The fifth factor is the lack of consensus. A portfolio approach gives many actors a chance to contribute. Different communities, institutions, and individuals have different resources and different ideas about which approaches will be useful and effective. In a world in which people disagree about which interventions are best and in which it will be valuable to engage many actors in the effort to deal with the problem, it would be a serious mistake to insist that only one approach be used.

To say that the committee decided to recommend a portfolio of approaches, however, is not to say that comparative judgments concerning the relative effectiveness of different instruments must be avoided or that individual components of the strategy cannot be implemented independently from the others However, we propose a comprehensive strategy that we believe will be cost-effective based on the notion that several instruments will be reinforced by the addition of other instruments as they help to reach a problem that is missed (or created) by a particular policy or as they provide hedges against uncertainty or opportunities to learn. Evidence from youth smoking prevention policy reinforces the notion that a comprehensive, multifaceted approach is likely to be more effective than any single approach (Lantz, 2004).

But the balance among these instruments has to reflect a clear conception of both the nature of the problem and the reasons for selecting the chosen strategy. We present our overall analysis of cost-effectiveness at the end of Chapter 12 after more fully discussing the individual components of the strategy.

The committee also considered what Congress meant by a “cost-effective” strategy and what data and analysis are needed to assess cost-effectiveness. We note that such an assessment involves more than the usual question in program evaluation, which focuses simply on whether a particular policy “works” to produce the desired effect (or effects).

What did Congress mean by effectiveness? Presumably, one key measure of effectiveness is simply reducing the numbers of youth who drink alcohol at all before they turn 21. To the extent that the law treats all drinking by people under 21 as illegal and to the extent that the goal of any law is to get to as close to complete compliance as possible, the ultimate test of effectiveness would be the degree to which underage drinking stopped altogether. However, given that alcohol use is regarded as entirely appropriate for adults and that this normative stance (and the policies it spawns)