Lott starts with a false claim ``Even my most determined critics concede one
point: violent-crime rates fell at the point on time that the
right-to-carry laws went into effect.'' As far as I know
none of Lott's critics have conceded this point, possibly
because there is no good evidence that it is true (as shown
Lott goes on to argue that the
reductions were not just the result of crime trends because
``the size of the drop is closely related to the number of
permits issued'' Since the number of permits just
increased with time, this is also consistent with a crime trend.
``it is not just the number of permits, but also the type of
people who obtain permits is important'' Lott doesn't have any data
on the type of people who obtain permits. What he does is fit a
model that predicts the number of people that obtain permits in a
state. (Incidentally, this model predicts that Texas issued a negative
number of permits in the years after it passed its shall-issue law.)
He then finds that the number of permits predicted by this model is
correlated with crime decreases. He then infers that since this
model contains a term that the number of permits issued is reduced
by an increased cost of a permit, and since poor people might be
expected to be more deterred by higher fees that this means that
poor people getting permits has a greater effect on crime than
However, because the model he fits contains a term that includes the
number of years since the carry law was passed, this result is also
consistent with a crime trend.
``why do violent-crime rates start rising in adjacent counties
in states without right-to-carry laws?''. The
answer is that they don't.
``as the period of time studied gets progressively longer, the
results are less likely to be due to crime cycles, since any
possible crime `cycles' involve crime not only going down but also
I reviewed Lott's results to see how they changed as the time period
got longer. Unfortunately, since the specification of the model in
table 9.1 is different from that in tables 4.8 and 4.13 the only
comparison that can be made is that between the results in
table 4.8 and 4.13. This shows the effect on violent crime
lessening from -0.9% to -0.5% when the period of time studied
increased by two years. This seems more consistent with a crime
cycle than with Lott's theory.
Here Lott makes a very strange argument--that including nonlinear
trends for regions is reasonable but including nonlinear trends for
individual states makes no sense. If it reasonable for crime rates to
follow nonlinear trends in regions, then surely it is reasonable
for crime rates to follow nonlinear trends in individual states. Lott
argues that a crime rate that was rising before the carry law and as a
result of the law started falling after the law would be fitted by a
nonlinear trend, making it look like the law had no effect. Equally,
you could argue that if a crime rate that following a nonlinear trend
which peaked before the carry law, Lott would fit it with a model that
showed an increase until the law was passed and then a decrease
afterwards, making it look like the law had an effect when nothing
special happened to crime rates when the law was passed.
To resolve the situation it would be necessary to fit a model with
both individual nonlinear trends and a parameter that allowed the
trend to change at the time of the carry law. We could then test to
see if the change in the trend was statistically significant.
Unfortunately, while some of the models considered by Lott contain a
parameter that measures the change in a trend at the time of the carry
law, none of them contain nonlinear trends for individual states (and
so would be expected to fail the Heckman-Hotz tests for
misspecification). And while the model considered by Black and Nagin
contains nonlinear trends for individual states it only allows the
carry law to change the crime rate, and not the crime trend.
To summarize, Lott is wrong when he claims that nonlinear trends in
each state make no sense, and we don't know whether his results on
changes in crime trends associated with the carry laws would still be
significant if nonlinear trends in each state were controlled for.
In the original paper written with David Mustard, the model with an
immediate and constant effect was clearly the primary one--the
results from it were the ones given in the abstract and the other
models considered were variations on it. Black and Nagin were
responding to that paper, so it was quite reasonable for them to focus
on that model. They found that that model failed the Heckman-Hotz tests for
Now that Lott has moved to a different preferred model, the
appropriate way to answer their objections would be test to see if the
new model satisfies the Heckman-Hotz tests. Lott has not done this.
Lott argues that the changes in crime rates are unlikely to have been
caused by changes in illegal drug use because:
``Neighboring counties without right-to-carry laws directly on
the other side of the border experienced an increase in violent
crime precisely when the counties adopting the law were experiencing
a drop.'' This is incorrect. Neighboring counties
did not have an increase in violent crime.
``The timing of changes in right-to-carry laws also makes their
argument less plausible.'' However, the changes in crime rates
caused by drug use do not have to exactly coincide with the carry
laws do make it look as if the carry laws reduced crime in the
states which did not experience an increase in drug use.
Lott argues that the data strongly suggests that criminals respond to the
actual increased risk rather than the announcement of the carry laws.
However, all of the data he mentions is also consistent with a crime
trend and Lott does not mention the data that shows that
the increased risk was negligible.
Lott actually responds again to the same Kleck quote in section 9.14,
except that in that section he mistakenly attributes the quote to me.
Lott writes ``Let me try to explain the meaning of Kleck's quote'',
and then fails to explain the meaning of the quote.
The explanation of the quote is simple.
As the quote from
Kleck inside the front cover of ``More Guns, Less Crime'' indicates,
Kleck has no problem with the design or methodology of Lott's study.
Kleck disagrees with Lott's conclusions because Kleck's own research
indicates that a significant number of people carry guns for
protection even without carry laws. The number of carry permits
issued represents a relatively small increase in the number of people
carrying, insufficient to cause the crime decreases that Lott claims
the carry laws caused.
It is true that permit holders are less likely to commit crimes than
the average person. This is because a people with criminal records
are not eligible for permits. However, Lott understates the risk they
pose to others. In
Texas, three permit holders have been convicted of murder, while
another permit holder used his gun to kill himself after committing
a murder .
Lott completely fails to address any of the criticisms of his use of
these surveys. For details,
In a later paper on safe storage laws  Lott and
Whitley do not
use these exit polls. Instead they use GSS surveys to measure how gun
ownership changes as a result of a state passing a safe storage law.
Using these polls they find that gun ownership declined by one
percentage point per year in the states with the laws and argue that
the laws caused increases in crime rates.
Lott does not explain why, after stoutly defending his use of the exit
polls to measure changes in gun ownership at the state level he
abandoned them for his later paper. One possible explanation is that
the exit polls say the opposite thing to the GSS surveys. The exit
polls show substantial increases in gun ownership in the
states that passed safe storage laws. I computed a regression
relating the change in gun ownership as measured by the exit polls to
the number of years that a safe storage law had been in place and
found that the laws were associated with a 0.06 percentage point per
year increase in gun ownership rates. This increase is not
statisitically significant, but it is the opposite sign to Lott's
result using the GSS surveys.
Since the exit polls give the opposite result from the GSS surveys it
is quite possible that if Lott has conducted his analysis in chapter 3
using the GSS surveys he would have gotten the opposite result and
found that more guns were associated with more crime.
It is not good practice to choose your data source to get the result
While Lott shows that if you look at changes in the trends rather than
the changes in the rates there is a still a significant change
associated with the law in the states other than Maine and Florida,
his results also show that issuing a permit in Maine and Florida had
four times the effect on violent crime as issuing a permit in the
other states. They also show that issuing a permit in Maine and
Florida reduced property crime just as much as it reduced violent
crime. This suggests hat the results in those states were the product
of general decreases in crime relative to other states and not the
Since Lott has county level data on permits for some states, an
appropriate response here would have been to have used this data to
test to see if permits were actually issued at a greater rate in more
populous counties. It is puzzling that he did not so this. The one
extra piece of information that he provides in his answer seems to be
evidence against his theory--he finds bigger drops associated with a
change from discretionary to non-discretionary permits than from no
permits to non-discretionary permits, even though you would expect a
greater change in the number of permits in a change from no permits
than in a change from discretionary permits.
I agree that because of different circumstances citizens do not
necessarily require the same training as police, but
Lott contradicts himself in his comments here, stating both that
``Training requirements improve the deterrence effect for
concealed-handgun laws'' and that the ``effect of increased training
is clearly to reduce the deterrent effect''.