Quasi-Experimental Examination of Truth-in-Sentencing and Three-Strikes Laws

During the late 1970s and 1980s, crime rates and the perception of crime as a problem grew throughout the country, putting pressure on politicians, particularly Democrats, to respond (Cullen, 2018a; Ford, 2016). By 1990, crime rates in the United States (US) had reached historic highs and had been consistently rising for nearly three decades (Ford, 2016). Violent crime grew exponentially in the United States (US) between 1985 and 1990 with rates of crimes, like homicide, doubling (Blumstein & Wallman, 2006). As the US began to shift away from rehabilitating offenders there was an influx of punitive legislation passed. During the Reagan administration’s “war on drugs”, several bills passed during the 1980s (e.g., the Comprehensive Crime Control Act of 1984 (CCCA), and the Anti-Drug Abuse Act of 1986 (ADAA)) (Taifa, 2021. In the early 1990s, the Clinton punctuated the US’s adoption of the tough-on-crime approach with the passage of the Violent Crime Control and Law Enforcement Act of 1994 (VCCLEA), commonly referred to as “the crime bill”. Two of the more well-known sections featured in the crime bill, three-strikes laws and truth-in-sentencing laws, have been heavily criticized for their impacts on prison populations and lack of effectiveness in preventing other social determinants of crime (Ray & Galston, 2020; Alexander, 2019). These two provisions are the primary focus of this study. Much of the research examining the impacts of three-strikes and truth-in-sentencing laws on crime and incarceration up to this point has had limited ability to infer causality (Chen, 2008; Helland & Tabbarrok, 2007; Kovandzic et al., 2004; Sheperd, 2002; Marvell & Moody, 2001; Stolzenber & D'Alessio, 1997). The current study uses archival data from the US Census Bureau, US Bureau of Justice Statistics (BJS), The Centers for Disease Control and Prevention (CDC), and The Federal Bureau of Investigation (FBI) to create a state- x level panel data set with data between 1984 and 2019. Using a quasi-experimental design and difference-in-differences estimator (Callaway & Sant’Anna, 2021; Sant’Anna & Zhao, 2020) the current study assessed causal policy effects while accounting for variations in when each state passed the policy of interest in a way not previously done. The analyses estimated the policy effects on the outcomes of 1) violent crime rate, 2) property crime rate, and 3) overall incarceration rate for both three-strikes and truth-in-sentencing laws.