This paper is available on-line at the Chicago Working Paper Series index at:
http://www.law.uchicago.edu/Publications/Working/index.html
This paper is also available from the Social Science Research Network at:
http://papers.ssrn.com/paper.taf?abstract_id=161637
CHICAGO
JOHN M. OLIN LAW & ECONOMICS WORKING PAPER NO. 73
(2D SERIES)
Multiple Victim Public Shootings, Bombings, and
Right-to-Carry Concealed Handgun Laws:
Contrasting Private and Public Law Enforcement
John R. Lott, Jr. and William M. Landes
THE LAW SCHOOL
THE UNIVERSITY OF CHICAGO
Multiple Victim Public Shootings, Bombings,
and Right-to-Carry Concealed Handgun Laws:
Contrasting Private and Public
Law Enforcement
John R. Lott, Jr. and William M. Landes
1
I. Introduction
Few events obtain the same instant national and even
worldwide news coverage as when several people are shot and killed
in a public place. The worst examples in the United States come
readily to mind: Colin Ferguson killed 6 people during his rampage
on the Long Island (NY) Railroad in December 1993; 22 people
were killed during the October 1991 shooting in Luby’s Cafeteria in
Killeen, Texas; 5 persons died at the Cleveland Elementary School
in Stockton, California in January 1989; and 21 persons were killed
at the San Ysidro, California McDonald’s in July, 1984. Shootings
by disgruntled post office employees, has led to the phrase “going
postal” entering our language. And with the recent public school
shootings, a sense of urgency has also been added to the discussion.
The most common suggestion for reducing the incidence of
public shootings (the term we use to denote shootings in public
places where two or more individuals are killed or injured) calls for
greater regulation of guns. For example, recent public shootings in
Tasmania, Australia and Dunblane, Scotland have lead to strict gun
1
Lott is John M. Olin Fellow in Law and Economics and Landes is Clifton
R. Musser Professor of Law & Economics at the University of Chicago Law
School. We would like to thank Edward Glaeser, Mitch Polinsky and
participants in seminars at the Arizona State University, Auburn University,
University of Chicago, Claremont Graduate School, University of Houston,
University of Illinois, University of Kansas, University of Miami, New York
University, University of Oklahoma, Rice University, University of Texas at
Austin, University of Texas at Dallas, William and Mary University, and
Yeshiva University School of Law as well as participants at the Economics of
Law Enforcement Conference at Harvard Law School, Association of
American Law Schools Meetings, American Economic Association Meetings,
Midwestern Economic Association Meetings, Southern Economic Association
Meetings, and Western Economic Association Meetings.
Chicago Law and Economics Working Paper 2
prohibitions in those countries. And after a 1997 shooting at the top
of the Empire State Building in which one person was killed, the
Mayor of New York called for national gun licensing laws. Other
laws restricting access to guns, such as waiting periods, are often
justified as producing a cooling off period to prevent shooting sprees.
Yet, the response to these shootings has not been uniform. In Texas
and other states, multiple shooting incidents may have helped pave
the way for passage of concealed handgun laws permitting law-
abiding citizens to carry handguns. Terrorist shootings in Israel have
lead to wider licensing of citizens to carry concealed handguns. (In
this paper, we use the term “shall issue law” or “right-to-carry” to
denote a state law that sets up objective criteria for a law-abiding
citizen to obtain a permit to carry a concealed handgun.)
Not surprisingly, those opposed to concealed handgun laws
point to the loss of life and injuries that result from these shootings.
Their argument is straightforward: “If you introduce a gun into a
violent encounter, it increases the chance that someone will die.”
2
Since a large number of murders may arise from fits of rage that are
quickly regretted, keeping guns out of people’s reach (even
temporarily) might prevent deaths in many instances.
3
Shootings in
public places may be the most visible manifestation of individuals
who might have refrained from such acts but for having access to
guns. For example, in the recent rash of school shootings in 1997
and 1998, the perpetrators obtained their guns from relatives or
neighbors. Had no guns been accessible, the acts may not have been
committed.
2
Philip Cook quoted in Editorial, Cincinnati Enquirer, Jan. 23, 1996, A8.
Others share this belief. "It's common sense," says Doug Weil, research
director at the Center to Prevent Handgun Violence, in Washington. "The
more guns people are carrying, the more likely it is that ordinary
confrontations will escalate into violent confrontations" (William Tucker,
“Maybe You Should Carry A Handgun,” The Weekly Standard, Dec. 16, 1996,
p. 30).
3
See P. J. Cook, “The Role of Firearms in Violent Crime,” in M.E.
Wolfgang and N.A. Werner, eds., Criminal Violence, Sage Publishers:
Newbury, N.J.(1982) and Franklin Zimring, “The Medium is the Message:
Firearm Caliber as a Determinant of Death from Assault,” Journal Legal
Studies, 1 (1972) for these arguments.
3 Multiple Victim Public Shootings
In contrast, proponents of concealed handgun laws point to the
potential use of guns for defensive purposes. They argue that the
prospect of encountering a victim who is armed may deter a criminal
from an attack in the first place. National polls indicate that people
use guns defensively against criminal attacks somewhere between
760,000 and 3.5 million times per year.
4
Data from the Department
of Justice’s National Crime Victimization Survey from 1979 to 1987
indicate that the risk of serious injury from a criminal attack is 2.5
times greater for women offering no resistance than for women
resisting with a gun (Southwick, 1996).
5
The most comprehensive
study of laws permitting individuals to carry concealed weapons
indicates that these laws reduce murder rates by about 10 percent,
with similar declines in other violent crimes (Lott 1998b but see
related comments by Bartley et. al., 1998; Black and Nagin, 1998;
Bronars and Lott, 1998; Plassman and Tideman, 1998; Lott and
Mustard, 1997; and Lott, 1998a). And contrary to a popular
misconception, the use of concealed handguns by permit holders are
virtually never involved in the commission of crime, let alone murder
(Lott, 1998b).
6
Just as advocates of greater regulation of guns point to shooting
sprees that kill or maim many individuals as evidence supporting
their position, opponents point to anecdotal evidence supporting
more permissive handgun laws. During the Luby’s Cafeteria
shooting one of the restaurant’s patrons, an expert marksman, had
left her handgun in her car to comply with the Texas state law
existing at the time. In an incident in 1997, a gunman in a Florida
restaurant was prevented from shooting people by the quick
4
Kleck, Gary, and Marc Gertz, “Armed Resistance to Crime: The Prevalence
and Nature of Self-Defense with a Gun,” 86 Journal of Criminal Law and
Criminology 86 (Fall 1995).
5
There are problems with the National Crime Victimization Survey both in
terms of its nonrepresentative sample (for example, it weights too heavily urban
and minority populations) and its failure to adjust for many people not
admitting to a law enforcement agency that they used a gun, even defensively.
Unfortunately, this survey provides the only available evidence how the
probability of significant injury varies with level and type of resistance.
6
We add that no data are available on whether handguns lawfully bought by
permit holders are used in crimes by another party at a later date.
Chicago Law and Economics Working Paper 4
reactions of two people with concealed handguns.
7
(See Lott
(1998b) for a more complete list of such cases.) Internationally,
similar examples can be found. On March 13, 1997, a Jordanian
soldier shot seven young Israeli girls to death while they were visiting
Jordan’s so-called “Island of Peace.” According to newspaper reports,
the Israelis had “complied with Jordanian requests to leave their
weapons behind when they entered the border enclave. Otherwise,
they might have been able to stop the shooting, several parents
said.”
8
Referring to the July 1984 massacre at a San Ysidro, California,
McDonald’s restaurant, Israeli criminologist Abraham Tennenbaum
wrote that:
what occurred at a [crowded venue in] Jerusalem some
weeks before the California McDonald's massacre: three
terrorists who attempted to machine-gun the throng
managed to kill only one victim before being shot down
by handgun carrying Israelis. Presented to the press the
next day, the surviving terrorist complained that his
group had not realized that Israeli civilians were armed.
The terrorists had planned to machine-gun a succession
of crowd spots, thinking that they would be able to escape
before the police or army could arrive to deal with them.
9
Obviously allowing Israeli citizens to carry concealed handguns
has not eliminated terrorist attacks. Indeed, terrorists may well have
responded to the difficulty of successfully shooting civilians in public
places by substituting bombings, which allow potential victims little
chance to respond.
Anecdotal evidence cannot resolve the question whether
allowing persons to carry concealed handguns will save or cost lives.
In this study, we provide a systematic empirical analysis of the effects
of different gun laws on multiple victim public shootings. We focus
7
Allison Thompson, Robber Gets Outgunned on Westside,” Florida Times-
Union ( Jacksonville, FL),September 24, 1997, p. B1.
8
Rebecca Trounson, “Anxiety, Anger Surround Return of Young Survivors,”
Los Angeles Times, March 14, 1997, p. A1
9
Baltimore Sun, Oct. 26, 1991. As referenced in an article by Don Kates and
Dan Polsby. “Of Genocide and Disarmament,” Journal of Criminal Law and
Criminology, 86 (Fall 1995): 252.
5 Multiple Victim Public Shootings
primarily on nondiscretionary (or “shall issue”) concealed handgun
laws although we also look at evidence on the role of waiting periods
and additional penalties imposed on individuals who use guns in the
commission of a crime.
We analyze multiple victim public shootings (hereafter, multiple
shootings or killings) in the United States in the period 1977 to
1995.
10
The main advantage of restricting our study to U.S. data is
that we can compare states with and without shall issue laws at
different points in time (other things constant) and, therefore, can
estimate the effects of a change in the law within a state during the
sample period. In contrast, time series data for a single country faces
the problem that many different events may occur at around the
same time which makes it difficult to disentangle the impact of a
change in the law from other factors. And an international study
across different countries faces problems in finding comparable data
on gun laws, crime rates, and gun ownership. Our study ends in
1995 because many of the variables we use are not available after
than date.
A few words about why we study multiple shootings. Aside
from the public attention these acts receive, multiple shootings allow
us to test the applicability of economics to an area believed to be far
outside the domain of economics. Perpetrators of these acts are often
thought to be psychotic or irrational and hence not responsive to
costs and benefits. Thus, legal sanctions or, as in this case, the
prospect of encountering an armed individual during a shooting
spree would have no deterrent effect on these individuals. Indeed,
the act itself is cited as powerful evidence of irrational or psychotic
behavior since a sane person would never kill helpless victims in a
10
While the recent rash of public school shootings during the 1997-98 school
took place after the period of our study, these incidents raise questions about the
unintentional consequences of laws. The five public school shootings took
place after a 1995 federal law banned guns (including permitted concealed
handguns) within a thousand feet of a school. The possibility exists that
attempts to outlaw guns from schools, no matter how well meaning, may have
produced perverse effects. It is interesting to note that during the 1977 to 1995
period, 15 shootings took place in schools in states without right-to-carry laws
and only one took place in a state with this type of law. There were 19 deaths
and 97 injuries in states without the law, while there was one death and two
injuries in states with the law.
Chicago Law and Economics Working Paper 6
public place. The claim is then made that a law permitting
individuals to carry concealed weapons couldn’t possibly deter
shooting sprees in public places (though it might reduce the number
of people killed or wounded). And since these laws might well
increase the availability of guns to potential perpetrators, the
combination of criminal irrationality and greater availability of guns
might lead to an increase in the number of incidents of multiple
shootings.
In contrast, the economic model of crime predicts that a shall
issue law will raise the potential perpetrator’s cost of acting (e.g., he
might be wounded or killed if he acts) and lower his expected benefit
(e.g., he will do less damage if he encounters armed resistance).
Although not all offenders will alter their behavior in response to a
shall issue law, some individuals will be deterred from carrying out a
shooting spree because the resulting changes in costs and benefits
will be sufficiently large to make their net gain from acting negative.
How large the deterrent effect is depends on how many potential
offenders are close enough to the margin so that the passage of a
shall issue law changes the net benefit from positive to negative.
Economics predicts, therefore, that shall issue laws will reduce the
number of mass shootings (subject to the “greater availability of
guns” qualification noted above) though the magnitude of this
response is uncertain.
A study of multiple shootings also allows us to compare
whether a shall issue law will produce a bigger deterrent effect on
multiple shootings than on ordinary murders and other crimes. This
may appear surprising in light of the claimed irrationality of
individuals who go on shooting sprees. But another consideration
points in the opposite direction. Suppose a shall issue law deters
crime primarily by raising the probability that a perpetrator will
encounter a potential victim who is armed. In a single victim crime,
this probability is likely to be very low. Hence the deterrent effect of
the law—though negative—might be relatively small. Now consider
a shooting spree in a public place. The likelihood that one or more
potential victims or bystanders are armed would be very large even
though the probability that any particular individual is armed is very
7 Multiple Victim Public Shootings
low.
11
In principle, this suggests a testable hypothesis: a shall issue
law will have a bigger deterrent effect on shooting sprees in public
places than on more conventional crimes. Finally, because the
presence of citizens with concealed handguns may be able to stop
attacks before the police are able to arrive, our data also allows us to
provide the first evidence that we know of concerning whether the
severity of those crimes that still take place can be mitigated.
II. Multiple Victim Public Shootings: A First Look
We define a multiple shooting as one in which two or more
people are killed or wounded in a public place. “Public place” refers to
a shooting that occurred in a church, business, bar, street,
government building, public transit, place of employment, park,
health care facility, mall or restaurant. We collected data on these
shootings from articles reporting multiple shootings during the 1977
to 1995 period in the Lexis/Nexis computerized database. Our
sample does not include all multiple shootings in the database. We
exclude the following: those that were the byproduct of another
crime (e.g., a robbery or drug deal); shootings that involved gang
activity (e.g., drive by shootings); professional hits or shootings
related to organized crime; and serial killings or killings that took
place over the span of more than one day.
12
Although there is no
bright line between different types of multiple shootings, there are
several reasons for limiting our sample as we do.
1. One of us has already studied the effects of shall issue laws on
murder rates in general. Hence by adding shootings from robberies,
11
To illustrate, let the probability (p) that a single individual is carrying a
concealed handgun equal .10. Assume further that there are 10 individuals in a
public place. Then the probability that at least one of them is armed is about .65
(= 1 – (.9)
10
).
12
In a recent paper (see T. Petee, K. Padgett and T. York, Debunking the
Stereotype: An Examination of Mass Murder in Public Places, 1 Homicide
Studies 317 (1997)) the authors find felony related mass murders account for 36
percent and gang motivated 5.8 percent of mass murder incidents over the 1965
to 1995 period. That study defines mass murders as the killing three or more
persons (so it has much fewer incidents than our sample).
Chicago Law and Economics Working Paper 8
gang activity and so forth to our sample we risk duplicating earlier
work.
2. Shall issue laws permit law-abiding citizens to carry guns so
these laws should have little impact on killings related to gang
activity, drug deals and organized crime. Individuals involved in these
activities are already engaged in unlawful activities that often require
them to carry guns. In short, these persons carry guns independent
of whether the law permits them to do so. Hence a “shall issue”
should have little effect on their behavior.
3. Restricting our study to shootings in public places allows us
to concentrate on places where economic theory suggests that shall
issue laws will have their greatest effect.
13
We expect this because
more citizens will be carrying concealed handguns in public places
after a law allowing them to carry a gun has been passed. And a shall
issue law should also be a more effective deterrent if the potential
perpetrator either knows or is uncertain whether the intended victim
is armed. Conversely the law should have no effect if the offender
knows in advance that the victim is not armed. But such knowledge
is unlikely for public places. So unless there are explicit restrictions on
carrying guns in certain places (e.g., near schools), a shall issue law
should increase the likelihood that a potential victim or bystander is
armed.
14
4. One can also question our definition of multiple shootings as
requiring two or more killings or injuries. Later in the paper we look
13
Alschuler (1997, p. 369) claims that concealed handguns should only deter
crimes involving strangers. Our response is that concealed handguns can deter
crimes involving acquaintances as well as strangers, though deterrence
involving acquaintances might be more easily thought of as similar to open
carrying of guns. The big effect of concealed handguns is that they may allow
people to be able to now defend themselves outside of their home or business.
The passage of the concealed handgun laws may deter crimes against
acquaintances simply to the extent to which it increases gun ownership.
14
Most states allow private businesses to decide whether permit holders are
allowed to carry concealed handguns on their premises. State rules may also
vary with regard to other places such as government buildings, churches, and
bars.
9 Multiple Victim Public Shootings
at alternative definitions requiring greater number of deaths and
injuries.
Tables 1 and 2 present data on multiple shootings for the
United States as a whole, and for states with and without shall issue
laws. We find that states without shall issue laws had more deaths
and injuries from multiple shootings per year (both in absolute
numbers and on a per capita basis) during the 1977 to 1995 period.
Note also that the number of states without shall issue laws declined
from 43 to 29 and the percentage of the U.S. population in these
states fell from 91.5 to 68 percent in this period. Yet states without
shall issue laws still account for the overwhelming majority (often
over 90 percent) of deaths and injuries. The different rates of
shootings, murders, or injuries are very consistent over time and do
not arise from a few unusual years.
Tables 3 and 4 look more closely at the 14 states that adopted
shall issue laws between 1977 and 1995.
15
(No state has ever
repealed this law.) Table 3 shows a sharp drop in multiple murders
and injuries per 100,000 persons after the passage of a shall issue law.
Murders fell by 89 percent and injuries by 82 percent. Table 4
indicates that this drop occurred largely during the first full year after
a state enacted its law (year “1” in the first column). Overall, the
decline is so large that we observe zero multiple killings in three of
the eight years after the passage of a law, an event that did not occur
15
The fourteen states that enacted “shall issue” or “right-to-carry” laws in the
1977 to 1995 period (dates in parentheses) are as follows: Alaska (1994),
Arizona (1994), Florida (1987), Georgia (1989), Idaho (1990), Maine (1985),
Mississippi (1990), Montana (1991), Oregon (1990), Pennsylvania (1989),
Tennessee (1994), Virginia (1988), West Virginia (1989), and Wyoming
(1994). The following eight states had “shall issue” laws over the entire period:
Alabama, Connecticut, Indiana, New Hampshire, North Dakota, South
Dakota, Vermont and Washington. Data on states having laws prior to 1993
are from Clayton E. Cramer and David B. Kopel, Shall Issue: The New Wave
of Concealed Handgun Permit Laws, 62 Tennessee Law Review, 679 (1995).
We used a Nexis search to determine the state and date for states passing laws
between 1993 and 1995. These two sources were also used in Lott and Mustard
(1997). Because of objections raised to the dates for “shall issue” laws in Maine
and Virginia (see the discussion in Lott and Mustard), the regression analysis
presented in part III examines the sensitivity of our findings to alternative
dates for Maine and Virginia.
Chicago Law and Economics Working Paper 10
during any year before passage of the law.
16
Table 3 also presents
data on a narrower sample consisting of multiple shootings that
appeared in the first section of the New York Times at the time the
shootings took place. We use this sample as an estimate of more
serious or, at least, more notorious multiple shootings. Similar to the
data on all multiple shootings, the New York Times data show a
decline of 91 percent in the rate of multiple shootings after a state
adopted a shall issue law. Finally, we consider the possibility that
shall issue laws lead criminals to substitute bombings for shootings.
Data on bombings (from the Bureau of Alcohol, Tobacco and
Firearms annual publication Arson and Explosives: Incidents
Report) suggest no systematic impact on the number of bombings.
After the passage of shall issue laws, actual and attempted bombings
increased slightly, incendiary bombings fell and other bomb-related
incidents (involving stolen explosives, threats to treasury facilities,
and hoax devices) declined compared to the before law period.
III. Accounting for Other Factors
Although the above tables suggest that shall issue laws reduce
mass shootings, other factors may explain these changes. To take
account of this possibility, we estimated regression equations with
the following state specific variables: the arrest rate for murder; real
per capita personal income; real per capita government payments for
income maintenance, unemployment insurance and retirement; the
unemployment rate; the poverty rate; state population and
population squared; and a set of demographic variables that subdivide
a state’s population into 36 different race, sex, and age groups. We
also include year and state specific dummy variables in the regression
analysis. Table 5 lists the variables included in the regression analysis.
Thus our results hold constant both the effects of any national
trends and state-specific effects on multiple shootings which may
coincide with the adoption of shall issue laws. So, for example, if the
multiple shooting rate declines nationally between two years, the
regression coefficient on the law variable tests for whether the
decline is relatively greater in states that adopted shall issue laws
16
Of course, there were zero mass shootings in individual states in particular
years before the passage of concealed handgun laws.
11 Multiple Victim Public Shootings
during the two year period. This approach may actually understate
the impact of shall issue laws since the year dummy variables may
also pick up some of the changes attributed to the increasing number
of states that passed these laws.
Table 6 presents regressions for twelve different dependent
variables (six for multiple shootings and six for bombings) using the
simplest specifications of the shall issue law variable—a dummy law
variable which equals one if a state has a concealed handgun or “shall
issue’ law and zero otherwise. The regression analysis contains 953
observations (50 states and the District of Columbia for 19 years
minus 16 observations for various states and years in which we lacked
data on the arrest rate).
17
To simplify the table, we only present the
regression coefficients (and t-statistics) on the dummy law variable.
The results of Table 6 indicate that concealed handguns laws
significantly reduce multiple shootings in public places (but have no
systematic effects on bombings). For example, shall issue laws appear
to lower the combined number of killings and injuries (equation (3))
in a state by 11.1 per 10 million people per year, or by more than 80
percent of a one standard deviation change in the murder and injury
rate from multiple shootings. Equations (4) and (5) imply that the
average state passing these laws reduces the number of murders and
injuries by 6.9 and 6.5 persons respectively. Indeed these estimated
effects are so large that they often exceed the annual average number
of murders and injuries from public shootings in a state (either
absolutely or per 100,000 persons). To be sure, we expect large
deterrent effects from these laws because of the high probability that
one or more potential victim or bystander will be harmed. Still the
drop in murders and injuries is surprisingly large. And as we shall see,
alternative measures of shootings and adding other control variables
do not seem to reduce the magnitude of the law’s effect.
Turning to the other variables in the regressions in Table 6, we
find that states with larger populations have more multiple shooting
17
The states and years of the missing observations are as follows: Florida
(1988); Illinois (1993-95); Iowa (1991); Kansas (1993-95); Kentucky (1988);
Montana (1994-95); New Hampshire (1984 and 1995); Pennsylvania (1995)
and Vermont (1978-79). As a further check on our results, we reestimated the
regressions in Tables 6 and 7 deleting the arrest variable and adding the 16
missing observations. The coefficients and levels of significance on the shall
issue law variable were virtually unchanged.
Chicago Law and Economics Working Paper 12
deaths and injuries per 100,000 persons though the rate increases at a
decreasing rate. We also find that personal income, poverty and
population density are insignificant while retirement payments and
unemployment have positive and significant (or marginally
significant) effects on the murder rate. Higher arrest rates for
murder are associated with fewer multiple murders and killings but
these results are never statistically significant
18
(The full regressions
are available from the authors on request.)
Finally, notice that the number of bombings in Table 6 (with
the exception of “other bombing incidents” in eq. 12) are not related
to shall issue laws. Some types of bombings appear to rise and others
fall after the passage of a law, the signs often depend on whether
bombings are expressed as a rate or an absolute number, and five of
the six coefficients are not statistically significant. In short, there
does not appear any significant substitution between shootings and
bombings in states enacting “shall issue” laws. (In the remaining
tables we do not report the results for the bombing regressions
because, in almost all cases, bombings are not significantly related to
shall issue laws.)
Table 7 replaces the simple dummy law variable with two time
trend law variables for those states that passed laws between 1985
(the first year a state passed a law during the 1977 to 1995 sample
period) and 1995. The first is a time trend variable before passage of
the law that takes the value 0 in the year the law is passed (and all
years following passage), -1 in the year before passage, -2 in the
second year before passage and so forth. The second is a post law
variable that takes the value 0 in the year the law is passed (and 0 in
all years before passage), 1 in the first year after passage and so on.
The main reason for this specification is that we expect the impact
of shall issue laws to increase over time as more people obtain
permits. It may take many years after enacting a handgun law for
states to reach their long run level of permits. For states in which
18
We note that the arrest rate variable understates the actual (or expected)
arrest rate of individuals who go on shooting sprees. More than 90 percent of
these offenders are either arrested or killed, which is slightly greater than the
overall arrest rate for murder. The 90 percent figure (which comes from a Nexis
search) represents perpetrators who were immediately captured or killed. We do
not know whether those who escaped were apprehended later.
13 Multiple Victim Public Shootings
data on permits are available the share of the population with
permits is still increasing a decade after the passage of the law (Lott,
1998b, p. 75).
19
In Table 7, we find that deaths or injuries from mass shootings
are rising before the passage of the law and falling afterwards
(though the before law trend is only marginally significant in most
cases). The F-test for the differences in these time trends is always
significant at least at the .02 level. As expected, therefore, the more
years a shall issue law has been in effect in the 14 states that passed
laws starting in 1985, the greater the decline in murders and injuries
(both absolutely and per 100,000) from mass public shootings.
Because of the relatively large number of shootings that
occurred in the year prior to enactment of the laws and the
possibility that our results might be picking up a simple regression to
the mean, we reestimated the regressions in Tables 6 and 7 after
removing the observations for that year. All of the shall issue
coefficients in the shooting regressions remained statistically
significant, with the single exception of the injury rate in Table 6
which was negative but no longer statistically significant.
20
19
We note three other points related to Table 7.
(1) Eight states in our sample had shall issue laws during the entire
period. All eight passed their laws before 1960 and so should have reached
their equilibrium level of permits before 1977 (the first year in our sample).
These value assigned to two time trend variables for these states and states that
never enacted laws is zero.
(2) A second reason for the split time trend specification is that if (relative
to other states) shootings in states that pass shall issue laws are rising before
the law goes into effect and falling thereafter, a dummy law variable would
underestimate the law’s impact (even though the regression contains year
dummy variables). For example, imagine that the increase in shootings before
the law is symmetrical with the decline after the law. A simple dummy variable
for the presence or absence of the law would indicate that the law had no effect
yet the law might well have caused a change in the trend from positive to
negative.
(3) We also estimated regressions adding two time-squared variables for
the law variables. Here we find the same pattern of declining murders and
injuries after passage of the law with the decline flattening out by the sixth year
after enactment of the law.
20
Because of the relatively large number of shootings that occurred in the year
prior to enactment of the laws and the possibility that our results might be
picking up a simple regression to the mean, we reestimated the regressions in
Chicago Law and Economics Working Paper 14
Table 8 adds other law variables that may influence the number
of mass shootings. The law variables include the following: a
dummy variable if a state has a waiting period before an individual
can obtain a gun and the length of this period in days and days-
squared (see Lott and Mustard for a discussion of this variable); the
probability of execution (equal to the number of executions per
murder in a given year); and a dummy variable for whether a state
imposes additional penalties for using a gun in the commission of a
crime.
21
Three conclusions emerge from Table 8. First, the statistically
significant negative impact of shall issue laws on mass public
shootings continues to hold. Second, the regression coefficients on
the shall issues variables are of the same magnitude as in Table 7.
Third, the other gun related law variables and the capital
punishment variable have no significant impact on mass shootings.
The point estimates on the waiting period variables sometimes imply
that longer waiting periods increase the risk of mass public shootings
and other times they imply the reverse. In no case is the waiting
period variable statistically significant. Although the execution rate
has a negative coefficient in the six regressions, it is never statistically
significant. The execution variable itself may be only weakly related
to the probability that a mass murderer will be executed given the
Tables 6 and 7 after removing the observations for that year. All of the shall
issue coefficients in the shooting regressions remained statistically significant,
with the single exception of the injury rate in Table 6 which was negative but
no longer statistically significant. For example, the t-statistics for the shall
issue dummies in specifications 1 to 3 in Table 6 are -3.02, -1.456, and -2.295;
and the F-statistics for the before and after trends corresponding to
specifications 1 to 3 in Table 7 are 20.15, 6.67, and 13.99. Similar results were
also obtained when the two years preceding a state’s adoption of the law were
deleted from the sample. The t-statistics for the shall issue dummies in
specifications 1 to 3 in Table 6 are -2.732, -1.50, and -2.281; and the F-
statistics for the before and after trends corresponding to specifications 1 to 3 in
Table 7 are 20.96, 7.92, and 15.29.
21
See the Tracy L. Snell, Prisoners executed under civil authority in the
United States, by year, region, and jurisdiction, 1977-1995, Bureau of Justice
Statistics, May 14, 1997. For the source of penalties imposed for when a gun is
used in a commission of a crime see Thomas B. Marvell and Carl E. Moody,
The Impact of Enhanced Priosn Terms for Felonies Committed with Guns,”
Criminology 33 (May 1995): 247, 258-61.
15 Multiple Victim Public Shootings
long delays before execution and its over-inclusiveness (i.e., the
variable measures the execution rate for all murders not mass
murders), so we also tried including a simple dummy variable for
whether the dead penalty was in effect. However, the coefficient
was never statistically significant and did not alter our other results.
Finally, the law variable denoting the imposition of additional
penalties for using a gun in a crime is insignificant.
The impact of the death penalty on public shootings stands in
sharp contrast to evidence that we have put together on murder rates
using county level data from 1977 to 1995. We find that a one
percentage point increase in the execution rate is associated with a
seven percent decline in the overall murder rate and the effect is
statistically significant at better than .01 percent level. The question
is why does the presence of concealed handgun laws deter multiple
victim shootings when other penalties fail to have an effect. One
possibility is that the execution rate does not deter mass public killers
because they already die at such high rates from their crime, but
concealed handgun laws deter attacks because the killers are
committing the crime to kill or injure a large number of people and
that their return from these attacks is reduced when citizens with
concealed handguns can limit the carnage.
We also reestimated the regressions by including both the
murder and total bombing rates as explanatory variables because there
may be some overlap between factors that explain public shootings,
overall murders and bombings that are not taken into account by the
independent variables in the previous regressions. Adding these
variables to the regressions in Tables 6, 7 and 8 produces little
change from our previous results. In 17 of the 18 regressions, the
shall issue variable has a negative and statistically significant effect on
multiple shootings.
22
IV. Alternative Measures of Multiple Shootings
22
Even in the one case where the coefficient is no longer statistically
significant it is still negative. The t-statistic is –1.63 in the regression on
injuries per 100,000 persons using the specification in Table 6 with the
addition of the murder and bombing variables. The murder rate is always
positively related to mass shootings but statistically significant in only six of
the 18 regression equations. Total bombings are never statistically significant
in any of these regressions.
Chicago Law and Economics Working Paper 16
The dependent variable in Table 9 is the number of multiple
shootings reported in the first section of the New York Times.
Because the Tobit regressions with state specific effects did not
converge, we substituted regional dummy variables in the Tobit
regressions. We also present OLS estimates that include state fixed
effects variables. Regional and even state fixed effects may be
important if the New York Times has a regional or state bias in its
coverage of shooting events. Overall, the results are consistent with
our earlier findings. Both the dummy shall issue law variable and the
split time trend variables indicate significant reductions in multiple
shootings after the passage of the law. The OLS estimates also
produce negative effects of the law but these estimates are less
significant (i.e., they are generally significant at the .10 but not .05
level).
The only other study that we know of on multiple victim
murders defines the offense more narrowly than we have, and it
limits the definition to four or more people killed in the shooting
(Petee et. al., 1997). This definition sharply lowers the number of
public shootings to 36 incidents during the 1977 - 1995 period. We
attempted to explain both the per capita and absolute number of
people killed in these shootings using the same specifications as in
Tables 6 and 7.
23
The results are similar to those reported earlier.
We find that shall issue laws reduce the number of deaths, and that
these deaths were increasing before passage of the law and falling
thereafter.
24
.
23
Again, the Tobit estimates do not converge when state fixed effects are used
for there is not enough variation in the data to distinguish the law's impact on
these shootings with state fixed effects. Consequently, the state fixed effects are
replaced with regional dummies (Northeast, Midwest, South, and West (the left
out region)).
24
In explaining the per capita number of people killed, the shall issue
concealed handgun dummy variable was -.1207502 (t-statistic = 1.966) and the
difference in the before and after trends equaled -.057 (f-statistic = 6.96). For
the regressions explaining the number of people killed in a state, the shall issue
concealed handgun dummy variable was -15.4 (t-statistic = 2.363) and the
difference in the before and after trends equaled -7.05 (f-statistic = 9.41). The
mean number of deaths per 100,000 persons per state per year is .0037
(standard deviation = .0294) and the mean number of deaths per state per year is
.249 (standard deviation = 1.52).
17 Multiple Victim Public Shootings
V. The Number of Shootings and the Number
of People Killed or Injured Per Shooting
The presence of concealed handguns should reduce both the
number of public shootings and the amount of harm caused by any
one event. Consider the following examples. During a recent
shooting spree at a public school in Pearl, Mississippi, an assistant
principal retrieved his gun and physically immobilized the shooter
before he caused additional harm (CNN, October 2, 1997, 2:40 PM
EST). And in the public school related shooting in Edinboro,
Pennsylvania, which left one teacher dead, a shot gun pointed at
offender while he was reloading his gun prevented additional harm
(Reuters Newswire, April 26, 1998). The police did not arrive for
another ten minutes. The examples mentioned in the introduction
also illustrate cases where shooters have been shot by citizens and
thus presumably prevented from harming even more people.
Although one can also imagine circumstances where shall issue laws
increase the availability of guns to potential offenders or where guns
used in self defense lead to more not less killings, our results so far
strongly indicate that these effects, if they exist, are not sufficient to
offset the overall negative impact of shall issue laws on multiple
shootings.
Here we separate the effects these laws have on the number of
shootings from the number of people harmed. Suppose, for example,
perpetrators are undeterred by legal penalties or the prospect of
encountering an armed defender. Then the number of persons
harmed per shootings could still fall (as the two school shooting
examples suggest) if concealed handguns interfered with the
offender’s ability to carry out his plans. Using either the dummy law
variable or the before-and-after time trends, the coefficients in Table
10 indicate that concealed handguns reduce both the number of
shootings and the number of people harmed.
25
The evidence on
whether shall issue laws have a bigger impact on the number of
25
Note that there are 234 observations in the deaths or injuries per shooting
regressions although Table 1 indicates that there were 396 shootings in the
sample period. The dependent variable in equations (1) – (3) in Table 10 equals
the average number of deaths or injuries per shooting in a state in a year.
Hence if there were two or more multiple shootings in a state in a year, this
counted as one observation in the regression.
Chicago Law and Economics Working Paper 18
people harmed relative to the number of shootings is mixed. The
absolute size of the coefficient in the regression on deaths and
injuries (equation (3)) is three times greater than in the shooting
regression (equation (4)) but the elasticity is greater in latter than in
the former regression (the means are 4.52 and .41 for the number of
persons harmed and the number of shootings respectively).
26
VI. Do Shootings Produce More Shootings?
Does a public shooting lead others to imitate or mimic the
behavior of the first gunman? One might reason that the attention
and notoriety surrounding a shooting by A might encourage B to
undertake a similar act, and B’s act might encourage C and so on.
The notion of crime “fad” or epidemic is not new. Landes (1978, pp.
16-18) investigated and rejected the hypothesis that the increase and
subsequent decrease in airline hijackings in Europe and the United
States over the 1961 to 1976 period could be explained by a fad
rather than by the increase in apprehension rates and penalties.
To test for fads or imitative behavior, we calculate the number
of mass shootings per month for the 228 months in the 1977 to
1995 period. We specified the dependent variable in two ways: the
number of monthly shootings and the difference in the number
between adjacent months. The regression includes dependent
variables denoting various monthly lags in either the number of
shootings (or number reported in the New York Times) or the
change in the number of shootings. We control for the increase in
the number of states with shall issue laws during this period by
adding a variable denoting the percentage of the U.S. population
covered by these laws. We also include month dummy variables and
a time trend (in months). Table 11 reports both Tobit and OLS
estimates of the regression equations.
In Table 11, only the percent of the U.S. population covered by
shall issue laws and the time trend variables are consistently
significant in all eight regressions. The positive coefficients on the
26
While individuals with permits produce a large social benefit, they risk
being shot by the attacker. We have no instances where people with permits
have indeed been shot, but this risk surely raises the prospects of whether
citizens with permits should be compensated or at least not have to pay large
fees for obtaining a permit.
19 Multiple Victim Public Shootings
lagged values of shootings in the first four regressions provide some
weak evidence of faddish behavior. But the lagged values of the
change in shootings in last four equations imply the opposite—i.e.,
the larger the increase in shootings during the prior months, the
greater the decline in the current month. In short, the evidence on
fads is mixed.
27
One reason we may not find significant evidence of faddish
behavior is that lagged shootings and lagged stories on shootings in
the New York Times are highly collinear. To test this we reestimated
the regressions in Table 11 using either lagged shootings or lagged
stories by themselves. We still find no systematic evidence of faddish
behavior. Lagged values of shootings are positively related to
monthly shootings while lagged differences are negatively related to
differences in monthly shootings. Again the percent of the
population covered by shall issue laws continues to have negative and
statistically significant effects on the number and change in the
number of monthly shootings.
While the evidence provides little support for the copycat
hypothesis, we note that our data contains shootings by adults. The
recent public school shootings involving children might be different
and more consistent with the imitative hypothesis. However, data
on school shootings involve such a small sample that it is not
possible to study these shootings separately.
VII. Explaining the Passage of Shall Issue Laws
The previous specifications assume that the passage of a shall
issue law is an exogenous event. Following Lott and Mustard (1997,
pp. 39-48), we assume that the likelihood a state will enact a law
depends on political influence variables that include the percentage
of its population that are members of the National Rifle Association,
the percentage of votes received by the Republican presidential
candidate in the state, lagged and changes in the rates of violent and
property crime, and fixed regional effects. Since presidential elections
occur every four years, we interacted the percentage voting
27
Note that October appears to be the most dangerous month although he
number of shootings in October is only significantly greater than the number
in January, September and November. Note, however, that the monthly
dummy variables are not jointly significant.
Chicago Law and Economics Working Paper 20
Republican with dummy variables for the years adjacent to the
relevant elections. Thus, the percentage of the vote obtained in 1980
is multiplied by a year dummy for the years 1979-82, and so on,
through the 1996 election.
The first stage (see the bottom half of Table 12) implies that
states adopting these laws are relatively Republican and have low
violent crime but high property crime rates. Neither NRA
membership nor rising violent and property crimes rates are
statistically significant (though the NRA variable is marginally
significant). The second stage regressions support our earlier results
contained in Table 6. Adopting a shall issue law is associated with a
significant decline in the combined number of multiple killings and
injuries (both absolutely and per 100,000 persons). In the separate
murder and injury regressions, the coefficients are always negative
and either significant or marginally significant (a t-statistic greater
that 1.65).
28
VIII. Conclusion
The results of this paper support the hypothesis that concealed
handgun or shall issue laws reduce the number of multiple victim
public shootings. Attackers are deterred and the number of people
injured or killed per attack is also reduced, thus for the first time
providing evidence that the harm from crimes that still occur can be
mitigated. The results are robust with respect to different
specifications of the dependent variable, different specifications of
the handgun law variable, and the inclusion of additional law
variables (e.g., mandatory waiting periods and enhanced penalties for
using a gun in the commission of a crime). Not only does the
passage of a shall issue law have a significant impact on multiple
shootings but it is the only law related variable that appears to have a
significant impact. Other law enforcement efforts from the arrest
rate for murder to the death penalty to waiting periods and
background checks are not systematically related to multiple
28
As a test of whether the shall issue laws were passed because of a shooting,
we reestimated just first stage regression by itself after including the lagged
murder or injury rate from the shootings to see if the law was adopted because
of the shooting. While the coefficients on these lagged values were positive,
neither variable was ever statistically significant.
21 Multiple Victim Public Shootings
shootings. We also find that shall issue laws deter both the number
of multiple shootings and the amount of harm per shooting. Finally,
because the presence of citizens with concealed handguns may be
able to stop attacks before the police are able to arrive, our data also
allows us to provide the first evidence on the reduction in severity of
those crimes that still take place.
Chicago Law and Economics Working Paper 22
Bibliography
Alschuler, Albert W., “Two Guns, Four Guns, Six Guns, More
Guns: Does Arming the Public Reduce Crime,” Valparaiso Law
Review 31 (Spring 1997): 365-373.
Bartley, William Alan; Cohen, Mark A. and Frobe, Luke. “The
Effect of Concealed Weapon Laws: Estimating Model
Uncertainty.” Economic Inquiry 36 (April 1998): 258-265.
Black, Dan A. and Daniel S. Nagin, “Do ‘Right-to-Carry’ Laws
Deter Violent Crime?” Journal of Legal Studies 27 (January 1998):
209-219.
Bronars, Stephen G. and John R. Lott, Jr., “Criminal Deterrence,
Geographic Spillovers, and the Right to Carry Concealed
Handguns,” American Economic Review, 82 (May 1998): 475-478.
Cook, P. J., “The Role of Firearms in Violent Crime,” in M.E.
Wolfgang and N.A. Werner, eds., Criminal Violence, Sage
Publishers: Newbury, N.J.(1982).
Kates, Don, and Dan Polsby. “Of Genocide and Disarmament,”
Journal of Criminal Law and Criminology, 86 (Fall 1995): 247-256.
Kleck, Gary, and Marc Gertz, “Armed Resistance to Crime: The
Prevalence and Nature of Self-Defense with a Gun,” 86 Journal of
Criminal Law and Criminology 86 (Fall 1995): 150-187.
Landes, William M., “An Economic Study of U.S. Airline
Hijacking, 1961-1976,” Journal of Law and Economics 21 (April
1978): 1-32.
Lott, John R., Jr. and David Mustard, “Crime, Deterrence, and
Right-to-Carry Concealed Handguns,” Journal of Legal Studies 26
(January 1997): 1-68.
23 Multiple Victim Public Shootings
Lott, John R., Jr., “The Concealed Handgun Debate,” Journal of
Legal Studies 27 (January 1998a): 221-243.
Lott, John R., Jr., More Guns, Less Crime: Understanding Crime and
Gun Control Laws, University of Chicago Press: Chicago, Illinois
(1998b).
Marvell, Thomas B., and Carl E. Moody, The Impact of Enhanced
Priosn Terms for Felonies Committed with Guns,” Criminology 33
(May 1995).
Petee, Thomas A., Kathy G. Padgett, and Thomas York,
“Debunking the Sterotype: An Examination of Mass Murder in
Public Places,” Homicide Studies 1 (November 1997): 317-337.
Plassman, Florenz and T. Nicolaus Tideman, “Does the Right to
Carry Concealed Handguns Deter Countable Crimes?: Only a
Count Analysis Can Say,” Virginia Polytechnic Institute Working
Paper (October 9, 1998).
Reuters Newswire, “Police praise man in PA dance hall shooting,”
Sunday, April 26, 1998.
Zimring, Franklin, “The Medium is the Message: Firearm Caliber as
a Determinant of Death from Assault,” Journal Legal Studies, 1
(1972): 97-123.
Chicago Law and Economics Working Paper 24
This paper is a draft dated February 27, 1999. All rights reserved.
Readers with comments should address them to:
John R. Lott, Jr.
John M. Olin Fellow in Law and Economics
University of Chicago Law School
1111 East 60
th
Street
Chicago, IL 60637
26
Table 1
The Number of Multiple Victim Murders and Injuries in
Public Shootings by Year and by the Presence of a Concealed Handgun Law
All States States Without Nondiscretionary Handgun Law (Including the District of Columbia)
Year
# of Murders
in Public
Shootings
# of Injuries
in Public
Shootings
# of Public
Shootings
# of States w/out
Non-discretionary
Concealed
Handgun Law
# of Murders
in Public
Shootings
# of Injuries
in Public
Shootings
# of
Shootings
% of Total
Deaths
(Column 5/
Column 1)
% of Total
Injuries
(Column 6/
Column 2)
% of Total
Deaths
(Column 7/
Column 3)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
1977 19 46 7 43 19 46 7 100% 100% 100%
1978 14 12 8 43 14 12 8 100% 100% 100%
1979 23 77 13 43 20 74 12 87% 96% 92%
1980 30 51 11 43 22 46 8 73% 90% 73%
1981 44 60 30 43 37 50 27 84% 83% 90%
1982 32 92 20 43 28 92 19 87% 100% 95%
1983 19 36 18 43 16 22 14 84% 61% 78%
1984 56 76 26 43 53 73 24 95% 96% 92%
1985 38 45 24 43 34 37 21 89% 82% 88%
27
1986 41 54 21 42 41 52 20 100% 96% 95%
1987 44 73 36 42 41 69 34 93% 95% 94%
1988 49 90 35 41 47 85 32 96% 94% 91%
1989 49 84 31 40 39 79 24 80% 94% 77%
1990 29 53 22 37 20 43 20 69% 81% 91%
1991 58 68 22 34 53 58 18 91% 85% 82%
1992 31 55 18 33 29 54 17 94% 98% 94%
1993 87 83 33 33 83 76 30 95% 92% 91%
1994 15 20 10 33 13 19 9 87% 95% 90%
1995 26 11 11 29 23 11 10 88% 100% 91%
28
Table 1 (continued)
States With Shall Issue Concealed Handgun Law
Year
# of States
w/ Law
# of Murders in
Public Shootings
# of Injuries in
Public Shootings # of Shootings
% of Total
Deaths
(Column 12/
Column 1)
% of Total
Injuries
(Column 13/
Column 2)
% of Total
Deaths
(Column 14/
Column 3)
(11) (12) (13) (14) (15) (16) (17)
1977 8 0 0 0 0% 0% 0%
1978 8 0 0 0 0% 0% 0%
1979 8 3 3 1 13% 4% 8%
1980 8 8 5 3 27% 10% 27%
1981 8 7 10 3 16% 17% 10%
1982 8 4 0 1 13% 0% 5%
1983 8 3 14 4 16% 39% 22%
1984 8 3 3 2 5% 4% 8%
1985 8 4 8 3 11% 18% 12%
1986 9 0 2 1 0% 4% 5%
29
1987 9 3 4 2 7% 5% 6%
1988 10 2 5 3 4% 6% 9%
1989 11 10 5 7 20% 6% 23%
1990 14 9 10 2 31% 19% 9%
1991 17 5 10 4 9% 15% 18%
1992 18 2 1 1 6% 2% 6%
1993 18 4 7 3 5% 8% 9%
1994 18 2 1 1 13% 5% 10%
1995 22 3 0 1 12% 0% 9%
30
Table 2
The Rate of Multiple Victim Murders and Injuries in
Public Shootings by Year and by the Presence of a Concealed Handgun Law
States Without Shall Issue Law States With Shall Issue Law Comparison of Rates Between Two Types
of States
Year # of States w/out
Shall Issue Law
(Including the
District of
Columbia)
Murders and
Injuries in
Public
Shootings Per
100,000 People
# of Shootings
Per 100,000
People
# of States
w/ Shall
Issue Law
Murders and
Injuries in Public
Shootings Per
100,000 People
# of Shootings
Per 100,000
People
Does the Murder
and Injury Rate in
States w/out Laws
Exceed the Rate in
States w/ Laws
Does the Shooting
Rate in States w/out
Laws Exceed the
Rate in States with
Laws
1977 43 0.033 .004 8 0.000 0 Yes Yes
1978 43 0.013 .007 8 0.000 0 Yes Yes
1979 43 0.046 .011 8 0.031 .003 Yes Yes
1980 43 0.033 .004 8 0.067 .010 No No
1981 43 0.041 .012 8 0.087 .009 No Yes
1982 43 0.057 .009 8 0.020 .004 Yes Yes
1983 43 0.018 .014 8 0.086 .012 No Yes
1984 43 0.058 .012 8 0.030 .021 Yes No
1985 43 0.032 .010 8 0.060 .024 No No
1986 42 0.042 .006 9 0.009 .002 Yes Yes
31
1987 42 0.050 .014 9 0.033 .004 Yes Yes
1988 41 0.063 .019 10 0.021 .004 Yes Yes
1989 40 0.057 .010 11 0.037 .031 Yes No
1990 37 0.034 .013 14 0.031 .002 Yes Yes
1991 34 0.061 .005 17 0.022 .002 Yes Yes
1992 33 0.045 .004 18 0.004 .001 Yes Yes
1993 33 0.085 .013 18 0.015 .001 Yes Yes
1994 33 0.017 .005 18 0.004 .0004 Yes Yes
1995 29 0.019 .003 22 0.004 .0008 Yes Yes
Ave. 0.0423 .009 0.0288 .0064 Yes Yes
32
Table 3
The Fourteen States That Changed Their
Concealed Handgun Laws between 1977 and 1995
(Each cell shows the mean and the standard deviation in parentheses.)
Fourteen States that Changed from Not Having to
Having a Shall Issue Concealed Handgun Law
Years during Which These States Did Not
Have Shall Issue Concealed Handgun Laws
(Observations = 200)
Years During Which They Did Have
Shall Issue Concealed Handgun Laws
(Observations = 66)
Murders in Multiple Victim Public
Shootings Per 100,000 People
.026
(.119)
.0028
(.0093)
Injuries in Multiple Victim Public
Shootings Per 100,000 People
.025
(.099)
.0044
(.014)
Murders and Injuries in Multiple
Victim Public Shootings Per 100,000
People
.051
(.190)
.007
(.021)
References to Multiple Victim Public
Shootings in the First Section of the
New York Times Per 100,000 People
.0033
(.023)
.0003
(.0016)
Number of Shootings Per 100,000
People
.0136
(.0499)
.002
(.0061)
Actual and Attempted Bombings Per
100,000 People
.600
(.632)
.6998
(.492)
Actual and Attempted Incendiary
Bombings Per 100,000 People
.1468
(.206)
.1257
(.124)
Other Bomb Related Incidents Per
100,000 People
.863
(.833)
.757
(.799)
Total Explosive Incidents Per 100,000
People
1.610
(1.277)
1.571
(1.059)
33
Table 4
Examining the Means for States that Changed Their
Concealed Handgun Laws During the 1977 to 1995 Period
States that Changed Their Concealed Handgun Laws During the 1977-95 Period: Using State Avg’s to Compute Rates
Years Before and
After the
Adoption of the
Law (Year 1 is
the first full
Year that the
Law is in Effect)
Number
of States
that Fall
into that
Category
Murders in
Multiple
Victim Public
Shootings Per
100,000 People
Injuries in
Multiple
Victim Public
Shootings Per
100,000 People
Murders and
Injuries in
Multiple Victim
Public
Shootings Per
100,000 People
The Number of
Shootings Per
100,000 People
Total Number of
Murders in
Multiple Victim
Public Shootings
for all States in
this Category
Total Number
of Injuries in
Multiple Victim
Public
Shootings for
all States in this
Category
Total Number of
Murders and
Injuries in
Multiple Victim
Public Shootings
for all States in
this Category
-8 14 0.0033 0.0175 0.0208 0.0060 4 20 24
-7 14 0.0033 0.0145 0.0177 0.0116 2 7 9
-6 14 0.0593 0.0349 0.0942 0.0307 12 11 23
-5 14 0.0055 0.0032 0.0087 0.0013 8 5 13
-4 14 0.0151 0.0235 0.0386 0.0141 16 16 32
-3 14 0.0064 0.0159 0.0223 0.0045 5 12 17
-2 14 0.0188 0.0041 0.0229 0.0102 6 4 10
-1 14 0.0570 0.0896 0.1467 0.0460 13 17 30
0 14 0.0152 0.0277 0.0428 0.0154 20 25 45
34
1 14 0.0034 0.0097 0.0131 0.0046 4 10 14
2 10 0.0033 0.0050 0.0083 0.0033 4 6 10
3 10 0.0062 0.0046 0.0108 0.0008 8 6 14
4 10 0.0000 0.0166 0.0166 0.0003 0 2 2
5 9 0.0000 0.0000 0.0000 0.0000 0 0 0
6 6 0.0049 0.0061 0.0110 0.0024 4 5 9
7 3 0.0048 0.0024 0.0072 0.0024 2 1 3
8 2 0.0000 0.0000 0.0000 0.0000 0 0 0
35
Table 5
Means and Standard Deviation of Variables
Obs. Mean Standard Deviation
Shall Issue Law Dummy 969 .22497 .41778
Arrest Rate for Murder 953 89.4275 53.9812
Murders in Multiple Victim Public Shootings Per 100,000 Persons 969 .01586 .066083
Injuries in Multiple Victim Public Shootings Per 100,000 Persons 969 .021886 .09133
Murders and Injuries in Multiple Victim Public Shootings Per 100,000 Persons 969 .03775 .13236
Murders in Multiple Victim Public Shootings 969 .72652 2.47185
Injuries in Multiple Victim Public Shootings 969 1.1207 3.850898
Murders and Injuries in Multiple Victim Public Shootings 969 1.84727 5.76515
Attempted or Actual Bombings Per 100,000 Persons 969 .568889 .488192
Attempted or Actual Incendiary Bombings Per 100,000 Persons 969 8.12461 19.16359
Attempted or Actual Other Bombing Incidents Per 100,000 Persons 969 .709398 .6875484
Attempted or Actual Bombings 969 26.20906 42.02015
Attempted or Actual Incendiary Bombings 969 8.12461 19.16359
Attempted or Actual Other Bombing Incidents 969 28.44376 41.44581
Deaths per shooting 239 1.707 1.54538
Injuries per Shooting 239 2.22 4.47260
Deaths or Injuries per Shooting 239 4.525 4.83836
Number of Shootings 969 .4087 1.01036
Number of Shootings per 100,000 Persons 969 .00888 0.0289133
Murders per 100,000 Persons 966 7.56856 7.41677
Death Penalty Execution Rate 966 .0008941 .0042598
Waiting Period Dummy 969 .3750877 .47944
NRA Members Per 100,000 Persons 969 1231.408 674.4301
State Population 969 4718730 5115287
State Population Squared 969 4.84e+13 1.20e+14
Real Per Capita Personal Income 969 12874.66 2293.915
Real Per Capita Income Maintenance 969 166.6172 66.8434
Real Per Capita Unemployment Insurance Payment 969 73.36105 44.88649
36
Real Retirement Payments Per Person Over 65 969 151.1192 55.0607
Population Density Per Square Mile 969 356.704 1399.525
Unemployment Rate 969 6.564 2.1023
Poverty Rate 969 13.55103 4.223485
Percent of the Population that is:
Black Males 10 to 19 Years of Age 969 1.00559 1.08956
Black Females 10 to 19 Years of Age 969 .9921156 1.10503
White Males 10 to 19 Years of Age 969 6.56392 1.560617
White Females 10 to 19 Years of Age 969 6.25813 1.526072
Other Males 10 to 19 Years of Age 969 .368235 .732602
Other Females 10 to 19 Years of Age 969 .3557895 .707472
Black Males 20 to 29 Years of Age 969 .946677 1.021947
Black Females 20 to 29 Years of Age 969 1.024541 1.207095
White Males 20 to 29 Years of Age 969 7.205986 1.252604
White Females 20 to 29 Years of Age 969 7.05257 1.291831
Other Males 20 to 29 Years of Age 969 .3611558 .7046579
Other Females 20 to 29 Years of Age 969 .3647368 .7125971
Black Males 30 to 39 Years of Age 969 .7368627 .8415113
Black Females 30 to 39 Years of Age 969 .8434159 1.00203
White Males 30 to 39 Years of Age 969 6.74614 1.214384
White Females 30 to 39 Years of Age 969 6.693695 1.203899
Other Males 30 to 39 Years of Age 969 .3143137 .6723823
Other Females 30 to 39 Years of Age 969 .3460475 .7103821
Black Males 40 to 49 Years of Age 969 .4893395 .5828483
Black Females 40 to 49 Years of Age 969 .5763158 .7112123
White Males 40 to 49 Years of Age 969 5.019412 1.049877
White Females 40 to 49 Years of Age 969 5.03807 1.024572
Other Males 40 to 49 Years of Age 969 .2131166 .502547
Other Females 40 to 49 Years of Age 969 .2391022 .5474948
Black Males 50 to 64 Years of Age 969 .5142621 .6773694
Black Females 50 to 64 Years of Age 969 .6462745 .8779527
White Males 50 to 64 Years of Age 969 5.727946 1.03609
White Females 50 to 64 Years of Age 969 6.147668 1.225103
37
Other Males 50 to 64 Years of Age 969 .2020537 .6068594
Other Females 50 to 64 Years of Age 969 .2359133 .6975253
Black Males Over 64 Years of Age 969 .3600516 .4907798
Black Females Over 64 Years of Age 969 .5555728 .8039973
White Males Over 64 Years of Age 969 4.337441 1.157958
White Females Over 64 Years of Age 969 6.324757 1.685616
Other Males Over 64 Years of Age 969 .1291434 .4826889
Other Females Over 64 Years of Age 969 .1490402 .5136774
Violent Crime Rate Per 100,000 Persons 969 482.7601 338.3358
Murder Rate Per 100,000 Persons 969 7.56856 7.416777
Rape Rate Per 100,000 Persons 969 33.76551 16.04484
Aggravated Assault Rate Per 100,000 Persons 969 282.3533 177.0785
Robbery Rate Per 100,000 Persons 969 161.7178 176.6475
38
Appendix: Coefficient Estimates for Regression Explaining
Murders from Multiple Victim Public Shootings
(Does not report the state and year fixed effects
that were included in the regression.)
Coefficient t-statistics Probability
Shall Issue Law Dummy -.0608 2.766 0.006
Arrest Rate for Murder -.00027 1.597 0.111
State Population 5.11e-08 2.167 0.031
State Population Squared -7.95e-16 1.618 0.106
Real Per Capita Personal Income 4.74e-06 0.365 0.715
Real Per Capita Income Maintenance .000078 0.317 0.751
Real Per Capita Unemployment Insurance Payment -5.28e-06 0.028 0.977
Real Retirement Payments Per Person Over 65 .00109 1.795 0.073
Unemployment Rate .0103159 1.999 0.046
Poverty Rate -.00212 0.832 0.406
Percent of the Population that is:
Black Males 10 to 19 Years of Age -.518476 1.169 0.243
Black Females 10 to 19 Years of Age .62572 1.416 0.157
White Males 10 to 19 Years of Age .508906 2.169 0.030
White Females 10 to 19 Years of Age -.55929 2.275 0.023
Other Males 10 to 19 Years of Age .94188 1.994 0.046
Other Females 10 to 19 Years of Age -1.12658 2.292 0.022
Black Males 20 to 29 Years of Age .50902 1.739 0.082
Black Females 20 to 29 Years of Age -.30357 1.069 0.285
White Males 20 to 29 Years of Age -.310526 3.067 0.002
White Females 20 to 29 Years of Age .3187005 2.815 0.005
Other Males 20 to 29 Years of Age .3769835 0.797 0.425
Other Females 20 to 29 Years of Age -.252396 0.514 0.607
Black Males 30 to 39 Years of Age -.92964 2.593 0.010
Black Females 30 to 39 Years of Age .60665 2.119 0.034
39
White Males 30 to 39 Years of Age .31494 1.815 0.070
White Females 30 to 39 Years of Age -.37101 2.178 0.030
Other Males 30 to 39 Years of Age -1.3244 2.135 0.033
Other Females 30 to 39 Years of Age .67917 1.402 0.161
Black Males 40 to 49 Years of Age -.201499 0.482 0.630
Black Females 40 to 49 Years of Age .20465 0.525 0.599
White Males 40 to 49 Years of Age -.12779 0.721 0.471
White Females 40 to 49 Years of Age .19292 1.081 0.280
Other Males 40 to 49 Years of Age .1767997 0.287 0.774
Other Females 40 to 49 Years of Age -.56335 1.037 0.300
Black Males 50 to 64 Years of Age .11955 0.273 0.785
Black Females 50 to 64 Years of Age -.72681 1.814 0.070
White Males 50 to 64 Years of Age -.14827 0.889 0.374
White Females 50 to 64 Years of Age .12739 0.929 0.353
Other Males 50 to 64 Years of Age .92883 1.627 0.104
Other Females 50 to 64 Years of Age 1.0097 1.785 0.075
Black Males Over 64 Years of Age -.29203 0.594 0.553
Black Females Over 64 Years of Age .37786 1.115 0.265
White Males Over 64 Years of Age .060696 0.420 0.675
White Females Over 64 Years of Age -.02943 0.312 0.755
Other Males Over 64 Years of Age 1.1684 1.490 0.136
Other Females Over 64 Years of Age -1.0168 1.528 0.127
Intercept -.02097 -0.017 0.987
Chi Square 356.36
Number of Observations 953
40
Table 6
The Impact of Shall Issue Concealed Handgun Laws on Public Shootings and Bombings
(The regressions use the tobit procedure. The regressions include the following independent variables: detailed
demographic information by sex, race, and age; population and population squared; state unemployment rate; state
poverty rate; real per capita personal income, unemployment payments, income maintenance payments; retirement
payments; arrest rate of murder; and state and year fixed effects. Absolute t-statistics are shown in parentheses.)
Endogenous Variables
Exogenous
Variables
Murders in
Multiple
Victim Public
Shootings Per
100,000 People
Injuries in
Multiple
Victim Public
Shootings Per
100,000 People
Murders and
Injuries in
Multiple Victim
Public Shootings
Per 100,000 People
Murders in
Multiple Victim
Public Shootings
Injuries in
Multiple Victim
Public Shootings
Murders and Injuries
in Multiple Victim
Public Shootings
(1) (2) (3) (4) (5) (6)
Shall Issue
Law
Dummy
Variable
-.061
(2.766)
-.0582
(1.658)
-.111
(2.467)
-6.887
(2.959)
-6.473
(2.208)
-13.38
(3.136)
Chi-
Square
356.36 384.04 361.96 619.5 720.60 723.22
Number of
Observatio
ns
953 953 953 953 953 953
41
Endogenous Variables
Exogenous
Variables
Attempted or
Actual
Bombings Per
100,000 People
Attempted or
Actual
Incendiary
Bombings Per
100,000 People
Other Bombing
Incidents Per
100,000 People
Attempted or
Actual Bombings
Attempted or
Actual Incendiary
Other Bombing
Incidents
(7) (8) (9) (10) (11) (12)
Shall Issue
Law
Dummy
Variable
-.0488
(0.874)
.0018
(0.047)
.00046
(1.266)
.965
(0.215)
-2.9798
(0.762)
13.76
(2.034)
Chi-
Square
1077.7 688.01 804.98 2688.8 1371.66 1820.26
Number of
Observatio
ns
953 953 953 953 953 953
42
Table 7
The Time Trends in Public Shootings and Bombings
Before and After the Implementation of the Shall Issue Laws
(The regressions use the tobit procedure and include the following independent variables: detailed demographic information by sex,
race, and age; population, population squared, and state unemployment rate; state poverty rate; real per capita personal income,
unemployment payments, income maintenance payments, retirement payments; arrest rate for murder; and state and year fixed effects.
Absolute t-statistics are shown in parentheses.)
Endogenous Variables
Exogenous
Variables
Murders in
Multiple
Victim Public
Shootings Per
100,000 People
Injuries in
Multiple
Victim Public
Shootings Per
100,000 People
Total Murders and
Injuries in
Multiple Victim
Public Shootings
Per 100,000 People
Murders in
Multiple Victim
Public Shootings
Injuries in
Multiple Victim
Public Shootings
Total Murders
and Injuries in
Multiple Victim
Public Shootings
(1) (2) (3) (4) (5) (6)
Time Trend for
Years Before the
Shall Issue Law
Went into Effect
.0093
(2.321)
.0102
(1.630)
.0148
(1.900)
.8217
(1.867)
.7738
(1.402)
1.1283
(1.461)
Time Trend for
Years After the
Shall Issue Law
Went into Effect
-.0222
(3.507)
-.01888
(1.879)
-.0384
(2.973)
-2.7703
(4.040)
-1.915
(2.229)
-4.606
(3.698)
F-test for
Differences in
Time trends
(probability in
parentheses)
18.37
(.0000)
6.49
(.0110)
13.00
(.0003)
20.61
(.0000)
7.57
(.0060)
16.24
(.0001)
Chi-Square 367.55 388.07 369.17 632.81 723.40 730.55
Number of Obs. 953 953 953 953 953 953
43
Table 8
Including Other Gun Control Laws and Death Penalty Execution Rates
(The regressions use the tobit procedure and include the following independent variables: detailed demographic
information by sex, race, and age; population, population squared, and state unemployment rate; state poverty rate; real
per capita personal income, unemployment payments, income maintenance payments, retirement payments; arrest rate for
murder; and state and year fixed effects. Absolute t-statistics are shown in parentheses.)
Endogenous Variables
Exogenous
Variables
Murders in
Multiple
Victim Public
Shootings Per
100,000 People
Injuries in
Multiple
Victim Public
Shootings Per
100,000 People
Total Murders and
Injuries in
Multiple Victim
Public Shootings
Per 100,000 People
Murders in
Multiple Victim
Public Shootings
Injuries in
Multiple Victim
Public Shootings
Total Murders
and Injuries in
Multiple Victim
Public Shootings
(1) (2) (3) (4) (5) (6)
Time Trend for
Years Before the
Shall Issue Law
Went into Effect
.0093
(2.270)
.0119
(1.842)
.0164
(2.047)
.7016
(1.556)
.7953
(1.392)
1.059
(1.325)
Time Trend for
Years After the
Shall Issue Law
Went into Effect
-.0247
(3.260)
-.0218
(1.911)
-.042
(2.789)
-3.113
(3.713)
-2.1359
(2.213)
-5.07
(3.472)
Waiting Period
Dummy
.025
(0.513)
.0558
(0.702)
.0501
(0.509)
-.7575
(0.127)
-1.23
(0.158)
-1.652
(0.151)
Length of
Waiting Period
in Days
.00017
(0.0111)
-.011
(0.427)
-.00084
(0.027)
3.7184
(1.199)
4.233
(0.969)
6.777
(1.148)
44
Length of
Waiting Period
Squared
-.00064
(0.474)
-.0007
(0.307)
-.0016
(0.614)
-.4137
(1.045)
-.6252
(1.101)
-.801
(1.058)
Additional
Penalty for Using
Gun in the
Commission of a
Crime Dummy
-.00048
(0.271)
.0179
(0.619)
.017
(0.485)
.0037
(0.002)
1.777
(0.711)
2.551
(0.755)
Death Penalty
Execution Rate
-.8077
(0.431)
-1.91
(0.624)
-2.28
(0.593)
-187.02
(0.878)
-.81499
(0.003)
-1882.35
(0.469)
F-test for
Differences in
Time trends
(probability in
parentheses)
16.73
(.0000)
7.27
(.0071)
12.71
(.0004)
17.74
(.0000)
7.55
(.0061)
14.87
(.0001)
Chi-Square 369.75 393.76 373.65 642.40 728.87 739.63
Number of
Observations 953 953 953 953 953 953
45
Table 9
News Stories on Multiple Victim Public Shootings
in the First Section of the New York Times
(Because the Tobit estimates reported in the first two columns did not converge when we included all the control
variables from the earlier regressions, we replaced the state fixed effects with regional dummy variables. The OLS
regressions in the third and fourth columns control for all variables used earlier. Absolute t-statistics are shown in
parentheses.)
Endogenous Variables
Tobit Estimates Ordinary Least Squares Estimates
Exogenous Variables
Multiple Victim Public
Shooting Stories
Appearing in the First
Section of the New York
Times Per 100,000 People
in a State
Multiple Victim
Public Shooting
Stories Appearing in
the First Section of
the New York Times
Multiple Victim Public
Shooting Stories
Appearing in the First
Section of the New York
Times Per 100,000 People
in a State
Multiple Victim Public
Shooting Stories
Appearing in the First
Section of the New York
Times
A) Estimates Using
Only a Simple
Dummy Variable
Shall Issue Law
Dummy Variable
-.0305
(3.645)
-2.485
(3.530)
-.0025
(1.635)
-.2151
(1.684)
Chi-Square 239.06 363.31
F-statistic 2.06 4.44
R
2 .1274 .3787
B) Estimates Using
Before and After
Time Trends
46
Time Trend for
Years Before the
Shall Issue Law
Went into Effect
.0023
(1.832)
.1913
(1.773)
.0001899
(0.823)
.0096
(0.502)
Time Trend for
Years After the Shall
Issue Law Went into
Effect
-.0055
(2.66)
-.4126
(2.434)
-.00058
(1.556)
-.06212
(2.009)
F-test for
Differences in Time
trends (probability in
parentheses)
10.48
(.0013)
9.63
(.0020)
2.86
(.0913)
3.61
(.0579)
Chi-Square 234.39 358.70
F-statistic 2.05 4.46
R
2 .1277 .3797
Number of
Observations
953 953 953 953
47
Table 10
The Impact of Shall Issue Concealed Handgun Laws on the
Number of Shootings and the Number of Deaths or Injuries from each Shooting
(Equation (3) uses ordinary least squares while the other regressions use the tobit procedure. The regressions include
demographic information by sex, race, and age; population, population squared, and state unemployment rate; state
poverty rate; real per capita personal income, unemployment payments, income maintenance payments, and retirement
payments; and state and year fixed effects. Absolute t-statistics are shown in parentheses.)
Endogenous Variables
Exogenous Variables
Deaths per
Shooting
Injuries per
Shooting
Deaths or Injuries
per Shooting
Number of Shootings Number of Shootings per
100,000 people
(1) (2) (3) (4) (5)
Shall Issue Law
Dummy Variable
-2.246
(2.749)
-4.315
(1.962)
-6.139
(2.020)
-1.405
(2.115)
-.0172
(2.145)
F-statistic 1.72
R
2 .6067
Chi-Square 182.01 214.90 758.63 355.67
(6) (7) (8) (9) (10)
Time Trend for
Years Before the
Shall Issue Law
Went into Effect
-.0237
(0.105)
.7029
(1.090)
.44796
(0.535)
.2140
(1.728)
.0030
(2.066)
48
Time Trend for
Years After the
Shall Issue Law
Went into Effect
-.6726
(2.385)
-.9100
(1.187)
-1.635
(1.568)
-.5870
(3.069)
-.0053
(2.404)
F-test for
Differences in Time
trends (probability
in parentheses)
3.81
(0.0534)
3.50
(0.0636)
2.5
(0.0885)
12.88
(0.0004)
9.92
(0.0017)
F-statistic 1.68
R
2 .6039
Chi-Square 180.23 214.57 767.41 361.23
Number of
Observations 234 234 234 953 953
49
Table 11
Do Shootings Produce Yet More Shootings?
(Absolute t-statistics in parentheses. The regression also includes monthly dummy variables.)
Endogenous Variable: Number of
Shootings Per Month
Endogenous Variable: First Differences in the
Number of Shootings Per Month (For these
specifications the explanatory variables
for the number of shootings and the number of
New York Times articles are also
differenced and then lagged)
Tobit OLS Tobit Cochrane-Orcutt
Exogenous Variables (1) (2) (3) (4) (5) (6) (7) (8)
Number of Shootings in
Previous Month
.1690
(1.785)
.1526
(1.588)
.1264
(1.324)
.1550
(2.000)
-.5362
(5.136)
-.5072
(4.956)
-.5048
(4.881)
-.5227
(8.124)
Number of Shootings Two
Months Ago
. . . .1039
(1.084)
.07761
(0.812)
. . . .0354
(0.415)
.0384
(0.443)
Number of Shootings
Three Months Ago
. . . . . . .1813
(1.880)
. . . . . . -.099
(1.057)
Number of New York
Times’ Stories in the Front
Section in Previous Month
-.2057
(1.116)
-.1825
(0.983)
-.1111
(0.601)
-.1367
(0.921)
.0411
(0.208)
-.313
(1.317)
-.320
(1.312)
0.0478
(0.373)
Number of New York
Times’ Stories in the Front
Section Two Months Ago
. . . -.09145
(0.498)
-.0634
(0.348)
. . . -.551
(2.869)
-.519
(2.649)
Number of New York
Times’ Stories in the Front
Section Three Months Ago
. . . . . . .0428
(0.237)
. . . . . . .0132
(0.070)
50
Percentage of the Nation’s
Population Covered by
Shall Issue Laws
-9.432
(3.616)
-8.7729
(3.212)
-7.5203
(2.711)
-6.9996
(3.313)
-8.860
(2.241)
-9.201
(2.330)
-9.152
(2.266)
-8.174
(2.660)
Monthly Time Trend .01449
(4.117)
0.139
(3.719)
.0113
(2.996)
.0107
(3.810)
.0133
(2.451)
.0136
(2.512)
.0143
(2.567)
.0118
(2.842)
Intercept .6236
(1.431)
.5393
(1.221)
.2627
(0.571)
.9917
(2.851)
-.7686
(1.117)
-.5465
(0.769)
-1.165
(1.408)
-.0769
(0.159)
Chi Squared or F-statistic 35.36 36.68 40.00 2.47,
DW=2.005
48.05 54.80 55.29 6.23,
transformed
DW=2.008
Number of Observations 227 226 225 227 226 224 222 225
51
Table 12
Simultaneous Logit-Tobit Estimates
(The regressions control for sex, race, age; population, population squared, state unemployment rate, state poverty rate,
real per capita personal income, unemployment payments, income maintenance payments, retirement payments, arrest
rate for murder and state and year fixed effects. The first stage estimates do not report the regional dummy variables that
were in the regression. Absolute t-statistics are shown in parentheses.)
Second
Stage
Estimates
Endogenous Variables
Exogenous
Variables
Murders in
Multiple
Victim Public
Shootings Per
100,000 People
Injuries in
Multiple
Victim Public
Shootings Per
100,000 People
Murders and
Injuries in
Multiple Victim
Public Shootings
Per 100,000 People
Murders in
Multiple Victim
Public Shootings
Injuries in
Multiple Victim
Public Shootings
Murders and Injuries
in Multiple Victim
Public Shootings
Shall Issue
Law Dummy
Variable
-.023
(1.647)
-.0743
(3.110)
-.0921
(2.271)
-2.6746
(1.674)
-10.36
(4.371)
-13.166
(2.989)
Chi-Square 330.90 421.88 365.44 580.15 723.25 690.13
Number of
Observations 892 892 892 892 892 892
52
First Stage
Estimates Exogenous Variables
Endogenous
Variable
NRA
Membership
Rate
Lagged
Violent
Crime Rate
Lagged
Property
Crime Rate
Change in
Violent
Crime Rate
Change in
Property
Crime Rate
% Rep. Pres. in
State Vote * Year
Dummy1977-78
% Rep. Pres. in
State Vote * Year
Dummy1979-82
Shall Issue
Law
Dummy
Variable
.000302
(1.682)
-.00259
(5.004)
.00025
(2.642)
.00159
(1.354)
.00013
(.916)
.043
(2.558)
.038
(2.641)
First Stage
Estimates
(cont’d)
Exogenous Variables
Endogenous
Variable
% Rep. Pres. in
State Vote * Year
Dummy1983-86
% Rep. Pres. in
State Vote * Year
Dummy1987-90
% Rep. Pres. in
State Vote * Year
Dummy1991-94
% Rep. Pres. in
State Vote * Year
Dummy1995
% of the
State that is
White
% of the
State that is
Black
Chi-Square
Shall Issue
Law Dummy
Variable
.033
(2.646)
.047
(3.468)
.0942
(4.932)
.083
(5.116)
.0314
(2.226)
.031
(1.683)
161.55