Do Changes in the Economy Affect Teenage Drug Use?

虽然社会和心理因素通常被认为是青少年吸毒的最重要决定因素，但其他因素也可能发挥重要作用。在这项拟议的研究中，我们将研究一个以前未探索过的非法药物使用因素，该因素有可能解释青少年用途：经济实力。1990年代，经济稳步改善，青少年吸毒人数大幅增加。然而，这两种趋势是否相关尚不确定。人们可以想象，一个更强大的经济将为青少年提供更多的就业机会，这将提供购买毒品的收入。与此同时，强劲的经济将提供更多的就业机会和更高的工资，使用毒品的机会成本将更高，因为使用毒品可能占用工作时间，并可能影响工作业绩。我们建议研究青少年吸毒的变化与国家失业率的变化。这项研究可能有助于我们更好地了解是什么驱使青少年吸毒，以及为什么青少年吸毒在20世纪90年代增加了这么多。此外，通过这项研究，评估人员可以分析出经济的影响，以获得更准确的估计如何有效的禁毒运动。为了确保未观察到的异质性不会影响估计值，我们将估计双变量模型，其中允许未观察到的使用药物的倾向与未观察到的额外工作时间的倾向相关。可以对此处制定的框架进行修改，以探讨吸毒的其他原因和后果。

DESCRIPTION (provided by applicant): While social and psychological factors are typically considered to be the most significant determinants of teenage drug use, other factors may also play an important role. In this proposed research, we will examine a previously unexplored factor of illicit drug use that has the potential to explain a significant portion of the trends in teenage drug use: the strength of the economy. During the 1990s, the economy steadily improved and there was a large increase in teenage drug use. It is uncertain, however, whether these two trends are related. One can imagine that a stronger economy would offer greater job opportunities for teenagers, which would provide income to buy drugs. At the same time, with more job opportunities and higher wages that a strong economy would provide, the opportunity cost of using drugs would be higher because using drugs could take away from time working and could affect work performance. We propose to examine how teenage drug use changes with variation in state unemployment rates. This research may help us better understand what drives teenage drug use and why teenage drug use increased so much during the 1990s. In addition, with this research, evaluators could parse out the effects of the economy to obtain more accurate estimates of how effective anti-drug campaigns have been. To ensure that unobserved heterogeneity would not affect the estimates, we will estimate bivariate models in which the unobserved propensity to use drugs is allowed to be correlated with the unobserved propensity to work additional hours. The framework developed here could be modified to explore other causes and consequences of drug use.