Standing Your Ground

A new NBER paper examines the effect of Stand Your Ground self-defense laws, which “eliminate the longstanding legal requirement that a person threatened outside of his or her own home retreat rather than use force.”  Chandler B. McClellan and Erdal Tekin exploit cross-state variations in implementation dates to determine the effect of these laws on homicides.  Their findings are grim: 

Our results indicate that Stand Your Ground laws are associated with a significant increase in the number of homicides among whites, especially white males.  According to our estimates, between 4.4 and 7.4 additional white males are killed each month as a result of these laws.  We find no evidence to suggest that these laws increase homicides among blacks.  Our results are robust to a number of specifications and unlikely to be driven entirely by the killings of assailants.

In the paper, the authors add that:

It is possible that some of the additional homicides associated with the SYG law may indeed be driven by the homicides of assailants.  However, note that the net increase in homicides cannot be accounted by a one-to-one substitution between killings of assailants and killings of innocent individuals unless multiple assailants are killed in some instances (Cheng and Hoekstra, 2012).  If at least some of the additional homicides are due to individuals resorting to the use of deadly force against each other in situations where the threat of death or serious bodily injury is not imminent to either party, this could indicate that these laws impose serious costs not only on criminals both also private citizens as well. It is also possible that potential intruders and attackers, who, in the absence of SYG laws, could have gotten away with their crimes without killing anyone, are now killed as a result of these laws.

Eighteen states have passed Stand Your Ground laws since 2005.



One reason there is so little resistance to pro-gun laws in the US is that the vast number of guns hasn't caused an appreciable spurt in crime. There seems little difference in crime between areas that have stricter or looser gun laws. But these stand your ground laws do seem to cause more gun use, which means they're bad.


I don't have any figures handy, but the gun-related murder rate is invariable lower in almost every Canadian city than almost every American city, per capita. Canada's big and small cities have very similar poverty problems to American ones, so the big difference seems to be the lack of gun control in the US. So it would seem that more gun control actually does reduce gun-related deaths.

Enter your name...

It's not just a "number of guns" or "gun regulation" issue.

Switzerland has half as many guns per adult compared to the US, but the have far fewer BB rifles (which aren't likely to kill anyone) and far more military-grade weapons. They also have almost half as many guns per household, with most households having a gun and most of them that have a gun having exactly one (their militia weapon). In the US, most households have zero guns, but those that do have gun commonly having multiple firearms (e.g., a shotgun, a deer rifle, a handgun, and a BB rifle for the kids).

So the Swiss have more access to firearms (more people have one in the home), but their rate of firearm homicide is one-sixth of the US's. I think there s a cultural component.

Steven Gangstead

Live free or die.

Mike B

While I disagree with the notion of being required to retreat in the face of a threat, SYG laws need to be tightened to discourage the use of force unless absolutely necessary. For example a person should actually be under threat of attack or attacked before being allowed to use force and they should be required to reasonably prove such a threat or attack took place. Also if such a threat goes away all use of force must stop. Also, as we saw in the Florida situation, one should not get a free pass if they themselves provoke the threat or attack.

However improved SYG laws can be worded will take careful thought, but following situations need to be clearly out of bounds.

* Harming an attacker who flees, surrenders or is otherwise no longer a threat. (Shoot 'em in the back)
* Initiating physical violence where the imminent threat of violence has not yet been established. (Shoot first and ask questions later)
* Any situation where one can induce another to trigger an SYG response. (Looking for trouble)

SYG fanatics will often bring up situations involving the home and finding someone sneaking around one's home at night and that's a legitimate concern. More leeway should be given when one is in an enclosed private space like a home (or car or tent), but in a non-enclosed area, even on one's own private property, the standard needs to be higher. Citizens need to be encouraged to deescalate potentially violent confrontations or summon trained law enforcement authorities. For example, if a crazy neighbor comes on your property because they are mad, that should not automatically empower a homeowner to use deadly force. The United States is not Iraq, we have mechanisms in place to handle that kind of thing.


Some Random Economist

Please provide one example of a SYG law in any state that allows someone to "shoot first and ask questions later".

Some Random Economist

At least this paper weights regressions by state population and somewhat acknowledges common law as affecting the treatment (another recent paper on the topic did neither); however, they appear not to understand how far common law can go in some states. Virginia, for example, has no "stand your ground" law on the books; but the courts have provided strong enough protection that gun-rights groups in the state have helped defeat SYG bills that they felt would lessen their protection under the law. My understanding is that many of the states that have SYG bills now had similar protections in common law before their passage.

It's not clear to me that there is a well-defined, exogenous treatment in this case.


One (maybe) uninteneded consequence of these laws that I noticed during the Trayvon Martin saga was the potential for a situation where it is perfectly legal for EITHER of two parties to kill each other.

For example, two men go to a place where it is legal for them to have concealed weapons. They start talking. The talk turns to politics. The talk escalates into an argument which escalates into a full-blown fight each going back and forth with progressively more severe threats/physical contact. If the standard for stand your ground laws is to feel reasonably "threatened" or "in danger", couldn't either man justifiably kill the other? How would you determine when the discourse went from acceptable to someone instigating the fight?

And is this "wild west" scenario something the voters thought about and were in favor of?

alex in chicago

Not true. Instigators lose the ability to claim a SYG defense unless they have actively retreated from the scenario and been re-engaged by the other party (for instance, you run out of the barfight to your car, and get chased as you are jangling you keys trying to open the door).


True. Understanding the "meat and potatoes" of the law helps instead of just reading blogs about it. Instead of only reading articles and blogs, anyone who is interested would be well served to find the SYG law in their own state, or in a neighboring state if their own state does not have SYG, and do some homework. Debate doesn't help too much, unless it's informed debate.


How do the authors account for reduced victim deaths as a result of the laws? Or do they look at only one side of the safety ledger?

The key logical gaps in the authors' analysis appear equally problematic. Obviously, permitting potential victims to confront their assailants will produce more deaths, but when the authors themselves concede they cannot determine what percentage of those killed were assailants, one can draw no conclusion regarding the laws' efficacy or impact on safety from an increase in deaths. Note two logical errors the authors employ to avoid that reality when they claim results are "unlikely" to be driven "entirely" by the killing of assailants: 1) They believe the result is "unlikely" because "multiple assailants" would need to be killed by putative victims in some circumstances, yet they offer no reason why that result is unlikely. 2) Whether the increase is "entirely" driven by the killing of assailants is irrelevant to the underlying question, of course. If the increase is predominately driven by the killing of assailants, one could easily argue the laws are effective and increase safety.

The only thing that's grim, frankly, is the sloppy "empirical" research some people are using to justify their desired public policy goals.


Seminymous Coward

The other obvious question along these lines is: how many rapes have been halted by shooting the assailants? I, for one, don't think that such incidents causing an increase in homicides is all that regrettable.

Analogous questions are obvious for other crimes, though it's obviously a sliding scale. For example, a society might be willing to trade a few carjackings for less homicides, even justified ones.


Absolutely. But not just rapes are prevented, other crimes too. And not just from shooting the assailant. Many crimes are prevented just by the victim producing a weapon. We just don't hear about them on the news.


"It is also possible that potential intruders and attackers, who, in the absence of SYG laws, could have gotten away with their crimes without killing anyone, are now killed as a result of these laws."

Isn't that exactly the intent of SYG laws, to raise the potential cost of such crimes to the criminals, while decreasing it to the intended victims?


The fact that the above paper is unable to distinguish between assailant death (justified homicide) and unjustified homicide (murder, manslaughter, etc) greatly limits its utility at determining the effects of SYG laws. As has been pointed out, increasing the costs to criminals of criminal behavior is a likely rationale for the existence of SYG laws. It should be apparent to any resident of a major US city that fear of the criminal justice system is an insufficient deterrent to violent (and recidivist) criminals.


1) These studies are relatively new. I hope that there are links to any followup ones that explore the statistical validity of the studies.

2) Also I hope that there is some attempt to determine the different types of people were killed in SYG states than others. If there were a lot of potential murders, rapists... killed in SYG states where as so far only a few victims were saved, then one might see a long term benefit even if in the short term there are more SYG deaths.


This whole little "paper" is a joke. What criteria, exactly, was used? And for what time period? And what, praytell, are these "robust specifications"? These Heroes had their "answer" before they ever "researched" anything.


I didn't see anything here that actually takes away from SYG. Both paragraphs of SYG working.

SYG is MEANT to escalate the level of homicides against assailants, including assailants in situations that would not have resulted in a death prior to SYG.

For me, it's proof that SYG is working.

Loren Pechtel

Justifiable homicides could legitimately increase with SYG laws without any changes to the number of dead victims. Note that there is a case where SYG permits killing in a situation that otherwise wouldn't have become violent:

Bad guy: "You're not welcome here. Leave before I kill you." (Due to the situation you believe the threat is real.)

no SYG: You of course run. That was their objective, no violence results.

SYG: You have the option of shooting instead.

Note that this isn't only an issue with gangs, domestic cases can go down this way, also. A couple, living together, breaking up, the owner of the property tries to throw the other out with threats of violence.