Philip F. Lee
July 8, 2008 (revised July 21, 2008)
[ Op Eds ] [ Briefs ] [ Home ]
The handgun ban wouldn't keep a D.C. resident from purchasing a rifle or shotgun to be used for suicide. So, a claim, that D.C. had fewer suicides just because of this ban on handguns, doesn't pass a smell test on first impression. But we don't have to limit ourselves to an initial impression. Suicide and homicide data should show whether the ban had real consequences; this data doesn't favor Vedantam's assertions. Vedantam's mistake is to rely on a data analyis performed by others without examining that analysis critically. Vedantam might have found the flaws in the analysis performed by Loftin, McDowall, Wiersema, & Cottey (LMWC) by performing "due-diligence" on their paper to see what others had to say about their analysis. Had he done so, he might have found comments that the LMWC paper based their conclusions on suicide counts and not rates. Using counts during a time of decreasing D.C. population was either an act of self-delusion or deception. Suicide counts are affected by population changes but suicide rates (per capita rates) are not (see an early criticism in the 1994 article "Evaluation of the Loftin (et.al) District of Columbia Study" by Dean Payne or the 6/26/08 comment by Jim Lindgren; see also Jim Lindgren's second comment made on 7/9/08). This note assembles and discusses the relevant data.
Annual suicide rates during the period of interest are shown in the figure "DC, VA, MD Suicide Rates". The ban became effective in September 1976 (three-quarters of 1976 had passed). Ignoring the ban year as a year of transition, we see D.C. suicide rates did decline in the year after the ban (1977) compared to the year before (1975), but suicide rates increased again in 1978 and 1979. The fluctuation from 1975 to 1977 to 1979 is of a similar size experienced by D.C. from 1979 to 1981 to 1982, from 1962 to 1971, and other fluctuations shown by the data that are unrelated to the handgun ban. So, the suicide decrease in 1977 after the ban appears to be only a coincidence fluctuation.Vedantam doesn't mention in his editorial, but the 1991 LMWC paper claimed that the D.C. handgun ban "coincided with an abrupt decline in homicides by firearms." Like LMWC's claim for suicides, much of the reductions in homicide counts in D.C. result from population reductions during the period. If LMWC had used homicide rates to obtain a population independent result, they might have seen that D.C. performed worse on homicides, with its handgun ban, than surrounding Maryland and Virginia without handgun bans.
Annual homicide and population counts are given in Table 1 at the end of this note. These data were obtained from a Disaster Center compilation of annual FBI Uniform Crime Reports (UCRs). Suicide counts, Table 2, were obtained from U.S. Public Health Service source documents between 1960 and 1980 and from the CDC WISQARS Injury Mortality Reports for 1981 to 1988 (see Suicide Data References below).
The data shows that D.C. decreased from 809,000 people in 1968 to 620,000 in 1988. During the five years prior to the ban (1971-1975) D.C.'s population averaged 734,800 and during the five years after the ban (1977-1981) it averaged 658,247 which was a decrease of 10.4% in population from before to after the D.C. handgun ban. This decrease in the number of people living in D.C. should account for a 10.4% average reduction in homicides or suicides over the same periods even if there had been no public safety effect from the handgun ban. To distinguish the effects on homicides by the D.C. handgun ban from any effects resulting from population changes, we compare D.C.'s homicide rates with the rates from the surrounding states of Virginia and Maryland.
D.C.'s homicide rates peaked at 38.3 per 100,000 in 1973, decreased to 26.8 in 1976, held steady around 28 for the next few years and then increased to 35 in 1981.
In 1988, the D.C. rate increased significantly to 59.5. The graph of these rates, titled "DC, MD and VA Homicide Rates," shows no evidence that the D.C. gun ban improved homicide rates.Virginia and Maryland state data are shown for comparison in the same figure. Unlike D.C., Virginia and Maryland have growing populations. Maryland's homicide rate peaked in 1972 at the rate of 12.5 and fluctuated in the range of 7.9 to 10.1 between 1976 to 1988. Virginia's homicide rate peaked at 11.8 in 1972 and thereafter trended down to 7.8 by 1988.
D.C.'s homicide rates averaged over five-year periods just before (1971-1975) and just after (1977-1981) the ban decreased from 35.4 to 30.0, a 15.3% decrease. By comparison, Maryland's and Virginia's homicide rates averaged over the same five-year periods decreased from 11.5 to 9.1 and 10.9 to 8.7 or a 20.9% decrease for Maryland and 20.0% decrease for Virginia. Over the period examined, both Maryland and Virginia had larger precentage decreases in homicide rates than did D.C. More importantly, by 1988 D.C. is setting a new record high for homicides while both Maryland and Virginia rates remain well below their 1970's peak.
D.C. averaged 260 homicides during 1971-1975 and 197 during 1977-1981, a decrease of 24% from before to after its handgun ban. It was such a decrease that led LMWC wrongly to claim the ban caused the decrease. During those periods, the D.C. population decreased by 10% as discussed above. So, 10% of the decrease in homicides can be attributed to population decreases leaving only 14% not attributable to population changes. By comparison Maryland's homicide averages changed from 467 to 379, a 19% decrease despite a 3% growth in population (a real decrease of 22% considering the population growth), and Virginia's from 526 to 456, a 13% decrease despite an 8% growth in population (21% considering population growth) for the same periods.
The combined homicides and suicides in the three jurisdictions since the 1976 District handgun ban shows the District has far higher combined homicide and suicide rates each year than either Maryland or Virginia. This higher combined violent death rate year after year in the District need to be explained before anyone should believe the it's handgun ban had a beneficial effect.