According to this study, it's a poor strategy:
> Adult-supervised settings for alcohol use resulted in higher levels of harmful alcohol consequences, contrary to predictions derived from harm-minimization policy. Findings challenge the harm-minimization position that supervised alcohol use or early-age alcohol use will reduce the development of adolescent alcohol problems. (J. Stud. Alcohol Drugs, 72, 418–428, 2011)
Influence of Family Factors and Supervised Alcohol Use on Adolescent Alcohol Use and Harms: Similarities Between Youth in Different Alcohol Policy Contexts. McMorris et al.
I also found this 2008 newspaper article that's clearly referring to a different study and which also warns against it. But I could not track down that study via the NIAA Website.