Graduate Driver Licensing: Lower Exposure/Better Driving

描述（由申请人提供）：关于青少年驾驶的文献表明，青少年发生机动车碰撞的风险较高。2003年，青少年（13-19岁）占美国人口的10%，占车祸死亡人数的13%。2001年，机动车碰撞是青少年死亡的首要原因。41%的女性青少年死亡和34%的男性青少年死亡是由机动车碰撞造成的。各种研究将这种风险水平的增加归因于以下因素，例如青少年驾驶经验水平低，倾向于在风险较高的时间（夜间）驾驶，并且通常在车内携带其他青少年乘客，导致驾驶员分散注意力的环境。从1996年开始，各州通过采用分级驾驶执照（GDL）政策来减少青少年的交通危险。以往的文献评估的影响，分级驾驶执照（GDL）的政策，20世纪90年代已经证明了政策的有效性，减少青少年涉及致命的碰撞。然而，许多问题仍然没有答案：它们为什么有效？它们是如何工作的？最初的效果会消失吗？如何改进它们？回答上述问题的困难是由于无法很好地估计在任何给定时间有多少青少年司机在路上，以及这如何随GDL政策而变化。特别是，有有限的证据表明，这些政策是否通过改善青少年驾驶行为，或通过简单地暂时限制青少年驾驶量，取得了有利的结果。为了填补这一空白，我们将首先建立一个模型，使用一个新的计量经济学框架，将青少年驾驶的数量和青少年驾驶的风险与每个州年对的州一级GDL政策联系起来。然后，我们将估计这个模型的参数使用的数据从致命事故报告系统（PARS）和国家公路交通安全管理局（NHTSA）的一般估计系统（GES），1990- 2005年。我们的主要目的是评估GDL政策在限制青少年驾驶量和改善青少年驾驶方面的相对有效性。我们还旨在了解GDL政策的哪些组成部分（成人监督，夜间限制，青少年乘客限制）有助于更好地驾驶，而不是限制驾驶。最后，我们想研究GDL政策的动态，并了解它们是否对改善青少年驾驶行为产生了持续的长期影响。

DESCRIPTION (provided by applicant): The literature on teenage driving indicates that teenagers are at elevated risk for motor vehicle crashes. In 2003, teenagers (ages 13-19) accounted for 10% of the U.S. population, and 13% of motor vehicle crash deaths. In 2001, motor vehicle crashes were the top reason for death among the teenagers. 41 % of the female teenage fatalities and 34% of the male teenage fatalities resulted from a motor vehicle crash. Various studies have attributed this increased risk level to factors such as the fact that teenagers have low level of driving experience, tend to drive at riskier times (night-time), and usually carry other teenage passengers in their car, leading to a distracting environment for the driver. Starting in 1996, states have responded to these concerns by adopting graduated driver licensing (GDL) policies to reduce traffic hazards for teenagers. Previous literature evaluating the impact of graduated driver licensing (GDL) policies of the 1990s has demonstrated the policies' effectiveness on reducing teenage involved fatal crashes. However, much remains unanswered: Why do they work? How do they work? Will the initial effects erode? How might they be improved? The difficulty in answering questions above have been due to the unavailability of good estimates of how many teenage drivers are on the roads at any given time, and how that varies with the GDL policies. In particular, there is limited evidence suggesting whether these policies achieve favorable results by improving teenage driving behavior, or by simply limiting the amount of teenage driving temporarily. To fill this gap, we will first build a model that relates the amount of teenage driving and riskiness of teenage driving to the state level GDL policies for each state-year pair using a novel econometric framework. We will then estimate the parameters of this model using data from the Fatal Accident Reporting System (PARS) and the National Highway Traffic Safety Administration's (NHTSA) General Estimates System (GES), 1990- 2005. Our primary aim is to assess the relative effectiveness of the GDL policies on limiting the amount of teenage driving and on improving teenage driving. We also aim to understand which components of the GDL policies (adult supervision, nighttime restrictions, teenage passenger limits) contribute to better driving as opposed to limited driving. Finally, we would like to study the dynamics of the GDL policies, and understand whether they have sustained long-term effects in terms of improved teenage driving behavior.