The RAND Corporation's Gun Policy in America initiative is a unique attempt to systematically and transparently assess available scientific evidence on the real effects of gun laws and policies. Our goal is to create resources where policymakers and the general public can access unbiased information that informs and enables the development of fair and effective policies. Good gun policies in the United States require consideration of many factors, including the law and constitutional rights, the interests of various stakeholder groups, and information about the likely effects of different policies on a range of outcomes. This study seeks to provide the third factor—objective information about what the scientific literature examining gun policies can tell us about the likely effects of those policies.
This study synthesizes the available scientific evidence on the effects of various gun policies on firearm deaths, violent crime, the gun industry, participation in hunting and sport shooting, and other outcomes.1 It builds and expands on earlier comprehensive reviews of scientific evidence on gun policy conducted more than a decade ago by the National Research Council (see National Research Council, 2004) and the Community Preventive Services Task Force (see Hahn et al., 2005).
We used Royal Society of Medicine guidelines for conducting systematic reviews of a scientific literature (Khan et al., 2003). We focused on the empirical literature assessing the effects of 13 classes of firearm policies or of the prevalence of firearms on any of eight outcomes, which include both public health outcomes and outcomes of concern to many gun owners. We reviewed scientific reports that have been published since 2003, a date chosen to capture studies conducted since the last major systematic reviews of the science of gun policy were published by the National Research Council (2004) and Hahn et al. (2005).
The 13 classes of gun policies considered in this research are as follows:
- background checks
- bans on the sale of assault weapons and high-capacity magazines
- stand-your-ground laws
- prohibitions associated with mental illness
- lost or stolen firearm reporting requirements
- licensing and permitting requirements
- firearm sales reporting and recording requirements
- child-access prevention laws
- surrender of firearms by prohibited possessors
- minimum age requirements
- concealed-carry laws
- waiting periods
- gun-free zones.
The eight outcomes considered in this research are
- violent crime
- unintentional injuries and deaths
- mass shootings
- officer-involved shootings
- defensive gun use
- hunting and recreation
- gun industry.2
Policy Analyses, by Outcome
Building on the earlier reviews (National Research Council, 2004; Hahn et al., 2005) and using standardized and explicit criteria for determining the strength of evidence that individual studies provide for the effects of gun policies, we produced research syntheses that describe the quality and findings of the best available scientific evidence. Each synthesis defines the class of policies being considered; presents and rates the available evidence; and describes what conclusions, if any, can be drawn about the policy's effects on outcomes.
In many cases, we were unable to identify any research that met our criteria for considering a study as providing minimally persuasive evidence for a policy's effects. Studies were excluded from this review if they offered only correlational evidence for a possible causal effect of the law, such as showing that states with a specific law had lower firearm suicides at a single point in time than states without the law. Correlations like these can occur for many reasons other than the effects of a single law, so this kind of evidence provides little information about the effects attributable to specific laws. We did not exclude studies on the basis of their findings, only on the basis of their methods for isolating causal effects. For studies that met our inclusion criteria, we summarize key findings and methodological weaknesses, when present, and provide our consensus judgment on the overall strength of the available scientific evidence. We did this by establishing the following relativistic scale describing the strength of available evidence:
- No studies. This designation was made when no studies meeting our inclusion criteria evaluated the policy's effect on the outcome.
- Inconclusive evidence. This designation was made when studies with comparable methodological rigor identified inconsistent evidence for the policy's effect on an outcome or when a single study found only uncertain or suggestive effects.
- Limited evidence. This designation was made when at least one study meeting our inclusion criteria and not otherwise compromised by serious methodological problems reported a significant effect of the policy on the outcome, even if other studies meeting our inclusion criteria identified only uncertain or suggestive evidence for the effect of the policy.
- Moderate evidence. This designation was made when two or more studies found significant effects in the same direction and contradictory evidence was not found in other studies with equivalent or strong methods.
- Supportive evidence. This designation was made when (1) at least three studies found suggestive or significant effects in the same direction using at least two independent data sets or (2) the effect was observed in a rigorous experimental study.
These ratings are meant to describe the relative strengths of evidence available across gun policy research domains, not any rating of our absolute confidence in the reported effects. For instance, when we find supportive evidence for the conclusion that child-access prevention laws reduce self-inflicted injuries and deaths, we do not mean to suggest that it is comparable to the evidence available in more-developed fields of social science. That is, in comparison to the evidence that smoking causes cancer, the evidence base in gun policy research is very limited. Nevertheless, we believe that it may be valuable to the public and to policymakers to understand which laws currently have more or less persuasive evidence concerning the effects the laws are likely to produce.
Table 1 summarizes our judgments for all policy and outcome pairings. Several outcomes show multiple judgments, and these correspond to different characterizations of the specific policy-outcome association. For instance, we identified limited evidence that background checks reduce total suicides and moderate evidence that they reduce firearm suicides.
Table 1. Strength of Evidence Across Gun Policies and Outcomes
|Background Checks||Bans on the Sale of Assault Weapons and High-Capacity Magazines||Stand-Your-Ground Laws||Prohibitions Associated with Mental Illness||Lost or Stolen Firearm Reporting Requirements||Licensing and Permitting Requirements||Firearm Sales Reporting and Recording Requirements||Child-Access Prevention Laws||Surrender of Firearms by Prohibited Possessors||Minimum Age Requirements||Concealed-Carry Laws||Waiting Periods||Gun-Free Zones|
|Purchasing||Possessing||Shall Issue||Permitless Carry|
|Total suicides||↓ L||I||↓ L||I||↓ L||I||I||I|
|Firearm suicides||↓ M||I||↓ L||I||↓ M||I||I|
|Firearm suicides among children||↓ L|
|Firearm self-injuries (nonfatal)||I|
|Firearm self-injuries (including suicides)||↓ S|
|Violent crime||↓ L||↓ M||I||I||↑ L||I|
|Total homicides||↓ L||I||↑ M||↓ L||I||I||I||I|
|Firearm homicides||↓ M, I*||I||↑ L||I||I||I||I||I||I|
|Intimate partner homicides||I||I|
|Other violent crime||I|
|Unintentional injuries and deaths|
|Unintentional firearm deaths||I|
|Unintentional firearm injuries and deaths among adults||↓ L|
|Unintentional firearm injuries and deaths among children||↓ S|
|Unintentional firearm injuries among adults||↑ L|
|Unintentional firearm injuries among children||I|
|Defensive gun use||I|
|Hunting and recreation|
|Prices of banned firearms in the short term||↑ L|
NOTE: I = inconclusive; L = limited; M = moderate; S = supportive. When we identified no studies meeting eligibility criteria, cells are light gray. ↑ = the policy increases the outcome; ↓ = the policy decreases the outcome.
* We concluded that there is moderate evidence that dealer background checks decrease firearm homicides, and there is inconclusive evidence for the effect of private-seller background checks on firearm homicides.
Rather than concerning how strong a policy's effects are, our findings concern the strength of the available scientific evidence examining those effects. Thus, even when the available evidence is limited, the actual effect of the policy may be strong. Presumably, every policy has some effect on a range of outcomes, however small or unintended. Until researchers design studies that can detect these effects, available evidence is likely to remain inconclusive or limited. But this fact should not be confused with the conclusion that the policies themselves have limited effects. They may or may not have the effects they were designed to produce; available scientific research cannot yet answer that question. Moreover, even a policy with a small effect may nevertheless be beneficial to society or worth its costs. For instance, a policy that reduces firearm deaths by just a few percentage points could save more than 1,000 lives per year. This kind of “small” effect might be very difficult to detect with existing study methods but could represent an important contribution to public health and safety.
The 13 types of policies reviewed in this study and the scope of the systematic review for the research synthesis were selected a priori and represent the central focus of our research synthesis efforts. Nevertheless, in reviewing evidence on these policies, other important themes emerged that the research team believed provided useful context for the policies or that were frequently cited in gun policy debates. Thus, we also researched what rigorous studies reveal about
- the possible mechanisms by which laws may affect outcomes
- how taxes, access to health care, and media campaigns might affect gun violence
- the effectiveness of laws used to target domestic violence
- methodological challenges in defining and estimating the prevalence of mass shootings and defensive gun use
- how suicide, violence, and mass shootings were affected by Australia's implementation of the National Firearms Agreement.
Conclusions and Recommendations
Of more than 100 combinations of policies and outcomes, we found that surprisingly few were the subject of methodologically rigorous investigation. Notably, research into four of our outcomes was essentially unavailable, with three of these four outcomes—defensive gun use, hunting and recreation, and the gun industry—representing issues of particular concern to gun owners or gun industry stakeholders. Here, we summarize the key conclusions and recommendations that can be drawn from the policy-outcome combinations with the strongest available evidence (conclusions 1 through 8). Thereafter, we draw conclusions and recommendations concerning how to improve evidence on the effects of gun policies (conclusions 9 through 13).
Conclusions and Recommendations Based on the Existing Evidence Base
Our first set of conclusions and recommendations describes the policy-outcome combinations with the strongest available evidence as identified through our review of the existing literature, as well as recommendations for policy based on this evidence.
Conclusion 1. Available evidence supports the conclusion that child-access prevention laws, or safe storage laws, reduce self-inflicted fatal or nonfatal firearm injuries among youth. There is moderate evidence that these laws reduce firearm suicides among youth and limited evidence that the laws reduce total (i.e., firearm and nonfirearm) suicides among youth.
Conclusion 2. Available evidence supports the conclusion that child-access prevention laws, or safe storage laws, reduce unintentional firearm injuries or unintentional firearm deaths among children. In addition, there is limited evidence that these laws may reduce unintentional firearm injuries among adults.
Recommendation 1. States without child-access prevention laws should consider adopting them as a strategy to reduce firearm suicides and unintentional firearm injuries and deaths. We note, however, that scientific research cannot, at present, address whether these laws might increase or decrease crime or rates of legal defensive gun use.
Recommendation 2. When considering adopting or refining child-access prevention laws, states should consider making child access to firearms a felony; there is some evidence that felony laws may have the greatest effects on unintentional firearm deaths.
Conclusion 3. There is moderate evidence that background checks reduce firearm suicides and firearm homicides, as well as limited evidence that these policies can reduce overall suicide and violent crime rates.
Conclusion 4. There is moderate evidence that stand-your-ground laws may increase state homicide rates and limited evidence that the laws increase firearm homicides in particular.
Conclusion 5. There is moderate evidence that laws prohibiting the purchase or possession of guns by individuals with some forms of mental illness reduce violent crime, and there is limited evidence that such laws reduce homicides in particular. There is also limited evidence these laws may reduce total suicides and firearm suicides.
Recommendation 3. States that currently do not require a background check investigating all types of mental health histories that lead to federal prohibitions on firearm purchase or possession should consider implementing robust mental illness checks, which appear to reduce rates of gun violence. The most robust procedures involve sharing data on all prohibited possessors with the National Instant Criminal Background Check System.
Conclusion 6. There is limited evidence that before implementation of a ban on the sale of assault weapons and high-capacity magazines, there is an increase in the sales and prices of the products that the ban will prohibit.
Conclusion 7. There is limited evidence that a minimum age of 21 for purchasing firearms may reduce firearm suicides among youth.
Conclusion 8. No studies meeting our inclusion criteria have examined required reporting of lost or stolen firearms, required reporting and recording of firearm sales, or gun-free zones.
Conclusions and Recommendations for Improving Gun Policy Research
Based on our review of the existing literature on the effects of firearm policy changes, we offer the following conclusions and recommendations for improving the evidence base on the effects of gun laws.
Conclusion 9. The modest growth in knowledge about the effects of gun policy over the past dozen years reflects, in part, the reluctance of the U.S. government to sponsor work in this area at levels comparable to its investment in other areas of public safety and health, such as transportation safety.
Recommendation 4. To improve understanding of the real effects of gun policies, Congress should consider whether to lift current restrictions in appropriations legislation, and the administration should invest in firearm research portfolios at the Centers for Disease Control and Prevention, the National Institutes of Health, and the National Institute of Justice at levels comparable to its current investment in other threats to public safety and health.
Recommendation 5. Given current limitations in the availability of federal support for gun policy research, private foundations should take further steps to help fill this funding gap by supporting efforts to improve and expand data collection and research on gun policies.
Conclusion 10. Research examining the effects of gun policies on officer-involved shootings, defensive gun use, hunting and recreation, and the gun industry is virtually nonexistent.
Recommendation 6. To improve understanding of outcomes of critical concern to many in gun policy debates, the U.S. government and private research sponsors should support research examining the effects of gun laws on a wider set of outcomes, including crime, defensive gun use, hunting and sport shooting, officer-involved shootings, and the gun industry.
Conclusion 11. The lack of data on gun ownership and availability and on guns in legal and illegal markets severely limits the quality of existing research.
Recommendation 7. To make important advances in understanding the effects of gun laws, the Centers for Disease Control and Prevention or another federal agency should resume collecting voluntarily provided survey data on gun ownership and use.
Recommendation 8. To foster a more robust research program on gun policy, Congress should consider whether to eliminate the restrictions it has imposed on the use of gun trace data for research purposes.
Conclusion 12. Crime and victimization monitoring systems are incomplete and not yet fulfilling their promise of supporting high-quality gun policy research in the areas we investigated.
Recommendation 9. To improve the quality of evidence used to evaluate gun policies, the National Violent Death Reporting System should be expanded to include all states with rigorous quality control standards.
Recommendation 10. The Bureau of Justice Statistics should examine the cost and feasibility of expanding its existing programs to generate state-level crime data.
Recommendation 11. The Bureau of Justice Statistics should continue to pursue its efforts to generate state-level victimization estimates. The current goal of generating such estimates for 22 states is a reasonable compromise between cost and the public's need for more-detailed information. However, the bureau should continue to expand its development of model-based victimization rates for all states and for a wider set of victimization experiences (including, for instance, crimes involving firearm use by an assailant or victim).
Conclusion 13. The methodological quality of research on firearms can be significantly improved.
Recommendation 12. As part of the Gun Policy in America initiative, we have published a database containing a subset of state gun laws from 1979 to 2016 (Cherney, Morral, and Schell, 2018). We ask that others with expertise on state gun laws help us improve the database by notifying us of its errors, proposing more-useful categorizations of laws, or submitting information on laws not yet incorporated into the database. With such help, we hope to make the database a resource beneficial to all analysts.
Recommendation 13. Researchers, reviewers, academics, and science reporters should expect new analyses of the effects of gun policies to improve on earlier studies by persuasively addressing the methodological limitations of earlier studies, including problems with statistical power, model overfitting, covariate selection, poorly calibrated standard errors, multiple testing, undisclosed state variation in law implementation, unjustified assumptions about the time course of each policy's effects, the use of spline and hybrid effect codings that do not reveal coherent causal effect estimates, and inadequate attention to threats of reciprocal causation and simultaneity bias.
In conclusion, with a few exceptions, there is a surprisingly limited base of rigorous scientific evidence concerning the effects of many commonly discussed gun policies. This does not mean that these policies are ineffective; they might well be quite effective. Instead, it reflects shortcomings in the contributions that scientific study can currently offer to policy debates in these areas. It also reflects, in part, the policies we chose to investigate, all of which have been implemented in some U.S. states and, therefore, have proven to be politically and legally feasible, at least in some states. This decision meant that none of the policies we examined would dramatically increase or decrease the stock of guns or gun ownership rates in ways that would produce more readily detectable effects on public safety, health, and industry outcomes. The United States has a large stock of privately owned guns in circulation—estimated in 2014 to be somewhere between 200 million and 300 million firearms (Cook and Goss, 2014). Laws designed to change who may buy new weapons, what weapons they may buy, or how gun sales occur will predictably have only a small effect on, for example, homicides or participation in sport shooting, which are affected much more by the existing stock of firearms. Although small effects are especially difficult to identify with the statistical methods common in this field, they may be important. Even a 1-percent reduction in homicides corresponds to more than 1,500 fewer deaths over a decade.
By highlighting where scientific evidence is accumulating, we hope to build consensus around a shared set of facts that have been established through a transparent, nonpartisan, and impartial review process. In so doing, we also mean to highlight areas where more and better information could make important contributions to establishing fair and effective gun policies.