酒駕罰則修訂對累犯及非累犯之政策效果評估

The Effects of DUI Regulation Revisions on Recidivists and Non-Recidivists.

康江良(Jiang-Liang Kang)

指導教授 ： 駱明慶

國立臺灣大學/社會科學院/經濟學系/碩士(2022年)

https://doi.org/10.6342/NTU202201082

摘要

我國酒駕相關法規於 2013 年及 2019 年大幅修訂, 除全面下修酒精濃度標準、持續提高罰鍰與刑度等, 更在2013 年首度訂定累犯處罰專項, 復於 2019 年大幅加重累犯之罰刑處遇。惟 Becker(1988) 理性成癮理論說明成癮者追求效用極大之計畫, 相對未成癮者較不具時間一致性; Sloan et al.(2014) 更實證論結酒駕者無法在喝酒前對法律規範先行理性計畫。爰此, 酒癮程度不同的酒駕累犯與非累犯, 對修法反應可能不同, 修法效果持續性亦可能互異。本研究乃利用交通違規資料註記酒駕累犯, 並過錄到酒駕交通事故當事人資料中, 將酒駕事故傷亡人數及人車相關特徵, 利用差異中的差異法 ( Differences - in - Differences ) , 估計兩次修法對累犯與非累犯酒駕事故每日傷亡人數之影響。

實證顯示, 酒駕累犯在兩次修法後 1 - 4 個月累計每日事故傷亡人數降幅均不顯著, 且小於-10% ; 主要受影響者為非累犯, 其各月累計效果均顯著下降, 且降幅逐月遞減並介於 -31% 至 -11% 。惟因2019 年累犯各月累計效果多數低於 2013 年, 致兩次修法雖整體酒駕事故每日傷亡人數多數仍顯著下降, 但 2019 年降幅普遍較 2013 年低落。此外, 非累犯無照酒駕事故傷亡人數顯著少於有照者, 但 2019 年二者差距已較 2013 年縮小; 累犯則 2019 年無照者所致傷亡人數反多於有照者,以吊扣駕照遏止酒駕效果減弱。高教人口比率及平均所得等居住特徵對酒駕事故傷亡人數之影響, 從累犯擴及至非累犯。

關鍵字

酒醉駕車 ； 累犯 ； 差異中的差異法 ； 交通事故

並列摘要

Drunk-driving regulations were significantly amended in 2013 and 2019, including lowering alcohol concentration levels, toughening penalties, increasing fines, etc. Regarding repeated Driving Under Influence (DUI), targeted penalties were first formulated in 2013, and the punishment for repeated DUI offenders has greatly increased in 2019. According to Becker (1988), rationality is defined as a consistent plan to maximize utility over time, but the consistency of time preferences is unstable for addicts. Moreover, Sloan et al.(2014) found that drunk drivers are less likely to plan alcohol-related events before drinking. Due to their different levels of alcohol addiction, DUI recidivists and non-recidivists may respond differently to revised regulations. This study aims to understand the impact of revised DUI regulations on both recidivists and non-recidivists. Repeated DUI offenders in traffic violation data were coded and compiled into DUI traffic accident data. Differences-in-Differences (DID) method was used to estimate the variations in daily casualties caused by DUI recidivists and non-recidivists, which were adjusted for regulations.

Results show that 1 to 4 month cumulative daily casualties of DUI recidivists did not significantly change after two amendments, however those of non-recidivists significantly dropped between -31% and -11%. The two amendments mainly affected non-recidivists. Overall daily DUI casualties were significantly decreased from -18.7% for the first month to -7.5% for the 4 months cumulative in 2013; significant decrease is shown from -11.0% for 2 months cumulative to -7.0% for 4 months cumulative in 2019. Besides, daily DUI casualties among unlicensed recidivists increased more than those with a license in 2019. A DUI's license suspension would therefore have a decreasing distinct effect in deterring them from drunk driving. Residents' characteristics, including tertiary education and income, are increasingly having an impact on the number of drunk driving casualties.

並列關鍵字

DUI ； Recidivist ； Differences-in-Differences ； Traffic accident

參考文獻

Becker, K. M., G. S.and Murphy (1988) “The theory of rational addiction.,” The Journal of Political Economy, 96, 675–701.

Bertrand, M., E. Duflo, and S. Mullainathan (2004) “How Much Should We Trust Differencesin-Differences Estimates?” Quarterly Journal of Economics, 119, 249– 275.

Carpenter, C. (2006) “Did Ontario’s Zero Tolerance Graduated Licensing Law reduce youth drunk driving?” Journal of Policy Analysis and Management, 25 (1), 183–195.

Chang, H., K. Chang, and E. Fan (2019) “The Intended and Unintended Effects of

DrunkDriving Policies,” Oxford Bulletin of Economics and Statistics, 82 (1), 23–