Universal Background Checks and Firearm Homicide Rates Study Notes
Abstract
Importance: A loophole in US gun policy permits gun purchases from private sellers without background checks. Some states have enacted universal background checks for all gun sales, either at the point of sale or through a permit system. Most studies examining the effectiveness of universal background checks have not analyzed these two policy mechanisms separately.
Objective: Assess the association of point-of-sale background check laws and gun permit laws, separately, with firearm homicide rates from 1976 to 2022.
Design: Cross-sectional study employing a difference-in-differences, fixed-effects regression model of firearm laws and homicide rates in 48 states (excluding New Hampshire and Vermont) from 1976-2022.
Participants: Data was analyzed for 48 states and included laws requiring a universal background check at the point of sale or laws requiring a permit for gun purchases.
Main Outcomes: Annual state-specific rates of firearm homicide per 100,000 people.
Results:
12 states adopted universal background check laws without permitting requirements.
7 states implemented gun permit laws for all firearms.
Mean firearm homicide rate was 4.3 per 100,000 people.
Universal background checks alone were not associated with homicide rates.
Permit laws were correlated with significantly lower homicide rates (overall: -15.4%, firearm: -18.3%).
Conclusions: Universal background checks are insufficient alone; combining them with permits correlates with reduced firearm homicide.
Key Points
Question: Are laws for universal background checks and/or permit requirements associated with reduced firearm homicide rates in the US?
Findings: In a study of 48 states, permit requirements significantly reduced firearm homicide rates, whereas universal background checks alone did not.
Meaning: Effective reduction in firearm violence necessitates pairing universal background check laws with permit requirements for all firearm purchases.
Introduction
Loophole in US Gun Policy: Allows purchases from private sellers without background checks.
Legislative Efforts: States mandating universal background checks aim to reduce gun violence.
Impact Evidence: State laws requiring universal background checks correlate with decreased firearm homicide rates.
Mechanisms of Background Checks:
Point-of-sale checks.
Permit system checks prior to purchase.
Research Limitation: Few studies assess the separate impact of these methods on homicide rates.
Previous Findings: Two prior studies indicated point-of-sale checks linked to increased homicide rates, while permit laws correlated with lower rates.
Methods
Study Design: Cross-sectional using panel data of firearm violence rates and firearm laws across 48 states from 1976 to 2022.
Ethics: Exempt from institutional review as it used deidentified datasets.
Statistical Model: Difference-in-differences, fixed-effects linear regression examining the impact of laws on homicide outcomes:
Where:
$ln(fhst)$: Natural log of homicide rate in state s at time t.
$Bst$: Indicator for point-of-sale background check law.
$Pst$: Indicator for permit-to-purchase law.
$Xst$: Control variables.
$yt$: Year fixed effects.
$zs$: State fixed effects.
$est$: Error term.
Control Variables: Population size, density, violent crime rate, property crime rate, suicide rate, demographic data (e.g., racial composition), poverty, and unemployment rates were included to mitigate confounding.
Sample
States Studied: 48 total, with 10 having only point-of-sale checks and 7 with permit laws by 2022.
Data Sources:
CDC WISQARS: Provides annual homicide rates.
Uniform Crime Reports: Provided crime statistics.
Census and Surveys: Provided demographic and economic data.
Statistical Analysis
Modeling Approach: Difference-in-differences with robust standard errors, adjusted for serial autocorrelation and heteroskedasticity using Prais-Winsten regression in Stata.
Durbin-Watson Statistic: Analyzed for evidence of serial autocorrelation, confirming model adequacy post-adjustment.
Interpretation of Results: Exponentiated coefficients calculated from log-transformed data represent percentage change in homicide rates attributed to law enactments.
Results
Descriptive Statistics:
Age-adjusted firearm homicide rates varied, from 1.0 in Maine to 18.5 in Mississippi.
Mean rate was 4.3 per 100,000.
Notable trends showed firearm homicide rates fluctuating significantly over the study period.
Regression Findings:
Point-of-sale checks: No significant association with overall homicide rates (1.3%) or firearm homicide rates (3.7%).
Permit laws: Associated with significant reductions (overall: -15.4%, firearm: -18.3%).
Control variables: Race and crime rates correlated with higher homicide rates, while increased incarceration rates correlated with lower rates.
Discussion
Implications: Findings indicate the ineffectiveness of point-of-sale background checks alone in reducing firearm violence, emphasizing the necessity of accompanying permit laws.
Internal Validity: Utilization of a difference-in-differences model enhances internal validity and accommodates potential confounding factors.
Reasons for Effectiveness: Permit laws likely necessitate more comprehensive background checks, engaging law enforcement directly, unlike point-of-sale checks.
Limitations:
Potential reverse causation influencing the adoption of permit laws could falsely imply efficacy.
Study limited to firearm homicide rates, not examining firearm suicide rates where correlations may exist.
Conclusions
Universal background checks alone lack sufficient association with firearm homicide rates. The combination of background checks and permit requirements shows promise in effectively reducing gun violence fatalities.
Article Information
Publication Date: August 1, 2024.
Authors: Michael Siegel, MD, MPH, Tufts University School of Medicine.
Conflict of Interest: Acknowledged grants received from various organizations.
References
Comprehensive list of studies examining firearm-related injuries, mortality rates, and legislative impact on firearm violence are included in this section.