Editors-in-Chief

Arild Bjørndal
Centre for Child and Adolescent Mental Health, Oslo & University of Oslo, Norway
Julia Littell
Bryn Mawr College, USA

Search for:

print

Corporate Crime Deterrence: A systematic review

Review Protocol
 
Sally Simpson, Melissa Rorie, Mariel Elise Alper, Natalie Schell-Busey, William Laufer, N. Craig Smith  
01.05.2014  
Crime and Justice  
10  
4  
BACKGROUND
Corporate crime is a poorly understood problem with little known about effective strategies to prevent and control it. Competing definitions of corporate crime affect how the phenomenon is studied and implications for reducing it. Therefore, in this review, we use John Braithwaite’s definition (1984: 6) which specifies that corporate crime is “the conduct of a corporation, or of employees acting on behalf of a corporation, which is proscribed and punishable by law.” Consistent with this approach, this review focuses on various legal strategies aimed at companies and their officials/managers to curtail corporate crime. Interventions may be punitive or cooperative, but the goal is to prevent offending and increase levels of corporate compliance.

OBJECTIVES
Our overall objective is to identify and synthesize published and unpublished studies on formal legal and administrative prevention and control strategies—i.e., the actions and programs of government law enforcement agencies, legislative bodies, and regulatory agencies on corporate crime. We then assess the impact of these strategies on individual and company offending. Included are legal and administrative interventions such as new laws or changes in laws, inspections by regulatory agencies, punitive sanctions and non-punitive interventions aimed at deterring or controlling illegal behaviors.

CRITERIA FOR INCLUSION OF STUDIES
We were highly inclusive in our selection criteria, including studies that encompass a wide variety of methodologies: experimental (e.g., lab studies or vignette surveys), quasi-experimental (e.g., pre/post-tests), and non-experimental (e.g., correlational statistics using secondary data). The studies included also contained a wide variety of data (e.g., data from official agencies, corporate reports, individuals’ survey responses, etc.). Our search included published and unpublished articles, reports, documents, and other readily available sources. The outcome of interest, corporate offending, could reflect actual behavior or behavioral intentions as reported by respondents.

SEARCH STRATEGY
Our search strategy included numerous online databases and other sources (41 in total) to identify published and unpublished studies consistent with our definition of corporate crime. Sixty-nine search terms were generated to conduct the search, including specific terms associated with corporate offending (misconduct, corruption, unethical conduct, and organizational crime), offense type, and legal/regulatory strategies. For each database and search term, we tracked the number of “hits” gleaned from the search and determined whether the article related to corporate crime deterrence (broadly defined) and if there was quantitative evidence that might be coded for a systematic review. Studies meeting both criteria were considered “potentially eligible” and went through the next phase of eligibility coding.

DATA COLLECTION AND ANALYSIS
This research proceeded in two phases—the first phase (through 2003) collected studies of any intervention type (legal and otherwise) that focused on corporate crime prevention and control. The second phase of the project focused exclusively on prevention and control strategies solely in the legal area. Reflecting the broader focus of the first phase, we created a coding protocol that included the entire domain of corporate crime prevention/deterrence research which included the specific treatment variables we are interested in here—legal restraints. Two hundred and sixty five studies were eligible for inclusion. From each included study, we coded various forms of data classified into and calculated multiple types of effect sizes. These included 1) the “standardized mean-difference effect size” for analyses comparing two groups’ performance on a continuous outcome, 2) the “product-moment correlation effect size” for relationships in which both the independent variable and dependent variable are continuous (or assumed to represent a continuous construct), and 3) the “Odds-Ratio Effect Size” when both the independent and dependent variables involved are dichotomous. Interventions were classified into six treatments and broken down by subgroups (unit of analysis and whether the design was cross-sectional or longitudinal). We also conducted moderator analyses.


MAIN RESULTS
Out of the 40 possible treatment categories, we were able to calculate a mean effect size for 19. Although most showed a positive albeit non-significant treatment effect, some (including a significant effect) were iatrogenic. Looking at the specific mechanisms, the impact of law on corporate crime showed a modest deterrent effect at the firm and geographical level of analysis (there was not enough data to calculate effect sizes for individuals). However, this finding is limited to cross-sectional studies. For punitive sanctions, where there was substantially more data from which to calculate effect sizes, we observe a similar pattern: A tendency toward deterrence across units of analysis, with relatively few significant effects regardless of whether data are cross-sectional or longitudinal. The one area where there appears to be a consistent treatment effect is in the area of regulatory policy, but only at the individual level. Effects for other levels are contradictory (with some positive and others iatrogenic) and none are statistically significant. Regarding moderator effects, the least methodologically rigorous designs— those that were not experimental versus experimental designs and those without statistical control variables versus controls were associated with a treatment effect. We also found that older studies were associated with stronger deterrent effects—perhaps because the older studies are less methodologically rigorous that those that are newer. Other moderator results were less clear (publication bias, country bias, disciplinary bias; offense type), but given how few of the analyses revealed strong treatment effects overall we think it is premature to draw any conclusions from these findings and call instead for more methodologically rigorous and focused studies particularly in the punitive sanction and regulatory policy areas.
 
Share: