Manuel v. State

677 So. 2d 116

Opinion

677 So.2d 116 (1996)

Jody W. MANUEL, Stacey P. Foret, Burke G. Pierrotti and Wendell J. Manuel
v.
STATE of Louisiana; Honorable Edwin W. Edwards, Governor; Richard P. Ieyoub, Attorney General; J. William Pucheu, District Attorney; and Terry Pitre, Commissioner, Louisiana Office of Alcohol Beverage Control, Department of Revenue and Taxation.

No. 95-CA-2189.

Supreme Court of Louisiana.

July 2, 1996. Concurring Opinion July 12, 1996.

Richard P. Ieyoub, Attorney General, E. Kay Kirkpatrick, Roy Achille Mongrue, Jr., James Marshall Ross, Barbara B. Rutledge, James C. Hrdlicka, III, Thomas S. Halligan, Asst. Attorneys General, for Applicant.

John Larry Vidrine, Ville Platte, Camille F. Gravel, Jr., Alexandria, David Overlook Stewart, Brian S. Chilton, Washington, DC, for Respondent.

Henry M. Jasny, New York City, for Amicus Curiae Highway & Auto Safety and Mothers Against Drunk Drivers.

John Dowling Rawls, New Orleans, for Amicus Curiae Wedon A. Brown.

Concurring Opinion by Chief Justice Calogero July 12, 1996.

LEMMON, Justice.^[*]

This rehearing was granted to reconsider the earlier decision which held unconstitutional the statutory provisions that raised the minimum drinking age in this state to twenty-one. On original hearing, a majority of this court ruled that these statutes violated the equal protection afforded by La. Const. art. I, § 3 against arbitrary discrimination based on age. On reconsideration, we conclude that the statutes establishing the minimum drinking age at a level higher than the age of majority are not arbitrary because they substantially further the appropriate governmental purpose of improving highway safety, and thus are constitutional.

I

In 1986, the Legislature raised the minimum drinking age in the state from eighteen to twenty-one. See La.Acts 1986, No. 33, amending and reenacting La.Rev.Stats. 14:91.1, 91.2 and 91.5. While the new statutes made drinking alcoholic beverages by a person under twenty-one a crime, there were no penalties for selling alcoholic beverages to persons below the minimum age.

In 1995, the Legislature amended and reenacted La.Rev.Stat. 26:90A(1)(a) and (b), and 26:286A(1)(a) and (b); enacted La.Rev. Stat. 14:93.10-93.14; and repealed La.Rev. Stat. 14:91.1-91.5. See La.Acts 1995, No. 639. These statutes imposed penalties for selling alcoholic beverages to persons under twenty-one, thus closing the previous "loophole hole" which had made the minimum drinking age law unenforceable as a practical matter.

Challenging the constitutionality of Act 639 of 1995 and seeking both declaratory and injunctive relief, plaintiffs commenced this action in the form of a class action with two plaintiff subclasses—purchasers and retailers. The trial judge granted a temporary restraining order, prohibiting defendants from enforcing any prohibition against the purchase, possession, service or sale of alcoholic beverages to persons between the ages of eighteen and twenty-one pending the injunction hearing.

After the hearing on the preliminary injunction at which both sides presented evidence, the trial judge granted plaintiffs declaratory relief, ruling that Act 639 of 1995 constituted arbitrary age discrimination in violation of La. Const. art. I, § 3. The trial court further granted injunctive relief,^[1] prohibiting enforcement of the unconstitutional statutes. This direct appeal followed. La. Const. art. V, § 5(D).

On original hearing, a majority of this court affirmed the trial court's decision declaring the minimum drinking age provisions unconstitutional under La. Const. art. I, § 3. On the State's application, we granted rehearing to reconsider the correctness of that decision.

II

La. Const. art. I, § 3, provides:

No person shall be denied the equal protection of the laws. No law shall discriminate against a person because of race or religious ideas, beliefs, or affiliations. No law shall arbitrarily, capriciously, or unreasonably discriminate against a person because of birth, age, sex, culture, physical condition, or political ideas or affiliations. Slavery and involuntary servitude are prohibited, except in the latter case as punishment for crime, (emphasis added).

In asserting the constitutional challenge to the minimum drinking age statutes, plaintiffs rely on the special protection against arbitrary discrimination based on age set forth in the third sentence of La. Const. art. I, § 3.

In Sibley v. Board of Supervisors of Louisiana State Univ. and Agric. and Mechanical College, 477 So.2d 1094 (La.1985),^[2] we construed this third sentence of Section 3 to mean that when a statute classifies persons on a basis therein enumerated, the statute is unconstitutional unless the proponents of the statute prove this legislative classification "substantially furthers an appropriate state purpose." Id. at 1108. In so holding, we expressly repudiated the notion that the federal system of equal protection review should be used as a model for interpreting or applying this third sentence of Section 3, concluding that "the state constitution calls for more than minimal scrutiny of certain types of classifications, and assigns the state the burden of showing that such legislation is not arbitrary, capricious or unreasonable." Id. at 1107.

As we recently explained in Moore v. RLCC Technologies, Inc., 95-2621 (La. 2/28/96), 668 So.2d 1135, La. Const. art. I, § 3 sets up a spectrum for analyzing equal protection challenges based on discriminatory classifications. At one extreme are laws that classify persons based on race or religious beliefs. Under the second sentence of Section 3, such laws are repudiated completely. See Louisiana Associated Gen. Contractors, Inc. v. State through Div. of Admin., Office of State Purchasing, 95-2105 (La. 3/8/96), 669 So.2d 1185.

At the other end of the spectrum are laws that classify persons on any basis other than those bases expressly enumerated in Section 3. These laws are reviewed under the standard set forth in the first sentence, which has been construed to mean that every statutory classification must pass the minimum standard of being rationally related to a legitimate governmental purpose.^[3] Whenever a person disadvantaged by such a classification seeks to have the law declared unconstitutional, that person has the stringent burden of demonstrating that the law does not suitably further any appropriate state interest.

In the middle of the spectrum are laws that classify persons on the basis of the six grounds enumerated in the third sentence of Section 3: (1) birth, (2) age, (3) sex, (4) culture, (5) physical condition or (6) political ideas or affiliations. A law containing a statutory classification based on.any of the six enumerated grounds does not enjoy the usual presumption of constitutionality. Moreover, with that reversal of the ordinary presumption of constitutionality comes a reversal of the rule that ordinarily places the burden of proof on the party seeking a declaration of unconstitutionality. When the court reviews such a law, the burden is on the proponent of the classification and the standard of review is heightened, requiring the proponent to establish that the classification is not arbitrary, capricious or unreasonable because it substantially furthers an appropriate governmental objective.^[4]Moore v. RLCC Technologies, Inc., 95-2621, pp. 9-10 (La. 2/28/96), 668 So.2d 1135, 1140-41.

In summary as to the two lower levels of scrutiny, the standard under the Louisiana Constitution for determining the constitutionality of a statute that sets up a classification on any basis not enumerated in Section 3 is whether the classification under the presumptively constitutional statute furthers any legitimate governmental interest, and the opponent has the burden of proof. On the other hand, the standard for determining the constitutionality of a classification based on age is stated expressly in La. Const. art. I, § 3—the classification cannot "arbitrarily, capriciously or unreasonably discriminate." Because age classification is specifically enumerated in Section 3 and because an age classification must have a non-arbitrary basis, the burden of proof is on the proponent of constitutionality to show that the statute establishing such a classification substantially furthers an appropriate governmental purpose.

The principal differences in the standards for the two lower levels of scrutiny are the placement of the burden of proof and the requirement for the middle level that the appropriate governmental purpose be a substantial reason for the classification rather than merely an incidental consideration.^[5]

III

As noted above, the standard of scrutiny appropriate for review of a statute that classifies persons on the basis of age is whether the classification substantially furthers an appropriate governmental purpose. There is general agreement that improving highway safety is an appropriate governmental purpose, and an important one. The narrow issue is thus whether the age classification in these statutes substantially furthers that purpose.

On original hearing, the majority applied the above-stated standard to the evidence and framed the relevant inquiry as "whether eighteen to twenty year olds are the age group responsible for the greatest number of alcohol related accidents in Louisiana." 95-2189 (La. 3/8/96), p. 14 (emphasis added). In concluding that the evidence failed to establish that this age group was responsible for the greatest number of alcohol-related accidents in this state, the majority compared age groups in three-year segments. The majority rejected evidence that the eighteen-to-twenty-year-old age group was "over-represented"^[6] to a greater degree than other three-year age groups in alcohol-related accidents, because other three-year age groups with a larger number of licensed drivers were involved in a larger number of alcohol-related accidents. The majority also rejected the national statistics presented by the State, reasoning that while the national data provided sufficient justification for Congress under the Commerce Clause to enact the National Minimum Drinking Age Act, see South Dakota v. Dole, 483 U.S. 203, 107 S.Ct. 2793, 97 L.Ed.2d 171 (1987), that national data did not provide sufficient justification for an increase in the minimum, drinking age in Louisiana. Using this focus on absolute numbers and Louisiana statistics, the majority concluded that "eighteen to twenty year olds are not the group responsible for the greatest number of alcohol-related accidents in Louisiana." 95-2189 (La. 3/8/96), p. 14 (emphasis added).

On rehearing, we frame the relevant inquiry differently and employ a different focus.

The statute under review involves a change in the minimum drinking age. The precise question is whether the raising of the minimum drinking age from eighteen to twenty-one, which discriminates against adults in the lowest range of adulthood by postponing their legal access to alcoholic beverages, substantially furthers the governmental objective of improving highway safety. The inquiry should focus on the reasons for and effect of removing eighteen-to-twenty-year-old persons from the group of licensed drivers who are allowed to drink alcohol legally. The analysis should examine the relation of the two classes to each other by comparing licensed drivers in the eighteen-to-twenty-year-old group in alcohol-related accidents with licensed drivers in the age group of twenty-one and above. The inquiry should also include consideration of the reasons, presented by the evidence, for raising the minimum drinking age in order to affect highway safety.

IV

A detailed summary of the evidence introduced at trial is attached to this opinion as an appendix.

The governmental objective at issue is improving highway safety. The means chosen by the Legislature to improve highway safety in these particular statutes are aimed generally at the problem of drinking and driving, a problem that pervades all age groups, but specifically at the problem of youthful drinking and driving. The eighteen-to-twenty-year-old age group, who are barely experienced at driving legally, are totally inexperienced at drinking legally. In the words of the Commander of the Louisiana State Police, allowing this group to use alcohol has a detrimental effect on highway safety because the group "is not only inexperienced at driving but is also inexperienced at drinking." The Commander, based on his experience, expressed his expert opinion that the statutes under review are a "most critical and fundamental improvement" in attacking the problem of intoxicated drivers.

Increasing the minimum drinking age clearly is a rational and non-arbitrary approach to solving the particular problem of youthful drinking and driving, as well as the overall problem of intoxicated driving.^[7] Forty-nine other states and the federal system have adopted this approach as a non-arbitrary method of attaining improvement in the problem. Admittedly, those legislative decisions are not subject to the same equal protection scrutiny as the Louisiana statutes. Nevertheless, unanimous utilization of this approach to the problem is a significant indication that this approach "substantially furthers [the] appropriate state purpose" contemplated in Sibley. Id. at 1108.

Logic and experience were the principal bases for the governmental conclusion that an increase in the minimum drinking age will substantially further the objective of improving highway safety. Statistics were primarily used at trial to provide corroborative support for that conclusion.

The trial court, affirmed by the majority on original hearing, concluded that the statistics did not support that governmental decision. On reconsideration of the evidence and arguments of counsel, we conclude that the trial court erred manifestly on the mixed question of law and fact by using an incorrect method of analysis, and the majority on original hearing fell into the same error.

The majority on original hearing took a mistaken view of the State's argument, noting that the State attempted to justify the classification on the basis that the statutes would reduce the incidence of intoxicated driving and alcohol-related accidents in the eighteen-to-twenty age group. Of course, if that had been the State's argument, there was insufficient justification for a conclusion that the classification substantially furthered the improvement of highway safety in general. As the majority on original hearing noted prohibiting use of alcohol by any age group would reduce the incidence of intoxicated driving and alcohol-related accidents in that age group and would not justify the discriminatory classification.

The State's principal argument, which we deem valid, is that prohibiting use of alcohol by eighteen-to-twenty-year-old licensed drivers will improve highway safety for all motorists by removing from the general group of alcohol-using licensed drivers the specific group of alcohol-using licensed drivers who are, by percentage, the drivers most frequently involved in alcohol-related accidents. The overall statistical evidence supports this argument.

Of the statistical evidence, the most significant by far was the data obtained by the National Highway Transportation Safety Administration (NHTSA) from studies that were undertaken for the specific purpose of evaluating the effectiveness on highway safety of laws that raised the minimum drinking age to twenty-one. These statistics compared the specific group disadvantaged by the discrimination with the general group of licensed drivers and established, among other things, that the disadvantaged group was involved in twice as many accidents per capita as the general group and that Louisiana ranked above forty-six other states in the percentage of alcohol-related fatalities involving drivers under twenty-one. Other studies combined to conclude that minimum drinking age laws were responsible for significant reductions in traffic fatalities among motorists generally.

The executive director of the Louisiana Highway Safety Commission concluded from Louisiana and national data that alcohol-related fatalities involving young drivers increased significantly at age seventeen and peaked at age twenty. Particularly important was her testimony that persons in the disadvantaged group represented only five percent of the licensed drivers, but were involved in ten percent of the alcohol-related accidents involving injuries or fatalities. The latter statistics are particularly significant because if an increase in the drinking age to twenty-one eliminates drinking by the group of licensed drivers that are involved in twice the number of alcohol-related accidents proportionate to the percentage of drivers in that group, then clearly the increase substantially furthers the goal of improving highway safety. Prohibition of drinking by persons who are proportionately the most dangerous group of drinking drivers has to increase highway safety substantially, as opposed to incidentally.

Other reports outlined in the evidence established that the increase in the drinking age substantially reduces alcohol-related traffic accidents. Although any prohibition in the use of alcohol would have some beneficial effect on alcohol-related accidents, the specific evidence referred to in the previous paragraph establishes that the increase in the drinking age to twenty-one would have a significantly greater effect in reducing alcohol-related accidents.

The trial court's finding that the challenged classification does not substantially further the State's interest in improving highway safety because members of the disadvantaged group in Louisiana are neither arrested in greater numbers for intoxicated driving nor involved in greater numbers of alcohol-related accidents was erroneous as a matter of fact and law. First, there was no reason to reject from consideration the statistical data, studies, reports and expert opinions from outside Louisiana, and especially those studies that were specifically performed to evaluate the effectiveness of the increase in the minimum drinking age to twenty-one.^[8] Second, the court erred in considering a reduction in the gross number of accidents as the only means of improving highway safety. The State sought to improve highway safety by removing access to alcohol from the age group that is most likely, by percentage of licensed drivers, to be involved in alcohol-related accidents. The government may choose one of several appropriate methods for improving highway safety, as long as the method chosen substantially furthers that purpose. The proper approach for evaluating the statistical support for the method chosen by the State was to compare the disadvantaged group with the group of licensed drivers above twenty-one and to determine if the disadvantaged group was involved proportionately in significantly more alcohol-related accidents. Use by the trial court, as approved by this court's original hearing majority, of absolute numbers of accidents in each group (or in three-year age groups), without reference to the number of drivers in each group, was an improper focus Predetermining whether the classification is substantially related to the achievement, by the means chosen by the government, of an improvement in highway safety for motorists of all ages. These errors in method of analysis led to a manifestly erroneous conclusion.

V

The trial court further erred in accepting plaintiffs' argument that the decision in Pace v. State, Through La. State Employees Retirement Sys., 94-1027 (La. 1/17/95), 648 So.2d 1302 imposed three additional requirements to the burden of proof established by Sibley.

In their trial brief, plaintiffs set forth their interpretation of Pace as adding three additional requirements to the Sibley standard for determining whether legislation falling within the intermediate level of scrutiny under La. Const. art. I, § 3 substantially furthers an appropriate governmental interest. These three "requirements" were reiterated in brief to this court as follows:

Accordingly, the Age Legislation must survive the three Pace tests:

First, each interest must actually be implicated by the statutory scheme, Pace, 648 So.2d at 1309;

Second, there must be no non-discriminatory "alternatives which deal directly" with the asserted interest, Id.; and

Third, the age discrimination must not "undercut" a "countervailing state interest." Id. at 1310.

Plaintiffs' argument in the present case that Pace imposed additional requirements for intermediate review is erroneous. In Pace, this court merely reaffirmed that the proper standard for intermediate review of the statutory classifications expressly enumerated in the third sentence of Section 3 is the standard articulated in Sibley. We now clarify that Pace was simply a specific application of the Sibley standard to a concrete factual record and that Pace neither altered nor expanded the Sibley standard.

The Pace decision involved a challenge to a retirement statute that classified persons on the basis of birth or legitimacy. This court, using the Sibley standard for the enumerated categories in La. Const. art. I, § 3, held that the statute which required an illegitimate child of a male member of the retirement system to obtain a judgment of filiation in order to receive survivors' benefits, while not imposing the same requirement on legitimate children of members or on illegitimate children of female members, did not substantially further the State's asserted interests in the orderly disposition of property at death and the prevention of fraudulent or stale claims. The decision additionally noted, clearly for purposes of that factual situation only, that the former interest, although usually an important concern as to parental inheritance by illegitimate children, is "not implicated by the instant statutory scheme" (which provided for fixed monetary benefits) and does not involve the ownership of immovables or the need for finality required in succession proceedings. Id. at 1309.

The Pace decision further mentioned that there was a more effective non-discriminatory means of preventing fraudulent claims by the use of extremely reliable evidence, available through advances in modern scientific technology, to prove filiation, and that the strength of the asserted state interest in preventing fraudulent claims was undercut by a countervailing state interest in ensuring that genuine claims for child support are satisfied.

While these three factors perhaps were relevant for rejecting the classification in the factual context of Pace, these factors were neither part of the holding in Pace nor were these adopted as mandatory requirements for the intermediate level of scrutiny. Nevertheless, because the trial judge in the present case found a deficiency in the State's case based on failure to prove these factors, we discuss each separately.

A

The trial court ruled that the State failed to prove that the asserted governmental interest in improving highway safety was actually implicated by the classifications in the statutory scheme, as required by Pace.^[9]

The innocuous language in the Pace decision that the State's interest in the orderly disposition of property at death was not actually implicated by the statutory scheme that provided for fixed benefits hardly suggests that Pace added a significant burden to Sibley's intermediate scrutiny standard. Rather, the language in Pace indicates that there simply was not a substantial relationship in that case between the classification and the asserted governmental interest, which is essentially an application of the Sibley standard.

Since few classifications provide for a perfect fit^[10] between the classification and the asserted governmental purpose or for a completely irrational mismatch of classifications with purposes, "[t]he key factor in reviewing classifications is the degree of correlation between the means and the ends that is required by the judiciary arid the extent to which the judiciary will analyze the permissible purpose of the legislation." Ronald D. Rotunda and John E. Nowak, 3 Treatise on Constitutional Law § 18.2 (2d ed. 1992). This factor is dependent upon the standard of scrutiny applicable to the legislation in question.

In applying the federal intermediate standard of scrutiny in a case involving gender classification, the Supreme Court in Craig v. Boren, 429 U.S. 190, 197, 97 S.Ct. 451, 456-57, 50 L.Ed.2d 397 (1976), framed the inquiry as whether the classification serves "important governmental objectives and [is] substantially related to achievement of those objectives,"^[11] and focused on the classificatory fit. Considering a challenge to a statute that prohibited the sale of "nonintoxicating" 3.2% beer to males under the age of twenty-one and to females under the age of eighteen, the Court held that statistics broadly establishing 0.18% of females and 2% of males in that age group were arrested for alcohol-related driving offenses could not support the use of gender as a classifying device. The Court commented that the correlation between the gender classification and the objective of enhancing traffic safety was "an unduly tenuous `fit.'" Id. at 202, 97 S.Ct. at 459. Noting that the limited statistical data failed to consider the dangerousness of "non-intoxicating" 3.2% beer as opposed to alcohol generally, the suggestion that young men who drink and drive are transformed into arrest statistics while their female counterparts are chivalrously escorted home, or the fact that the statute only prohibited selling 3.2% beer to young males (and not their drinking the beverage), the Court held that the relationship between the gender classification and the goal of traffic safety was far too tenuous to support a conclusion that the classification was substantially related to achievement of the statutory objective.^[12]

In the present case, there is a common sense and experience-based relationship between the classification resulting from the increase in the minimum drinking age and the statutory objective of reducing youthful drinking and driving to improve highway safety. That relationship is supported by statistical data, when viewed in the proper focus. Although there is the difficulty noted in Craig of proving broad sociological propositions by statistics, the statistics in the present case were only necessary to provide corroborative support for the proposition, widely accepted by experts, that raising the minimum drinking age is substantially related to the improvement of overall highway safety by reducing alcohol-related accidents.^[13]

We conclude that there is a substantial relationship in the present case, that was absent in the Pace case, between the classification and the asserted governmental interest, which is all that Sibley requires.

B

The trial judge also imposed a burden on the State to prove that there were no nondiscriminatory alternative methods for reducing accidents caused by intoxicated driving. The judge listed alternatives, among others, such as public education, lowering the legal blood-alcohol level for drivers, and mandating lengthy driver's license suspensions.

This burden imposed on the State by the trial judge, at plaintiffs' suggestion that Pace required this burden, is seldom appropriate in intermediate scrutiny cases. In the federal system, cases requiring strict scrutiny because suspect classes or fundamental rights are involved sometimes mandate that the classification be narrowly tailored to achieve the specific governmental objective.^[14] Ronald D. Rotunda and John E. Nowak, 3 Treatise on Constitutional Law § 18.3 (2d ed. 1992). The intermediate scrutiny standard adopted by Sibley for review of age classification statutes falls far below the strict scrutiny standard in the federal system. Indeed, the Sibley standard—the classification must substantially further an appropriate governmental interest—is virtually the same as the intermediate scrutiny standard enunciated in Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976)—the classification must serve important governmental objectives and must be substantially related to achievement of those objectives. Neither the Sibley intermediate standard nor the Craig intermediate standard requires that the proponent of the statute prove there are no non-discriminatory methods of achieving the objective sought by the statute, as the trial judge required in the present case.

C

The trial judge indicated that the State had the burden to prove that the age classification does not undercut any countervailing governmental interest. However, the judge did not discuss this "requirement" any further in reasons for judgment. Suffice it to say, the Sibley intermediate standard contains no "requirement" that the proponent of the statute prove the classification does not undercut any countervailing governmental interest, and the trial judge erred in imposing this requirement.^[15]

VI

Plaintiffs' final argument is that even if implementation of the prohibitions in the challenged statutes would substantially further the governmental objective of improving highway safety, the statutes contain so many exceptions that the asserted objective "is not even implicated, much less substantially furthered."

La.Rev.Stat. 14:93.12 makes it a crime for a person under the age of twenty-one to purchase or to have public possession of alcoholic beverages. La.Rev.Stat. 14:93.10 defines "public possession" to include "consumption, on any street or highway or in any public place or any place open to the public, including a club which is de facto open to the public," but carves out of that definition the exceptions at issue. The exceptions exclude from "public possession" the possession or consumption of alcoholic beverages in the following five settings:

(i) For an established religious purpose.

(ii) At a function sponsored by a bona fide nonprofit organization under 26 U.S.C. 501(c) where an individual had received or purchased a ticket for admittance.

(iii) When a person under twenty-one years of age is accompanied by a parent or legal custodian twenty-one years of age or older.

(iv) For medical purposes when prescribed or administered by a licensed physician, dentist, nurse, hospital, or medical institution.

(v) In private residences.

La.Rev.Stat. 14:93.10(2)(a).^[16]

The State argues that almost all of the other forty-nine state statutes have similar exceptions permitting under-aged persons to drink in certain situations. Plaintiffs counter that while the minimum drinking age laws of other states contain similar exceptions, no minimum drinking age law of any state contains as many exceptions as the Louisiana statute. Plaintiffs contend that the inconsistencies presented by the exceptions undermine any furtherance of the governmental purpose. Noting that under-aged persons can nevertheless drink alcohol legally in their residences, at religious events, with parents and at bona fide nonprofit functions, plaintiffs submit that the statutes, when considered with the numerous exceptions, authorize public possession and consumption of alcohol by this age group in so many settings that the statutes cannot substantially further the governmental interest of improving highway safety.

The issue is determined by examining the State's purpose—improving highway safety—and determining the effect, if any, the exceptions have on that purpose. Therefore, the statutory exceptions must be individually and collectively examined to determine whether they, as the trial court concluded, undercut the State's appropriate objective of improving highway safety.

No one disputes the exceptions for medical and religious purposes. See Felix v. Milliken, 463 F.Supp. 1360 (E.D.Mich.1978) (noting accepted notion that religious and medical purposes are exceptions from such statutory and constitutional drinking age provisions). Rather, the dispute is over the other three exceptions.

The exceptions for consumption in a private residence or consumption with a parent or legal custodian share the common notion of family occasions, and we therefore address them together.

The trial judge concluded that since these exceptions allowed under-aged persons to consume alcoholic beverages "as much as they want" in certain settings uncontrolled by governmental supervision over alcoholic beverages, the exceptions effectively undermined the stated objectives of improving protection of the motoring public from youthful driving and drinking, as well as undermining the governmental policy favoring temperance.

Contrary to the trial judge's suggestion, this is not a temperance statute. Rather, the purpose of this statute is to curtail the ready availability of alcohol to inexperienced drivers and drinkers in this age group. The real danger at which the minimum drinking age law is aimed is the situation of under-aged persons' buying drinks at bar rooms and convenience stores before driving around in their vehicles. This is the setting in which the greatest danger of accidents is presented. On the other hand, when under-aged persons drink alcohol at home or with parents or guardians, theoretically their parents or guardians are not likely to allow them to drive on the highways after such drinking, and thus theoretically far less danger is presented.

This theoretical division between public settings and family settings is based on family controls versus societal controls on drinking and driving. Tracking that division, the exceptions restrict alcohol consumption by this age group to family-controlled settings, leaving the prohibition applicable primarily to non-family settings in public places. The statutory objective of reducing the ready availability of alcohol to under-aged persons likely to drive after drinking thus is not undercut by the two exceptions which permit consumption in family-controlled settings.

The exceptions also recognize the reality that the Legislature can control alcohol acquisition, possession and consumption in certain settings, but cannot exercise control, short of enacting a temperance statute, in certain other settings. When these exceptions are viewed in accordance with the division of control between the State and parents, the statutory exceptions do not undermine the statute's furtherance of the governmental purpose.

As to the final exception for consuming alcohol at ticketed functions sponsored by bona fide nonprofit organizations, we take judicial notice of the fact the Legislature recently amended La.Rev.Stat. 14:93.10(a) and totally repealed this exception. La.Acts 1996, No. 78. While this exception no longer affects the constitutionality of the present statute, we nevertheless analyze the effect of the exception as it existed at the time of the trial of this case.

Allowing persons between eighteen and twenty-one to drink alcohol at a ticketed event sponsored by a bona fide nonprofit organization is a very narrow exception to the prohibitory law against "public" possession and consumption of alcohol by this age group. The exception distinguishes between (1) a public place or a club which is de facto open to the public and (2) a single ticketed event held by a good faith nonprofit organization. While there is some "public" nature to this type of function, the effect of this infrequently anticipated occasion for drinking alcohol under this now repealed exception cannot truly be said to undermine the effect of the overall prohibition on the improvement of highway safety.

For these reasons, the judgment of the trial court is reversed, and plaintiffs' action is dismissed.

APPENDIX:^[1]

At the injunction hearing in the trial court, both sides submitted extensive affidavit testimony and documentary materials, and two live state expert witnesses testified.

Both sides introduced portions of the Louisiana Highway Safety Commission's raw data compilations. Particularly, plaintiffs introduced relevant portions of the 1986, 1991 and 1992 Louisiana Traffic Records Data Reports ("Louisiana Reports").^[2]

The 1986 Louisiana Report showed that the 18-20 year olds had 854 total alcohol-related fatal and injury accidents. By comparison, the 21-23 year olds had 924 such accidents; and the 24-26 year olds, 837. Adjusted to reflect the number of licensed drivers, however, the 1986 data report reflected that 18-20 year olds had fatal and injury-producing accidents at a per capita rate of 1 for every 191 drivers in that age group; whereas, all other older age groups had substantially higher per capita rates of alcohol-related fatal and injury-producing accidents, with the nearest group being the 21-23 year olds, having a per capita rate of 1 for every 217 drivers, followed by the 24-26 year olds, having a per capita rate of 1 for every 263 drivers.

The 1991 Louisiana Report employed a different methodology, using 4-year age grouping, and splitting the relevant 18-20 year old category into 2 categories: 15-19 and 20-24. Nonetheless, the 1991 report reflected that only 9.61% of all accidents for the 15-19 years olds were alcohol-involved; whereas, the corresponding statistic for the 20-24 year olds was 13.70%; the 25-29 year olds, 14.06%; the 30-34 year olds, 13.84%; the 35-39 year olds, 11.47%, and the 40-44, 11.89%.

The 1992 Louisiana Report employed yet another methodology, reflecting that the 15-20 year olds had 337 total alcohol-related fatal and injury accidents; whereas, the 21-24 year olds had 402 total such accidents; the 25-29 year olds, 479; the 30-34 year olds, 426; and the 35-39 year olds, 330.

The 1993 Louisiana Report, discussed in more detail below in connection with Bette Theis' affidavit, was introduced in fall by the State and, as plaintiffs point out, reflected that of the 509 total accidents involving 18-20 year olds, 352 (69.12%) involved alcohol impairment. In comparison, the report reflected that of the 517 total accidents involving 22-24 year olds, 377 (72.92%) involved alcohol impairment.

Plaintiffs also submitted 3 affidavits and documentary materials to establish that 18-20 year olds are arrested and convicted less often on DWI charges than other older 3-year age groups. Particularly, plaintiffs introduced affidavits from Charles Lombardino, Mary Jane Marcantel and Joseph W. Demourelle.

Lombardino, a Shreveport citizen, attested to DWI statistics he acquired from Caddo, Bossier City and Shreveport police departments. Lombardino attested that he requested and received statistical evidence from those departments listing the number of DWI citations issued, by age, for the year 1994. The Shreveport data was of no use in this case given that it was divided into age groups which split the 18-20 year olds into different categories. The Bossier City data showed that there were 22 DWI citations issued to drivers 20 and under and 119 DWI citations issued to drivers 21-30 years old. Assuming that the 21-30 years old group had an even distribution of citations, every 3year age group in that range had an average of 36 DWI citations during this same time period.

Marcantel, an experienced paralegal, was employed by plaintiffs to conduct statistical research, analyzing certain official DWI statistical records in Evangeline Parish for 1986, 1994 and 1995. Her research revealed that of the 92 DWI arrests in Evangeline Parish in 1986, 9 were of persons 18-20 years old, 13 were of persons 21-23 years old, and 12 were of persons 24-26 years old. Still further, she found that of the 179 DWI arrests during the period January 1, 1994 to July 28, 1995, 10 were of persons 18-20 years old as compared to 16 of persons 21-23 years old and 25 of persons 24-26 years. Summarizing, her research revealed that fewer 18-20 year olds were arrested for DWI in Evangeline Parish in either time period than either 21-23 or 24-26 years old.

Demourelle, an Evangeline Parish detective, also attested regarding DWI statistics in Evangeline Parish and regarding his experience with alcohol issues with young drivers. He stated that he researched and reviewed the criminal arrest and probation records of Evangeline Parish for the period January 1, 1995 to July 31, 1995, to determine the number of DWI convictions in that parish during that time frame and the age of the persons convicted. He stated that his research revealed of the 79 DWI convictions in Evangeline Parish during that time, only 1 was of a person under 21 years old.^[3]

Plaintiffs also introduced the affidavit of Robert Gramling, a professor of sociology at the University of Southwestern Louisiana, who has studied the relationship between drinking age legislation and alcohol consumption. His research, consisting of studies done in 1986 and 1992 comparing drinking habits of young adult college students in Louisiana and North Carolina, revealed a lack of empirical evidence to support the assumption that raising the drinking age to 21 resulted in a decrease in alcohol consumption by 18-20 year olds. Indeed, his research strongly suggested that increased quantities of alcohol may be consumed by 18-20 year olds where the drinking age is raised to 21.

Focusing on the Louisiana statutes at issue, Professor Gramling concluded that raising the legal drinking age in 1986 did not significantly change the alcohol consumption of 18-20 year olds in Louisiana. Commenting on the statutory exceptions, he stated that the exceptions "make it extremely likely that consumption of alcoholic beverages by persons between eighteen and twenty-one years of age will change locations from supervised public places to unsupervised places, further resulting in a potential increase in amounts consumed by this age group."

Discussing the latter shift in location of consumption, Dr. Gramling made a division between alcohol consumption in controlled and in uncontrolled setting. "Controlled" settings included those at which an under-aged youth may be required to provide proof of age as well as places at which societal controls are in place (public places and on the highway). "Uncontrolled" settings included those in which the sole controls are from family and guardians.^[4] Dr. Gramling concluded that raising the drinking age merely shifts the location of consumption from "controlled" settings—bars, cars and highway— to "uncontrolled" settings—private residences, fraternity houses, and friends houses—and found that it had no, or perhaps even an inverse effect, on consumption.

Dr. Gramling referred to this location shift as "front loading," which he defined as "consuming alcohol at home to the desired intoxication level before going out to situations in which consumption will be illegal." Hence, fewer drinks are needed in controlled situations throughout the evening in order to maintain a desired level of intoxication. He noted the obvious implications of this on highway safety given that the youth will be intoxicated while travelling to and from their designation.

Turning to the State's evidence on the highway safety issue, the State introduced affidavits from James Hedlund and Bette Theis.

James Hedlund was the Acting Associate Administrator for Traffic Safety Programs, and Director of the Office of Alcohol and State Programs, National Highway Traffic Safety Administration (NHTSA), U.S. Department of Transportation. Hedlund attested that "NHTSA has monitored the involvement of teenage drivers in fatal traffic crashes and the effectiveness of state laws establishing 21 as the legal age for purchase and public possession of alcoholic beverages. The Office of Alcohol and State Programs is responsible for NHTSA's impaired driving (resulting from alcohol or other drugs) and traffic records program." Hedlund further attested that 23 U.S.C. § 158 documented that state age 21 laws have saved lives, encouraged all states to adopt such age 21 laws, and solved the notorious "blood border" problem—"teenagers in an age 21 State no longer could drive to an adjoining State, drink legally, and then crash on their way home."

Relying on statistical data obtained by NHTSA through its Fatal Accident Reporting System ("FARS") and from reports studying age 21 drinking law effectiveness, Hedlund provided the following data.

First, he noted that over-involvement of 18-, 19- and 20-year-olds in alcohol-related crashes has been established by the following four findings:

(1) In 1994, 44% of 18-, 19-, and 20-year-old traffic fatalities were alcohol-related; this compares to 40.8% for all traffic fatalities.

(2) Alcohol-related traffic fatality rates (fatalities per capita) are over twice as great for 18-, 19-, and 20-year olds as for the population over 21.

(3) More 18-, 19-, and 20-year olds died in low (.01 to .09) blood alcohol level traffic crashes than 21- 22- and 23-year olds, or any other 3-year group, in 1994.

(4) In 1994, Louisiana had the 4th highest percentage (57.2%) of alcohol-related traffic fatalities of 15-20 year olds of all States (behind Alaska, Delaware, and New Mexico).

Second, Hedlund noted that the effectiveness of age 21 drinking laws has been established by the following three reports:

(1) A Government Accounting Office (GAO) report to Congress, Drinking-Age Laws—An Evaluation Synthesis of Their Impact on Highway Safety, March 1987, examined 14 high quality traffic accident studies. The report concluded that "[r]aising the drinking age has a direct effect on reducing alcohol-related traffic accidents among youths affected by the laws, on average, across the states." The report further stated that "[a]lmost all studies have found statistically significant reductions in traffic-accident outcomes, even though the studies often varied in scope, design, analysis methods and outcome measured."

(2) A January 1989 NHTSA study, The Impact of Minimum Drinking Age Laws on Fatal Crash Involvements: An Update of NHTSA Analyses, estimated that minimum drinking age laws were responsible for a 12% reduction in fatal crash involvements of affected drivers.

(3) In 1994, NHTSA estimated that state 21-year-old minimum drinking age laws have reduced traffic fatalities involving drivers 18-20 years old by 13% and have saved an estimated 14,816 lives since 1975.

The State also introduced the affidavit of Bette Theis, the Executive Director of the Louisiana Highway Safety Commission. She attested, based on the Louisiana Highway Safety Commission, 1993 Traffic Records Data Report, discussed above, and the NHTSA FARS 1994 Data Summary Report, that Louisiana data for 1993 reflected alcohol-involved fatalities for young drivers increased significantly at age 17, peaked at age 20, and began to decline between ages 22-24. She further attested that "[i]t is proven that in Louisiana and throughout the nation, young drivers are involved in alcohol-related crashes at a disproportionately high rate."

Consistent with the "over-representation"^[5] evidence established by the 1986 Louisiana Report and federal data discussed above, Ms. Theis attested that the 1993 Louisiana Report reflect that 18-20 year olds represented only 5% of the licensed drivers in Louisiana in 1993, yet were involved in 10% of the alcohol-involved fatal and injury accidents. In the 46 alcohol-involved fatal crashes during 1993, she attested that 72% of the 18-20 year old age group had been drinking. In the 17-20 year old age group, 26 drivers were killed in alcohol-involved crashes, representing 47% of all traffic fatalities in this age group. Moreover, she commented that "this number (Driver Fatalities) represents the proverbial `tip on the iceberg' and does not include other fatalities in crashes (e.g. passenger and pedestrian) or driver, passenger, and other injuries among young people in this age group."

The State also introduced an affidavit from a doctor and a nurse regarding the effect of alcohol on accidents in general involving 18-20 year olds.

Dr. Melvin Kohn attested that "injuries— particularly due to motor vehicle crashes, homicide and suicide—are the leading cause of death among persons aged 18-20 years. According to the best data available, alcohol use contributes to the occurrence of 30-50% of injuries." Continuing, he attested that alcohol use can result in "impaired judgment, delayed reaction time, and impulsive behavior" and that "[a]lcohol can exaggerate [the increased] willingness [of adolescents] to take risks, and transform what normally might be a harmless prank or activity into a serious injury either to the risk-taker or to an innocent bystander."

Similarly, David Lawrence, a registered nurse, attested that "[a]lcohol-related injuries are an important cause of death and disability in Louisiana but especially among the young." He further attested that "[t]he use of alcohol has been shown to increase the risk of virtually all types of injuries" and noted that studies of adolescents and young adults in other states reflected that "there is a great amount of evidence that a reduction in alcohol consumption also reduces the number and severity of injuries in this age group."

The State also introduced a copy of the report rendered to Congress by the Government Accounting Office (GAO), entitled Drinking-Age Laws—An Evaluation Synthesis of Their Impact on Highway Safety (March 1987), which is mentioned in Mr. Hedlund's affidavit, discussed above, and in Dr. Scribner's testimony, discussed below. The report cautioned that "[i]t is generally acknowledged that drinking-age laws do not affect traffic accidents directly but are mediated by a variety of intervening variables." Id. at 20. Nonetheless, the report noted that the studies that have analyzed the effect of changes in the drinking age on total crash fatalities for age groups affected by the law have found "statistically significant" effects; particularly, one study found that raising the drinking age resulted in a 7% average reduction in fatalities in the states with the higher drinking age. Id. at 39. The report further noted that "[s]tates with an older minimum [drinking] age seem to have better control over drinking and driving among youths." Id. at 63.^[6]

As noted, the State also called two live expert witnesses to testify.

The State's first witness was Captain Ronald Jones, Commander of Troop A of the Louisiana State Police, who was qualified as an expert in traffic enforcement and policy development. Captain Jones testified that, based on his personal experience, the challenged minimum drinking age statutes could be "the most critical and fundamental improvements in traffic safety when it comes to alcohol in this state." He further testified that in his opinion, access to and use of alcohol by 18-20 year olds has a detrimental effect on highway safety because that age group "is not only inexperienced at driving but is also inexperienced at drinking." Continuing, he testified that "when you combine those two levels of inexperience you are really facing serious problems. I think all of us were young at one time and we all drove at one time and we know that we took risks when we were younger on the highway that we clearly would not take today. When you combine that with alcohol consumption it can be a deadly combination."

On cross-examination, Captain Jones agreed that 21-23 year olds likewise have a high rate of injuries related to drunk driving, stating that "I think that persons in their 20's also have been identified with having problems with drinking and driving in accidents." He further testified on cross-examination that various alternative measures exist that the State could employ to reduce highway accidents, including school educational programs, public advertising about the risk of drinking and driving, using designated drivers, improving automobile and highway design, and tougher DWI laws such as mandatory jail time.

The State's second witness was Dr. Richard Scribner, who was qualified as an expert in public health and preventive medicine. Dr. Scribner testified that, in his opinion, alcohol is the leading cause of death among 18-20 year olds. He testified that this means that although the 3 leading causes of death among 18-20 year olds are accidents, homicide, and suicide, alcohol plays a part in a significant number of deaths attributed to these causes.

Relying on the 1987 Governmental Accounting Office (GAO) Report evaluating other statistical studies of alcohol involved accidents, Dr. Scribner further testified that raising the drinking age decreases alcohol related fatal and injury accidents and that the younger the driver the greater the over-representation in alcohol involved auto accidents. Based on that GAO Report, he estimated that enforcement of the challenged minimum drinking age law will result in a 5-28% decrease in alcohol involved fatal injury accidents in Louisiana.

Dr. Scribner testified on cross-examination that his own research has been in the area of alcohol availability and has revealed that peak alcohol consumption occurs between ages 20-24. He further testified on cross-examination that the rate of violent deaths associated with alcohol was consistent across all age groups. In response to questioning regarding the impact of alcohol involvement on violent deaths, he expressed his view that given the "social cues, the social milieu ... the younger you are the less experienced you are in many of the social situations in which alcohol is involved which leads to the types of situations which may result in either a violent altercation or a traffic crash."

BLEICH, Justice, concurring in the majority opinion.

A basic cornerstone of creditable government service is public safety. At times the measures chosen are evident and readily acceptable; at other times the protective provisions are somewhat less popular. It is within the province of the people though, through their elected legislative officials, to make those choices, subject to their own constitution.

Occasionally, the laws chosen to effect the goals of public safety are passed without sufficient public scrutiny. Such is not the case here. The subject legislation has been the source of great public debate.^[1] There are those who seek to invalidate these statutes because of the effect on profits stemming from the sale of alcoholic beverages.^[2] Some groups, individuals and associations support the subject legislation for safety, moral or social reasons. There are others who are genuinely concerned about any distinction based on age in light of our constitution.^[3] These persons, groups and associations provided significant input, information and debate to the legislature prior to the enactment of this legislation. Whether any members of this court disagree with the subject legislative action from a personal standpoint is of no moment. This court's only function is to interpret the statutes—the expression of the public will. It is not the function of this court to legislate.^[4]

The people of Louisiana have seen fit to prohibit, based in part on statistical information,^[5] the sale and possession of alcoholic beverages to some of their fellow citizens, those above the age of eighteen and less than twenty-one. The people of Louisiana have arguably created these prohibitions for safety and financial^[6] reasons. They have concluded that there is too great a risk imposed, not only on those in the subject age category but also on other travelers on our highways, to allow the possession and purchase of alcoholic beverages by persons in this age category.^[7] The people of Louisiana have said that they want to remove the "blood borders" that have heretofore existed.^[8] The citizenry have taken the position that despite the fact that they discriminate against their fellow citizens of the subject age category concerning the purchase and consumption of alcoholic beverages, the distinction is justified action to prevent further carnage. The people, through the legislature, have concluded that it is better to prevent a higher incidence of accidents and to keep more of their fellow citizens alive and uninjured.^[9]

The test adopted by the majority clearly meets the test of constitutionality. The subject statutes will certainly substantially promote the legitimate state interest of safety. To save any lives must be considered of "substantial" import.

The test adopted in the original opinion of the court and by the dissenters today defies the very intent of our constitution. If one were to take the rationale of the dissenters to its conclusion, one would find it impossible to ever draw a distinction based on age. As the dissenters today protest that our constitution is being misinterpreted, they must be reminded that the redactors of the constitution and the people said that sometimes a reasonable distinction must be drawn.

Obviously the subject legislation will not eliminate the carnage, death and destruction on our highways. As total prohibition earlier in this century did not eliminate all social ills, these statutes^[10] will not eliminate all highway safety problems. The subject statutes will, however, based on the record and evidence, save a significant number of lives.

The people of Louisiana have expressed their will through their legislators and have made it clear that they desire to impose these protective measures. This court would have acted inappropriately if it declared the judgment of the people unconstitutional and substituted its judgment for that of the people, especially when the people seek their own safety and protection.

I conclude that the legislature did not act arbitrarily, capriciously or unreasonably in enacting the subject statutes. Therefore, with this concurring opinion, I also respectfully adopt the well-reasoned opinion of Justice LEMMON.

KIMBALL, Justice, dissenting.

This case is not about whether 18-20 year olds should be allowed to purchase and consume alcoholic beverages in Louisiana. This case, instead, is about the proper interpretation of the Constitution of the State of Louisiana, as it exists in its present form. If the people of the State of Louisiana desire to prohibit 18-20 year olds from purchasing or consuming alcoholic beverages, the proper method for accomplishing such a result is to amend the Constitution of State of Louisiana. This, of course, is already in the process of being done, as the legislature has placed just such an amendment on the ballot for congressional general elections this year.^[1] The function of this Court is to enforce the Constitution of the State of Louisiana, not find ways to justify unconstitutional legislative enactments that we believe, on the basis of "experience and logic" (as opposed to the record evidence), are nevertheless sound social policy. Because the majority herein has conveniently ignored the Constitution of the State of Louisiana, ignored the record evidence, sidestepped the manifest error rule, and disavowed this Court's prior jurisprudence in this area to reach a result not otherwise properly obtainable, I respectfully dissent.

Article I, Section 3 of the Louisiana Constitution, in pertinent part, states: "No law shall arbitrarily, capriciously, or unreasonably discriminate against a person because of birth, age, sex, culture, physical condition, or political ideas or affiliations." Though the majority scarcely acknowledges it, ante at 121, the fact is that Article I, Section 3 of the Louisiana Constitution clearly and explicitly requires greater protection against discrimination on the basis of age than either the United States Constitution or any other State Constitution.^[2] Relying on "common sense and logic," ante at 121, 125, instead of the record evidence which clearly shows that, in Louisiana, eighteen to twenty year olds are not the group responsible for the greatest number of alcohol related accidents, the majority holds that it is not "arbitrary, capricious, or unreasonable" for the State to skip over other age groups responsible for the greatest number of alcohol related accidents and penalize a group of adults that has been shown by the record evidence to be responsible for less accidents than the other, non-penalized age groups. In doing so, the majority has conveniently ignored the specific directive of Art. I, Sec. 3 of the Louisiana Constitution that the government shall not arbitrarily or capriciously discriminate against persons on the basis of age, and conveniently ignored the record evidence. Classifying persons on the basis of age, where the record evidence shows the age group singled out is not the group most responsible for the evil which the government seeks to address, is inherently arbitrary and capricious.^[3]

Under the majority's reasoning, the legislature could constitutionally decide, on the basis of "logic and experience," that since men are overrepresented in alcohol related accidents, men may not purchase or consume alcohol. This, of course, is what the Oklahoma legislature decided in a statute which was struck down as unconstitutional under federal intermediate scrutiny by the United States Supreme Court in Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976), a standard the majority herein acknowledges "is virtually the same as the standard for review of age classification applicable in this case." Ante at 124, n. 11. Recognizing both the similar nature of the issues involved and the applicable levels of review, the majority herein nevertheless holds that there is "a common sense and experience-based relationship" between the challenged classification and the State's stated objective, despite the fact that the record evidence shows otherwise. In our legal system, contested issues should be decided on the basis of evidence properly adduced at trial. I therefore cannot in good conscience ignore both the Constitution of this State and the record evidence to decide this case on the basis of my, or any other judge or group of judges', "common sense and logic."

Of course, by ignoring the prohibitions of Art. I, Section 3 of the Louisiana Constitution, the majority has also conveniently sidestepped the manifest error rule by first deciding that the trial judge erred as a matter of law in his analysis. When the issue is properly analyzed under Article I, Sec. 3, as was done by the trial judge and as was done by this Court in its original opinion, there is no basis whatsoever for reversing the trial court's factual findings and, therefore, no basis for reversing his determination that the challenged statutes' classification of 18-20 year olds arbitrarily, capriciously and unreasonably discriminates against them on the basis of age.

Finally, it has been said that "hard cases make bad law." Such is clearly the case in the instant matter. In Pace v. State, Through La. State Employees Retirement System, 94-1027 (La. 1/17/95), 648 So.2d 1302, this Court unanimously decided that a statute which classified persons on the basis of birth or legitimacy was unconstitutional. In deciding that case, this Court employed several factors, including whether each governmental interest was actually implicated by the statutory scheme, whether any nondiscriminatory alternatives which dealt directly with the asserted governmental interest existed, and whether the discriminatory classification undercut any countervailing State interests, to determine whether the discriminatory classification at issue could withstand constitutional challenge. Now, only 18 months later, the majority effectively completely disavows this Court's decision in Pace, ostensibly because the factors described were "simply a specific application of the Sibley standard to a concrete factual record." Ante at 123. In my view, Pace and the analysis employed therein would not be so lightly discarded today but for the purpose of reaching a desired result in the matter at issue herein.

I respectfully dissent.

JOHNSON, J., concurs in result.

BLEICH, J., concurs and files additional reasons and subscribes to the majority opinion.

KIMBALL and MARCUS, JJ., dissent and assign reasons.

MARCUS, Justice (dissenting).

In my opinion, there is not a sufficient correlation between restricting the sale of alcoholic beverages to persons under twenty-one and the enhancement of traffic safety to support the conclusion that the classification is substantially related to the statutory objective. The majority permits this based upon tenuous statistical data.^[1] To further support its position, the majority relies upon "logic and experience"; however, this depends upon whose logic and experience are being employed. Accordingly, I respectfully dissent.

CALOGERO, Chief Justice, assigns additional concurring reasons.

I fully agree with the outcome and the reasoning of the Court's opinion in this case. In my view, the majority on original hearing erred in holding that the State is constitutionally prohibited from enacting a statute that raises the legal age for the purchase and public consumption of alcohol from eighteen to twenty-one.

On original hearing, the majority was led astray, in my view, by two factors. First, we misinterpreted and misapplied our decision in Pace v. State, Through Louisiana State Employees Retirement System, 94-1027 (La. 1/17/95), 648 So.2d 1302. In Pace, as the current opinion notes, this Court used three factors (whether a law implicates a state interest, whether there are non-discriminatory alternatives, and whether the challenged law undercuts countervailing state interests) to determine whether a provision of the State retirement system was valid.

These three factors were taken from Pace and used, for the first time, to stand for the proposition that they are the factors that "Louisiana courts examine" in order to determine whether a law "substantially furthers an important governmental objective." Manuel v. State, 95-2189, p. 5-6 (La. 3/8/96) (quoting Pace, 648 So.2d at 1305) (on original hearing). This declaration by the original majority was an entirely new statement of the law. Prior to our original opinion, Pace had not been interpreted as establishing three additional factors by which challenges to statutes under the Individual Dignity Clause would be evaluated. By applying the Pace factors to this case, the majority on original hearing, added a significant and unsupported burden to Sibley's intermediate scrutiny standard. However, as the current opinion notes, all that is needed in order to support the statute is evidence that it substantially furthers an appropriate governmental interest, not compliance with the factors found applicable and essential in Pace.

The original opinion's addition of the Pace factors to the Sibley standard was a critical error. However, an equally critical error, and the principal one which led us astray, in my opinion, was this Court's interpretation of the statistical evidence in the case. Key to the original opinion was the conclusion that there was no difference in the level of risk— incident to alcohol related accidents—posed by drivers in the eighteen to twenty year old group versus the twenty-one to twenty-three year old group and the twenty-four to twenty-six year old group of drivers.

The original opinion framed the issue as "whether eighteen to twenty year olds are the age group responsible for the greatest number of alcohol related accidents in Louisiana," Ante, at p. 120, and concluded that because they were not responsible for the greatest number of accidents, the statute's singling them out, among all post-eighteen year old age groups could not stand. The Court simply counted the number of alcohol related fatal and injury accidents within the eighteen to twenty year old and twenty-one to twenty-three year old groups, found more accidents in the latter group, then concluded that the eighteen to twenty year olds were less dangerous, not more dangerous, than the others. The fallacy in this argument is the fact that there were fewer eighteen to twenty year old licensed drivers than twenty-one to twenty-three year old licensed drivers. In fact, in terms of the number of licensed drivers, there was a greater percentage of eighteen to twenty year old alcohol related accidents than twenty-one to twenty-three year old alcohol related accidents. The eighteen to twenty year old group was thus over-represented in the percentage of alcohol related accidents. This group represented only five percent of licensed drivers, but was involved in ten percent of all alcohol related accidents involving injuries or fatalities, whereas the twenty-one to twenty-three year olds had a lesser percentage of alcohol related accidents notwithstanding that numerically they were involved in more accidents. The statistics, therefore, support the State's position that eighteen to twenty year olds are more dangerous where drinking and driving are concerned.

The prohibition on drinking for a group of drivers (by age) who are proportionately the most dangerous increases highway safety not insubstantially and without illegally discriminating against eighteen to twenty year olds. The legitimate state interest in promoting highway safety is substantially furthered by the statute.

NOTES

[*] Justice E. Joseph Bleich participated in the decision on rehearing, having been elected to fill the vacancy created by the resignation of Justice James L. Dennis. Watson, J., not on panel. Rule IV, Part 2, § 3.

[1] The parties subsequently agreed to convert the preliminary injunction to a permanent injunction, and the trial court's decision was stayed pending this court's decision.

[2] In Sibley, this court held that a legislative classification between medical malpractice victims with slight or medium class injuries, who were entitled to full recovery of damages, and seriously injured medical malpractice victims, who were only entitled to a limited portion of their damages, violated the prohibition of La. Const. art. I, § 3 against arbitrary discrimination based on physical condition.

[3] The first sentence sets forth a general rule against discrimination and empowers the courts to expand the equal protection guarantee to other types of classifications besides those expressly enumerated.

[4] The standards of the federal system, because of the Supremacy Clause, establish the baseline minimum standard of scrutiny. Pace v. State, Through La. State Employees Retirement Sys., 94-1027 (La. 1/17/95), 648 So.2d 1302. Nevertheless, a state may adopt a greater degree of protection of individual rights. In the federal system, the standard for reviewing a statute that discriminates on the basis of age utilizes the minimum standard of scrutiny. See Vance v. Bradley, 440 U.S. 93, 99 S.Ct. 939, 59 L.Ed.2d 171 (1979); Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 96 S.Ct. 2562, 49 L.Ed.2d 520 (1976). When such a statute is reviewed under the Louisiana Constitution, however, a higher level of scrutiny is applicable and the burden of proof is changed because the framers of the Constitution expressly enumerated in the third sentence of Section 3 a number of classifications that affect significantly important interests that do not reach the level of constitutionally suspect categories.

[5] The phrase "substantially furthers" in the standard for reviewing discriminatory statutes based on age imposes the requirement that the government purpose must be a substantial, as opposed to merely an incidental, reason for the classification. The intermediate standard of scrutiny thus accords less deference to the legislative branch than the rational relationship standard.

[6] "Over-represented" means that a certain age group has more licensed drivers involved in alcohol related accidents on a percentage basis than the group's respective percentage of total licensed drivers. Illustrating this concept, the majority noted that the State's evidence showed that eighteen to twenty year olds in 1993 accounted for five percent of licensed drivers in this state, yet were involved in ten percent of the alcohol-related fatal and injury accidents. Hence, that age group was "over-represented." 95-2189 (La. 3/8/96), p. 14 n. 7.

[7] As noted, the challenged statutes focus on the problem of youthful drinking and driving. If youthful drinking and driving is a substantial problem in highway safety and if increasing the minimum drinking age is an appropriate means of attacking that problem, the minimum drinking age can only be increased at the lowest level—the eighteen, nineteen and twenty-year-old group. Thus, the statutes do not single out the members of this three-year age group for unequal treatment as compared to other three-year age groups. This is the only age group that can be affected by a three-year increase in the minimum drinking age.

[8] The trial court's apparent reliance on statistics of DWI arrests in Evangeline Parish was particularly inappropriate. Not only did the statistics not show the percentage of drivers in the disadvantaged group in the parish, but also there was no consideration that young drinking drivers are frequently not arrested for the first incident.

[9] The majority on original hearing went far beyond the Sibley intermediate scrutiny standard and required that the age classification chosen by the Legislature be the one "which most directly implicates or furthers the asserted governmental interest." 95-2189, p. 6. (emphasis added). Sibley has no requirement that the Legislature's choice of means for furthering an appropriate objective be the best choice of several means. This requirement has only been applied in federal strict scrutiny situations involving suspect classes or fundamental rights.

[10] The relationship between a classification and the governmental purpose is sometime referred to as classificatory fit. Classificatory fit is generally analyzed for legislative rationality as a function of underinclusiveness and overinclusiveness of the classification, an approach posited by a seminal article, Joseph Tussman and Jacobus tenBroek, The Equal Protection of the Laws, 37 Cal.L.Rev. 341 (1949).

An underinclusive classification is one that "contains all similarly situated people but excludes some people who are similar to them in terms of the purpose of the law." Ronald D. Rotunda and John E. Nowak, 3 Treatise on Constitutional Law § 18.2 (2d ed. 1992). An overinclusive classification is one that "includes all persons who are similarly situated in terms of the law plus an additional group of persons." Id. Often, as in the present case, classifications are both under and overinclusive. Id.

The age classification at issue here is underinclusive because it does not address a major portion of the perceived problem, adults between twenty-one and thirty. It is overinclusive because it prohibits young adults from drinking even when they would not be driving. South Dakota v. Dole, 483 U.S. 203, 214-15, 107 S.Ct. 2793, 2799-2801, 97 L.Ed.2d 171 (1987) (O'Connor, J., dissenting). The overinclusiveness of the Louisiana minimum drinking age laws at issue, however, is somewhat overcome by the exceptions which permit young adults to drink in specified settings, such as private residences and with parents.

[11] The intermediate scrutiny standard used in Craig to review gender discrimination—the classification must serve important governmental objectives and must be substantially related to achievement of those objectives—is virtually the same as the standard for review of age classification applicable in this case.

[12] While the percentages for males was ten times that for females, the percentages were so small as to be of questionable accuracy, especially in view of the value problems pointed out by the Court for such limited statistics.

[13] See Government Accounting Office, Report to Congress entitled Drinking-Age Laws—An Evaluation Synthesis of Their Impact on Highway Safety (March 1987).

[14] Less restrictive alternatives may be required in cases involving overinclusive classifications. Laurence H. Tribe, American Constitutional Law § 16-1 n. 23 (2d ed. 1988).

[15] The majority on original hearing stated that the age classification undercuts the countervailing governmental interest of according "adult" or "major" status to members of this age group. The majority pointed out many responsibilities and obligations assigned to persons of the age of majority. While these observations are persuasive arguments against legislative adoption of the classification, the function of this court is not to pass on the desirability or wisdom of the Legislature, but to determine the constitutionality of the legislative enactment.

[16] We take judicial notice of the fact the Legislature amended these exceptions after our decision on original hearing, totally repealing the exception for bona fide nonprofit organizations and slightly amending the wording of the remaining four exceptions. See La.Act 1996, No. 78. More precisely, the statutory exceptions, as amended, now read as follows:

(i) For an established religious purpose.

(ii) When a person under twenty-one years of age is accompanied by a parent, spouse, or legal guardian twenty-one years of age or older.

(iii) For medical purposes when purchased as an over the counter medication, or when prescribed or administered by a licensed physician, pharmacist, dentist, nurse, hospital, or medical institution.

(iv) In private residences.

La.Rev.Stat. 14:93.10(2)(a).

[1] For clarity and consistency sake, we depart from proper Bluebook form in this appendix and utilize Roman numbering throughout.

[2] While the State, on rehearing, requested that this court expand the record to include, among other things, additional portions of these data compilations, we decline to do so and decide the case on the record as presented in the trial court.

[3] Plaintiffs also introduced federal statistics from the Federal Bureau of Investigation to establish that the Louisiana arrest statistics were consistent with the federal statistics and that the statistics for alcohol abuse, as measured by nationwide arrests for drunkenness, followed a similar pattern.

[4] "Controlled" was defined as "situations in which proof of legal age was required for the purchase of alcohol (bars, restaurants, spectator events) or where there was a higher likelihood of legal sanctions being imposed because of the greater surveillance of formal social control agents (public areas, motor vehicles)." "Uncontrolled" in contrast, was defined as "situations in which these criteria were absent or at least significantly reduced because of their tendency to be private rather than public locations (residences, dorms). Although informal social control may be present (e.g., parents), the threat of state-backed surveillance and enforcement is minimal (residences, dorms, fraternity and sorority houses, relatives' homes, and private houses)." Linda A. Mooney, Robert Gramling and Craig Forsyth, Legal Drinking Age and Alcohol Consumption, 13 Deviant Behavior: An Interdisciplinary Journal 59, 65 (1992).

[5] See footnote 6 of opinion on rehearing for the definition of "overrepresentation."

[6] The State also introduced affidavits from John Womack and Sharon F. Lyles on the federal funding issue, an issue that we do not reach in our decision on rehearing.

[1] This is confirmed by action of the legislature in addressing this issue again in the First Extraordinary Session of 1996, Act No. 78.

[2] The prohibition of the purchase of alcohol by a person in this age category has been prohibited, generally, since 1986. See Act No. 33, 1st Ex. Sess., 1986. The former set of statutes did not prohibit the sale of alcohol to these persons, and created a contradiction in legislative intent. The legislature "closed the loophole" in the instant Act 639 to prohibit vendors from selling alcohol to the subject age category.

[3] Noteworthy is the fact that Louisiana has delineated other distinctions based on age in the subject category. In the area of gambling, see LSA-R.S. 4:544(B)(1): and 4:660.

LSA-R.S. 4:544(B)(1):

"A person under the age of twenty-one shall not play, or be allowed to play, any licensed gaming device or slot machine, ..."

LSA-R.S. 4:660:

"A. A person under the age of twenty-one shall not:

(1) Play, or be allowed to play, any licensed game or slot machine.

(2) Loiter, or be permitted to loiter, in or about any room, premises, or designated gaming area wherein any licensed game is operated or conducted.

(3) Be employed as a gaming employee.

B. Any casino operator, licensee, or other person who intentionally violates or permits the violation of any of the provisions of this Section and any person under twenty-one years of age who violates any of the provisions of this Section may be punished by imprisonment of up to six months or a fine of up to one thousand dollars, or both.

C. In any prosecution or other proceeding for the violation of any of the provisions of this Section, it shall be a defense that the casino operator, employee, dealer, or other person had a reasonable factual basis to believe and in good faith believed the person was twenty-one years old or over.

D. The casino gaming operator shall withhold all winnings from patrons who are determined to be under the age of twenty-one and shall remit such winnings to the corporation."

[4] The doctrine of separation of powers is sacred. Whether the legislative act is in the area of criminal justice, tort law, taxation or as here, public safety, this court must refrain from invading another branch of government by attempting to substitute its own judgment.

[5] See the Appendix to the original opinion which contains an exhaustive summary of the evidence that appears of record. It is more than abundantly clear that the percentage of accidents among the subject age group due to alcohol was much higher. The proponent of the legislation has clearly met its burden.

[6] The question of potential loss of federal revenues is not of primary concern to this writer. However, this writer's personal judgment cannot be substituted for that of the legislature.

The plaintiff urged in oral argument that there might not be, or would not be, a loss of money and argues that this should not be a consideration. The record indicates to the contrary. The amount of money that might be lost by the state is estimated to be in the range of eighteen million ($18,000,000.00) dollars per year. The state legally has a justification in this regard.

[7] There was mention in oral argument that there might have been other considerations than highway safety and funding; e.g., that the early usage of alcohol would create a "gateway to crime," implying that early use of alcohol would lend itself to a higher incidence of crime in the subject age category. The record makes direct reference to this proposition. See the testimony of Dr. Richard Scribner (R-154) to the effect that alcohol was heavily involved in deaths of those age 18-20, including traffic accidents, homicides and suicides. It would defy logic and common sense to conclude that alcohol consumption in this age category does not significantly contribute to criminal activity other than traffic violations.

[8] See the record, 152, and the testimony of State Police Commander, Troop A, Ronald B. Jones. "Blood borders" is the term used to describe the situation wherein younger persons in the subject age category from other states drive to Louisiana in order to purchase alcoholic beverages.

[9] Although not argued, there is another important outcome that would result from this legislation designed to promote highway safety, i.e. a reduction in the cost of insurance. The Louisiana automobile policyholder should feel less burden via insurance premium when there is a reduction in accidents in this state.

[10] Indeed, the statutes are attempts to protect the public, both those in and not within the subject age category, from the dangers of excessive and early use of alcohol. It is ironic that the greater problems of dependence on alcohol stem in part from an instability and lethargy that our governments have helped create. The "if it feels good, do it" mentality that pervades our society has been exacerbated by government attempting to please the whim and fancy of its varied constituencies. The reluctance to abide by those great fundamental principles that made this country great, including reliance on family first and government last, have seemed to escape our focus in this country. Indeed, it may well be argued that it is the function of the family unit to address the problems of alcohol and substance abuse. Parenthetically, one of the exceptions in the subject legislation is within the confines of the family home.

However, the subject legislation, even if a substitute for what families should be doing in many instances, is not constitutionally repugnant. To the contrary, as long as there is not an unreasonable or arbitrary basis for this legislation, which this writer finds that there is not, it creates two small steps—one a stride forward to safety, and the other back to the principles that sober, healthy minds are preferred, not condemned. Louisiana is a uniquely beautiful state. Our people have regularly shown their desire to preserve the sanctity of life. Our family values demand such. And although citizens cannot and should not wholly depend on government to preserve their values in each and every instance, the subject legislation is a sensible expression that life will and should be preserved, and that it is appropriate for government to assist in that function.

[1] See 1996 La.Sess.Law Serv., Act No. 100 (West 1996). Of course, by upholding the constitutionality of the statutes at issue herein, this Court has effectively removed the issue from the ballot and pretermitted a decision on the issue by the people of the State of Louisiana, as Section 2 of Act 100 of 1996 states that this issue shall not be submitted to the electors if this Court upholds the validity of the statutes at issue on rehearing.

[2] Though Article II, Section 4 of the Constitution of the State of Montana (the "Individual Dignity" clause) does not list "age" as a protected class, Article II, Section 14 of that Constitution states: "A person 18 years of age or older is an adult for all purposes, except that the legislature or the people by initiative may establish the legal age for purchasing, consuming, or possessing alcoholic beverages." As previously noted, see supra note 1 and accompanying text, the people of this State are, of course, free to amend their Constitution to remove the age classification protection entirely, or to explicitly except protection regarding the purchase, possession or consumption of alcoholic beverages, as has been done in Montana. Whatever their decision, amendment of the Constitution by the people of this State is clearly preferable to alteration of the document by judicial fiat.

[3] Obviously, this Court's analysis of the statutes at issue, as was done in our original opinion using the Sibley and Pace factors, can have no effect on any other particular age classification which exists in Louisiana law. It is elementary that a determination as to the constitutionality of a particular statute or statutes can have no effect on a determination as to the constitutionality of any other statute, as each case is decided (or should be) on the record evidence at issue in that particular case. There is, therefore, no basis in fact for any expressed or implied concern that a determination of unconstitutionality of the statutes at issue herein would automatically lead to a determination that all other statutes which classify persons on the basis of age are unconstitutional.

[1] Applying the same standard of scrutiny, the United States Supreme Court in Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976) held, in a case similar to the present case, that the relationship between the gender classification and the goal of traffic safety was far too tenuous to support a conclusion that the classification was substantially related to achievement of the statutory objective. The Court commented that the correlation was "an unduly tenuous `fit.'"