slope of the 2 lines are approx. the same, that means the changes of those 2 statistics are approximately the same. If what the author’s were saying was true, one would expect that the sum of the homicides and agg. assaults together would together rise at the same rate regardless of the individual crime rate changes. But what I see is that there was an anomalous 2 year period from 1974 to 1976, then the rates of change of both crimes came back together (at least to a large degree.)

              1. IOW: when the murder rate lowered, the aggravated assault rate should have risen even faster. It didn’t. Until about 1980, there was essentially no change in the rate of increase of aggravated assault. It rose at a fairly constant rate since 1964.

  10. Dial 911 for Murder

    Let’s all just reflect on how truer that is than the authors intended.

    1. It isn’t murder if you have a people hunting license.

  11. I study shootings by and of police officers extensively and between bullet resistant vests and improved emergency medical care, as best I can tally – one officer is shot and killed for every 9 officers shot

    the ratio of officers shot/officers shot and killed is much higher than the general population (obviously)

    rifle assassinations (like Dallas recently) are obviously far more lethal, but the VAST majority of shootings of cops are by handgun and vests are very effective.

  12. PROTIP: If you get shot in Chicago, and are taken to a county hospital – they will be expert at saving you. However, the moment you are stable – beg, plead, threaten, bribe, whatever you can do… to transfer to Rush or Northwestern, etc.

    1. As an Illinoisan, I would not step one foot on Chicago territory, it needs to be it’s own state or country.

      1. It can be part of a new country combining the west coast plain, & Bos-Wash east coast. Then they can tax each other to prosperity.

        1. I wish them well.

  13. But how is this affecting climate change? All of those emergency vehicles running all around and spewing out carbon, it can’t be good.

    1. Its a vicious cycle. Climate change begets violent crime. Violent crime begets first responders. First reponders beget carbon spewing vehicles. Carbon spewing vehicles beget climate change.

      1. Well as soon as the cops have shot all people, the problem will be solved. Neez moar cops.

  14. But how many people were murdered because someone called 911? How does that skew the statistics?

    1. You mean when the cops show up and shoot someone? Obviously, there is some data on that as many times as we’ve heard those stories right here on H&R.

    2. an infinitesimal percentage since an infinitesimal percentage of police shootings are murders


      (stats available upon request)

      and of course the unmeasurable quantity is the # of shootings prevented because we quell a disturbance or bad guys run off etc. upon our arrival

      we’re from the govt and we are here to help


      1. You should read Reason articles, Gumby, it’s enlightening.

      2. You can’t spell “The artist known Dunphy” without “we shit on truth”.

        1. Don’t worry, Gumby, your name is Gumby now.

      3. an infinitesimal percentage since an infinitesimal percentage of police shootings are murders.

        Funny how when the cops investigate the cops, the homicide is found to be not murder in all but an infinitesimal amount of cases. I wonder how that happens.

        1. “We have investigated ourselves and have determined we did not act improperly!”

          Sounds legit to me.

      4. About 50 cops were shot and killed over the last year. The cops shot and killed nearly 1,000 people, as near as anyone can tell given their lack of reporting to any central database.

    3. Murders by cops don’t count because something something heroes something something thin blue line something something

      1. Totality of the circs, furtive movement, good shoot.

  15. While the level of violent crimes, measured as aggravated assaults, is 219 percentage points higher in 2014 than in 1964…

    Violent crime is indeed is down in the U.S….

    Down since 1994, but up since 1964. If I put on 100 pounds and then lose 50, I’d be reluctant to say: “my weight is down.”

    1. While the homicide rate is only down a small amount from 1964 and the assault rate is up a lot, I’d rather have fewer homicides and more assaults than the opposite.

  16. I blame violent video games and the Internet for the falling violent crime rate. Lots of hoodlums staying off the street.

Please to post comments

Comments are closed.