|
Article Excerpt
I. INTRODUCTION
The Federal Uniform Drinking Age Act (FUDAA), signed by President Ronald Reagan on July 17, 1984, threatened to withhold highway construction funds from states that failed to increase their minimum legal drinking age (MLDA) to 21 by October 1, 1986. Some states complied without protest, but many states balked and sued the federal government to prevent implementation of the Act. In South Dakota v. Dole (1987), however, the U.S. Supreme Court ruled the Act constitutional. The Court decided that the "relatively small financial inducement offered by Congress" was not so coercive "as to pass the point at which pressure turns into compulsion." The Court argued, in particular, that reducing traffic fatalities among 18- to 20-yr-olds was sufficient reason for the federal government to intervene in an arena traditionally reserved to states. (1)
Research subsequent to the Court's decision appears to confirm that raising the MLDA saves lives and much of it points to the FUDAA in particular. Relying on this research, the National Highway Traffic Safety Administration (NHTSA) attributes substantial declines in motor vehicle fatalities to federal and state traffic safety policies, particularly the MLDA21. For example, NHTSA estimates the cumulative number of lives saved by the MLDA21 at 21,887 through 2002 (NHTSA's National Center for Statistics and Analysis 2005a).
We challenge the view that MLDAs reduce traffic fatalities based on three findings. First, the overall impact estimated in earlier research is driven by states that increased their MLDA prior to any inducement from the federal government. Second, even in early-adopting states, the impact of the MLDA did not persist much past the year of adoption. Third, the MLDA has at most a minor impact on teen drinking.
The remainder of the article is organized as follows. Section II outlines the history of the MLDA and reviews the preexisting literature. Section III examines aggregate trends in the key variables. Section IV describes the state-level data set and presents panel estimates of the relation between the MLDA and the traffic fatalities. Section V investigates the effects of the MLDA on teen drinking.
II. HISTORICAL BACKGROUND AND PRIOR LITERATURE
When the United States repealed Alcohol Prohibition in 1933, the Twenty-First Amendment left states free to legalize, regulate, or prohibit alcohol as they saw fit. Most legalized but also enacted substantial regulation. This new regulation typically included an MLDA.
Table 1 gives the MLDA set by each state after Prohibition ended. (2) State reactions to federal repeal varied from Alabama maintaining state-level prohibition to Colorado legalizing alcohol without a minimum drinking age. In general, states set an MLDA between 18 and 21. In 1933, 32 states had an MLDA of 21 and 16 states had an MLDA between 18 and 20. With few exceptions, these MLDAs persisted through the late 1960s.
Between 1970 and 1976, 30 states lowered their MLDA from 21 to 18. These policy changes coincided with national efforts toward greater enfranchisement of youth, exemplified by the Twenty-Sixth Amendment granting 18- to 20-yr-olds the right to vote. The reasons for lowering the MLDA are not well understood and may have varied by state. Perhaps the changes reflected Vietnam-era logic that a person old enough to die for America is old enough to drink (Asch and Levy 1987; Mosher 1980). Whatever the reasons, the lower MLDAs "enfranchised" over five million 18- to 20-yr-olds to buy alcohol (Males 1986, p. 183).
Soon after the reductions in the MLDAs, empirical studies claimed that traffic collisions and fatalities were increasing in states that lowered their MLDA. Most prominently featured in congressional discussion were two comprehensive, multistate studies on the "life-saving" effects of raising the MLDA--the Insurance Institute for Highway Safety study and the National Transportation Safety Board study. According to Males (1986), both studies were referred to more than 50 times in the House and Senate debates, "almost to the exclusion of all other research on the question" (p. 182). (3) These research findings played a key role in reversing the trend toward lower MLDAs. The justification for the FUDAA, espoused by organizations including the Presidential Commission on Drunk Driving, the American Medical Association, and the National Safety Council, was that higher MLDAs resulted in fewer traffic fatalities among 18- to 20-yr-olds (Males 1986).
After passage of the FUDAA, all states adopted an MLDA21 by the end of 1988. Table 2 gives the most recent date each state switched to an MLDA21. Several states were early adopters (Michigan, Illinois, Maryland, and New Jersey), increasing their MLDAs before passage of the FUDAA. Other states were less eager to change. For example, Colorado, Iowa, Louisiana, Montana, South Dakota, Texas, and West Virginia passed MLDA21 legislation, but each provided for repeal if the FUDAA were held unconstitutional (DISCUS Office of Strategic and Policy Analysis 1996). Texas and Kansas enacted "sunset provisions" allowing the MLDA to drop back to previous levels once federal sanctions expired (DISCUS Office of Strategic and Policy Analysis 1996). When the Supreme Court upheld the constitutionality of the FUDAA, states faced a strong incentive to maintain an M LDA21. Nevertheless, the differences in how states initially responded suggest a policy endogeneity that needs to be addressed.
Several authors have recently summarized the MLDA literature, so we do not review specific articles in detail (Shults et al. 2001; Wagenaar and Toomey 2002). Overall, the existing research finds a negative relationship between the MLDA and traffic fatalities, but most studies omit key variables and mainly analyze either cross-sectional data from 1 yr or time series data in one state (Ruhm 1996). (4)
The most important exception to this summary is Dee (1999), who uses state-level panel data and controls for state fixed effects, state trends, year dummies, and other variables. Dee's estimates "suggest that the movement to [a] higher MLDA reduced ... traffic fatalities by at least 9%" (Dee 1999, p. 314). Dee's analysis forms the starting point for the empirical work below.
In addition to considering the impact of the MLDA on traffic fatalities, earlier literature also considers how the MLDA affects teen drinking. (5) Kaestner (2000) explains that most studies use cross-sectional data and fail to control for unmeasured state characteristics affecting both alcohol consumption and minimum drinking ages. Again, Dee (1999) is an exception. Using the same techniques just described, Dee concludes that moving away from an MLDA of 18 is associated with a reduction in heavy teen drinking of 8.4%. More recently, Carpenter et al. (2007) extend Dee's sample by 11 yr to include data through 2003. They find that "nationwide increases in the MLDA ... reduced youth drinking by about four percent relative to pre-existing levels (p. 2). (6) They acknowledge, however, that adoption of the MLDA21 might have increased underreporting.
III. AN OVERVIEW OF THE AGGREGATE DATA
Before examining state-level regressions that relate traffic fatality rates (TFRs) to MLDAs, we examine aggregate plots of the key variables. The reason is that state-level data on traffic fatalities are not available until the mid-1970s, but aggregate data on total and 15- to 24-yr-old fatalities exist back to 1913. The 18- to 20-yr-old population is most relevant for the issues in this article, but data for this age range are not available until 1975. The 18-20 fatality rate and the 15-24 fatality rate are highly correlated, however, as shown in Figure 1, so examination of the 15-24 TFR is likely informative.
Figure 2 presents the TFR for the total population and for 15- to 24-yr-olds for the period 1913 2004. These two series follow similar patterns over the past 90 yr. Both TFRs increased from 1913 to 1969 and then decreased thereafter. This similarity fails to suggest a major impact of the MLDA, which should have affected the 15-24 TFR more than the total TFR. The marked decline in the TFR during this period also contravenes claims of a rapid increase in traffic fatalities after several states decreased their MLDAs between 1970 and 1973. The declines in the total and 15-24 TFR that began around 1969 long precede the adoptions of an MLDA of 21 in the mid-1980s.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
The data in Figure 2 do not control for the vehicle miles traveled (VMT) each year, which have increased enormously over the past century. Figure 3 shows that fatalities per VMT exhibit a persistent downward trend over the entire sample period. The 1524 TFR does seem to increase slightly beginning in the 1960s, even when controlling for VMT, but the decline returns around 1969 prior to passage of the FUDAA.
[FIGURE 3 OMITTED]
Figure 4 plots the average MLDA for all 50 states against the (VMT based) TFR for the 15-to 24-yr-old age cohort. (7) While the average MLDA remained at approximately 20 between 1944 and 1970, traffic fatalities continued to decrease for years and then increased. Then, in the early 1970s, several states lowered their MLDAs, reducing the average to below 19. Yet, the brief increase in TFRs that occurred in the latter half of the 1970s looks modest in comparison to the larger,...
|
