65. For example, a December 1998 press release on children and gun violence had South Carolina and Colorado ranking similarly in terms of how liberal their right-to-carry laws were, but by January 1999, in a press release examining the change between 1992 and 1997, Colorado was listed as having a more restrictive law than South Carolina. The only motivation that I can conjecture for the change was that it helped get them the different results that they wanted.
66. "In stark contrast, a review of the national Uniformed Crime Reporting data, which is compiled by the FBI each year from state and local law enforcement agencies, indicates that the violent crime rate has fallen in all states by an average of 19 percent from 1992-97" (Richard Cook, "Don't Buy the Pro-Gun Arguments," Kansas City Star, Mar. 11, 1999, p. B7).
67. Peter Squires, "Review of More Guns, Less Crime," British Journal ofCriminology 39, no. 2 (Spring 1999): 318-20.
68. My book does not even cite this quotation, though I mentioned it in an earlier research paper because it was "quite relevant" to the debate over concealed handguns: it illustrates both the possibility of deterrence and the fears about the possible disasters that such laws could lead to.
Still other recent discussions in medical journals continue claiming that the nondis-cretionary concealed-handgun laws for "several counties ... were misclassified" and that the National Academy of Sciences deemed it inappropriate to account for arrest rates when researchers tried to explain changes in crime (see Arthur Kellermann and Sheryl Heron, "Firearms and Family Violence," Emergency Medicine Clinics of North America, Aug. 1999, pp. 699—708). Of course, responses 4 and 9 on pages 132-33 and 142 in this book addressed the first concern and page 18 discussed the second one.
69. http://www.handguncontrol.org/gunowner/statflaw.htm.
70. Doug Weil, Handgun Control's research director, provided the only response that I know of to my research on the Brady law by claiming that "Since John's data does not cover the years following implementation of the Brady Act, it's hard to know how he can claim to have studied the impact of the Brady law on crime rates or criminal access to guns" ("More Guns, Less Crime?: A Debate between John Lott, Author of More Guns, Less Crime, and Douglas Weil, Research Director of Handgun Control, Inc.," an online debate sponsored by Time magazine, transcript from July 1, 1998.) In fact, my book examined data up through 1994, the first year that the Brady law was in effect.
71. Romesh Ratnesar, "Should You Carry A Gun? A New Study Argues for Concealed Waspons," Time, July 6, 1998, p. 48.
72. I responded by saying that he was doing more than simply reporting these statements as claims when he used phrases like "Lott dropped" or "the book does not account." More importantly, readers were likely to believe that he had looked at the material and that he would not print something, even if the critics claimed it was true, unless it was true. Again, he emphasized that his role was that of a reporter and not to take sides in the debate.
NOTES TO PAGE 207/301
I had called Romesh in part to tell him that I planned to send in a letter clarifying these points, and Time magazine did print a letter. Undoubtedly he played some role in guaranteeing that the letter was published, but it seems doubtful that the letter carried the same weight as a statement by the reporter about whether he could verify if the claims made against me were true. The letter in Time magazine was printed in the Aug. 3, 1998, issue under the heading "More about Concealed Weapons." It read:
While your piece "Should You Carry a Gun?" [July 6] was generally favorable toward my new book, More Guns, Less Crime, it contained seriously misleading statements. Despite accusations by some critics, my study on the effect that carrying concealed weapons has on crime absolutely did not ignore "counties that had no reported murders or assaults for a given year." In contrast to the tiny samples in previous work by others, I used data on all the counties in the U.S. that were available when I did the study on the years from 1977 to 1994. It is likewise false that I did "not account for fluctuating factors like poverty levels and police techniques." Among the factors I included in the analysis were poverty, income, unemployment, arrest and conviction rates, the number of police officers and police expenditures per capita, as well as the impact that the prevention of less serious crimes has on more serious ones.
73. Ivins, "More Guns, Less Crime? Are You Sure?"
74. Tom Teepen, "A Modest Proposal: Let's Arm the Teachers," Atlanta Journal and Constitution, Sunday, May 17, 1998, p. 2G.
75. The following letter of mine appeared in the Atlanta Journal and Constitution, May 24, 1998, p. 6B:
Tom Teepen's column "A modest proposal: Let's arm the teachers," Perspective, May 17), an attack on my new book "More Guns, Less Crime" (University of Chicago Press), contained misleading information. He claimed that "Lott can't fairly compare 1988 and 1996 exit polls on gun ownership, as he does, because the questions were asked differently." Yet on pages 36—37 in my book, I point out this fact and discuss in detail what impact this has on estimates of changing gun ownership.
Citing a paper in the Journal of Legal Studies, Teepen claimed that I make a "fundamental gaffe" by failing to consider other anti-crime variables. My book provides the first systematic national evidence and examines the crime, accidental gun death, and suicide rates for all 3,054 counties in the United States by year from 1977 to 1994. No other study on crime has attempted to account for anywhere near as many different factors that could have affected crime rates over time. Unlike the Centers for Disease Control and Prevention's claim that homes with guns were "more likely to experience suicide," or have "a member of the family killed by another member or by an acquaintance," I did not focus on data from only one or a few cities for only one year. There is no evidence that these claims are correct.
Obviously, bad things can happen with guns, but guns also prevent bad things from happening to people. The evidence in my book indicates that many more lives are saved than lost from gun ownership.
76. An editor at the Fort Worth Star-Telegraph, Bob Davis, was very helpful, and he took the time to read my book to evaluate whether a mistake had been made. He printed a response by me in his newspaper, and he asked Creators Syndicate, which distributes Ms. Ivins's commentary, to make the response available to other newspapers around the country that carried Ms. Ivins's column. Unfortunately, despite repeated promises by Creators to do so, they never followed through on this.
302 / NOTES TO PAGES 207-214
Even John Lott admits that 58 percent of homicides are committed either by family members or friends and acquaintances, not criminals. (Richard Scribner, [director of the Injury Control Research Center], "More Guns Don't Mean a Safer Society," New Orleans Times-Picayune, Apr. 28, 1999, p. B6)
Dr. Lott's own analysis accounts for only about 10 percent of why some crime rates have fallen. We need to explain the other 90 percent before concluding that the "best" social policy is to carry more handguns. (Shela Van Ness, "More Guns, Less Crime? This Isn't Just a 'Good Guy' vs. 'Bad Guy' Issue," Chattanooga Times / Chattanooga Free Press, May 9, 1999, p. HI)
For the first point, not only do I not "admit" this, but my book points out that this claim is extremely misleading because the term "acquaintances" primarily includes rival gang members killing each other or drug buyers and drug sellers killing each other. As to the second point, the estimates shown in this book explain about 80—95 percent of the variation in crime rates.
78. The Chronicle of Higher Education noted that the opposition to my book also showed up in the University of Chicago Press, this book's publisher. The Chronicle reported that "The book also caused a mini-revolt at Chicago, where salespeople initially blanched at the prospect of pitching it to bookstores. Some cited personal views about guns; others thought that the book would alienate booksellers" (Christopher Shea, '"More Guns, Less Crime': A Scholar's Thesis Inflames Debate over Weapons Control," Chronicle of Higher Education, June 5, 1998, p. AH).
79. In this case, the dummy must be interpreted as whether the law ra
ised or lowered the crime rate as quickly as the quadratic time trend would predict.
80. This example is taken from David D. Friedman's Web site, www.best.com/~ddfr/ Lott_v_Teret/Lott_Mustard_Controversy.html.
81. Virtually identical complaints have been posted on the Handgun Control, Inc., Web site, where Handgun Control writes: "To this day, John Lott has failed to provide any statistical evidence of his own that counters Black and Nagin's finding that Lott's conclusions are inappropriately attributed to changes in concealed-carry laws. Until Lott can do this, it is inappropriate for him to continue to claim that allowing more people to carry concealed handguns causes a drop in crime."
82. Dan A. Black and Daniel S. Nagin, "Do Right-to-Carry Laws Deter Violent Crime?" Journal of Legal Studies 27 (Jan. 1998): p. 213.
83. What is mystifying to me is how others have also continued to make this claim. Hashem Dezbakhsh and Paul H. Rubin claim that "We believe that Lott and Mustard's findings are suspect, mainly because of the way they parameterize and measure the effect of permissive handgun laws on crime. They model the effect as a shift in the intercept of the linear crime equation they estimate at the county level. This approach is predicated on two assumptions: (i) all behavioral (response) parameters of this equation (slope coefficients) are fixed (unaffected by the law), and (ii) the effect of the law on crime is identical across counties" (Hashem Dezbakhsh and Paul H. Rubin, "Lives Saved or Lives Lost? The Effects of Concealed-Handgun Laws on Crime," American Economic Review Papers and Proceedings, May 1998, p. 468).
84. http://www.best.com/~ddfr/Lott_vJTeret/Friedman_on_B_and__N.html. A great deal of debate about my research and other gun-related research takes place on the Internet in discussion groups such as talk.politics.guns or on Web sites such as David Friedman's, which allows for a very detailed discussion of the issues. The give and take also allows people to ferret out the weaknesses and strengths of different arguments.
85. Benson, "Review of More Guns, Less Crime," p. 312.
86. Ayres and Donohue mention in a footnote that "Lott was not unaware of the possibility that crack influenced the level of crime and some regressions in the book control for the price data for cocaine (p. 201, fh. 8), but the quantity of crack sold in
NOTES TO PAGES 216-218/303
discrete geographic markets instead of its national price would be much more probative." Even though I had given them the price data, they apparently had not had time to examine it and realize that it was county-level data.
While simply using the price does not allow one to perfectly disentangle local differences in demand and supply, arbitrage basically assures that, except for short periods of time, the differences in prices between these local markets will equal differences in selling costs. If the total cost of selling cocaine were the same in two different cities, any price differentials resulting from sudden shifts in demand would cause distributors to send cocaine to the city with the higher price until the price had fallen enough that the prices between the two cities were equal. Distributors could even remove cocaine from the low-price city and move it to where it could obtain a higher price. Sellers in a city could also hold inventories and not sell their cocaine during periods with unusually low demand. To the extent that it is costly to move drugs instantly between different cities or to store drugs, any price differentials in the short run can be due to demand shifts, but since we are dealing with a period of a year, it seems difficult to believe that any non-cost-based price differentials will not be arbitraged away.
The bottom line is that if price differentials exist for long periods of time (and we would pick up precisely these differences by using average yearly prices), any price differentials will be cost based. Now cost differences can arise for many reasons (e.g., differences in law enforcement, wage differentials for workers, differences in rental prices for "business" spaces, etc.). The concern is not why these cost differentials exist, but simply that they do and that this will be related to the accessibility of drugs.
Suppose, for example, that it costs $8 to sell an ounce of cocaine in Atlanta and $3 in Washington, DC. If the price of a one-ounce bag of cocaine were $10 in Washington and $13 in Atlanta, then minus these selling costs an importer of illegal drugs would make $7 in Washington and $5 in Atlanta. Where is he going to ship more of his drugs? Clearly Washington, and he will continue doing so until the relative price net of these costs in Washington falls until the difference between the two markets is $5.
87. An example of one of the other criticisms is by Ayres and Donohue where they write that "the ultimate criticism of Lott will be that the model is too flawed to provide any information on the effect of the law.... One of the strongest results to emerge from Lott's book is that shall issue laws, as he models them, lead to higher property crime. If you don't believe this, then you cannot endorse any of Lott's findings. But, to believe that property crime rose you must believe that the rate of robbery fell, because the only reason that more concealed handguns would cause property crime to go up is that some other money-generating activity became less available or less attractive. One would hardly expect that someone desiring to beat up an individul would instead decide to steal a car if the assaultive option were foreclosed. But since the robbery results are arguably weak, it is hard to tell a convincing story that would explain the alleged shift from violent crime to property crime that the Lott model attributes to shall issue laws" (Ian Ayres and John J. Donohue HI, "Nondiscretionary Concealed Weapons Laws: A Case Study of Statistics, Standards of Proof, and Public Policy," American Law and Economics Review 1, nos. 1—2 (Fall 1999): 436-70.
88. The "recidivism" referred to by Ayres and Donohue is actually not a good measure for what they are discussing, since recidivism refers to whether criminals keep on committing a crime after they have been punished by the legal system.
89. Ayres and Donohue raise another issue that should be discussed at least briefly, and that is the use of the percentage of a state's population that is in prison as an enforcement variable. They find that including this variable strengthens the results, but while the variable provides some information, there are some important theoretical problems with it. One problem is that the prison population and the crime rate are simply in different units. The prison population measures a "stock," while the crime rate represents a "flow." The simplest comparison is between the amount of water in a bathtub (a stock)
304 / NOTES TO PAGES 219-232
and the rate at which water is flowing into the bathtub (a flow). The amount of water in the bathtub is only loosely related to the current flow into it because it depends upon not only flows in previous periods but also the rate at which water is flowing out of it. A second problem is that I have focused on county-level data because of the heterogeneity in law enforcement across counties within a state, and this variable is available only at the state level.
90. For example, Sarah Brady, "Q: Would New Requirements for Gun Buyers Save lives? Yes: Stop Deadly, Unregulated Sales to Minors, at Gun Shows and on the Internet," Insight, June 21, 1999, p. 24; or "More Guns, Less Crime? A Debate between John Lott and Douglas Weil."
91. Gary Kleck, Targeting Guns: Firearms and Their Control (Hawthorne, NY: Aldine de Gruyter Publishers, 1997), p. 371.
92. This is true whether one uses the 430,000 instances in 1997 in which crimes with guns were reported to police in the Uniform Crime Report or the number that is about twice as large from the National Crime Victimization Survey.
93. Stephanie Elizondo Griest, "Group: Arrest Data Show Flaw in Concealed-Gun Law: Permit Holders Have Been Arrested 2,080 times: NRA Says Low Conviction Rate Proves Licensees Abide by Law," Austin American-Statesman, Mar. 23, 1999, p. B2.
94. John Lott, Jr., "License to Kill? Careful Look at Critical Study Actually Backs Gun Permit Holders " Dallas Morning News, Feb. 8, 1998, p. J6.
95. Jonathan Rauch, "And Don't Forget Your Gun," National Journal, Mar. 20, 1999.
96. Based upon a telephone conversation with the Al
aska bureau responsible for issuing permits.
97. Source: Lt. Bill Whalen, Arizona Department of Public Safety, 602-223-2704. Peak issuance of permits was November 1998, when 63,040 permits were issued. The renewal rate is about 42 percent. Based on that number and the fact that about 26,000 permits were issued by July 1995, over 76,000 permits must have been issued during the period.
98. http://licgweb.dos.state.fl.us/stats/cw_monthly.html.
99. Based upon a telephone conversation with the Indiana State Firearms Bureau.
100. Revocation rate obtained from "North Carolina Handgun Permit Statistics by County from Dec. 1,1995 and Aug. 4,1999" (available on the North Carolina state government Web site). The other information is based upon a telephone conversation with Julia Nipper and Susan Grissom.
101. James N. Thurman, "As More Carry Hidden Guns, Who's Safer?" Christian Science Monitor, Sept. 1, 1999, p. 1.
102. Information from Steve Anderson with the Utah State Firearms Bureau.
103. Based on an E-mail message sent to me by Mr. Wilson of the Criminal Department of Investigation, Wyoming). His telephone number is 307-777-7181.
104. Frank Main, "Taxpayers Pay Big Part of Gunshot Victims' Bills," Chicago Sun-Times, Aug. 4, 1999, p. 30.
105. Another survey by gun-control advocates claims that "four million legal handgun owners sometimes carried guns for protection 'in connection with work.' Two-thirds of those who carried handguns said they kept them in their vehicles, while the others
said they sometimes carried them The researchers said about 56 percent of those who
carried handguns outside of work did so fewer than 30 days per year, while 22 percent said they rarely left home without a gun" (Will Hacker, "Majority of Owners Cite Security Concerns " South Bend Tribune, June 29, 1997, p. A6).
More Guns Less Crime Page 38