By Maëlle Stricot (Paris School of Economics & University of Paris I: Panthéon-Sorbonne)
Picture this: it’s 8pm in France. Millions of people are watching the country’s prime-time national TV newscasts. Tonight’s program features a recent case of sexual assault or intimate partner violence.
Does that burst of public attention change how the justice system responds, in the days that follow, to the hundreds of sexual and intimate partner violence cases already moving through the system?
In my job market paper, I show that it does, but only at a very specific point in the process. Media coverage does not make professional judges harsher. Instead, it affects prosecutors, who make the initial decision about whether a case moves forward, making them more likely to prosecute these cases.
This matters because violence against women is not only massively under-reported worldwide; it also faces steep attrition once cases enter the justice system. In many countries, most cases are closed early in the process, when prosecutors decide whether to prosecute or dismiss them. In France, nearly 80% of cases recorded by the justice system are closed at this stage (Figure 1), which raises questions about the factors that influence these decisions.
Figure 1: Judicial handling of sexual and intimate partner violence cases
A large body of research has examined how context and emotions can shape judges’ decisions, including the role of the media (Ash & Poyker, 2024; Lim et al., 2015; Philippe & Ouss, 2018). Yet despite how central prosecutors are to criminal justice systems, we still know surprisingly little about what shapes prosecutors’ decisions (Garoupa, 2012; Tonry, 2012; Voigt & Wulf, 2019).
When evidence is scarce, discretion matters
Why do so many of these cases stop early? Part of the answer lies in evidentiary constraints. Sexual and intimate partner violence often occurs in private settings, with few witnesses and limited material proof, which makes cases difficult to investigate and prosecute. Those constraints also leave substantial room for discretion: how much investigation to request, how to assess credibility, and where to set the threshold for moving forward. In that space, broader norms and beliefs about these crimes may also influence how cases are handled.
If discretion and norms matter, then salience and public attention may matter too. For instance, by making violence against women more visible, news coverage can shape public expectations about how these crimes should be handled and potentially influence decision-making inside the justice system.
So the question becomes: does greater visibility of violence against women affect how the justice system handles these cases?
A high-frequency visibility shock
To answer this question, I bring together two rare sources of data. First, I use administrative records covering nearly all sexual and intimate partner violence cases processed in French correctional and juvenile courts between 2012 and 2019. These data let me track each case through the process, from the initial report to the prosecutor’s decision to prosecute or dismiss and, when prosecuted, to conviction outcomes and sentence lengths.
Second, I use the complete, day-by-day content of the prime-time evening news on France’s two largest TV channels, TF1 and France 2, watched by millions every night (around 10.4 million viewers per day on average). About once every nine days, these broadcasts feature a story about violence against women.
The key insight is timing. In the short run, it is hard to predict when the 8pm news will cover violence against women: some days include such stories, while others do not. Crucially, I focus on coverage of crimes against women that are not the same cases currently being processed. This generates high-frequency, plausibly quasi-random variation in TV news coverage, so that some case decisions happen shortly after a day with coverage, while others happen after similar days without coverage.
By comparing these otherwise similar moments, while accounting for seasonal patterns and case characteristics, I estimate how shifts in media visibility translate into changes in prosecutorial and judicial decisions in the week that follows. To support this design, I show that coverage is not systematically related to observable case characteristics and that there is no sign of strategic timing around coverage, making cases decided just after coverage comparable to those decided on other days.
One more story at 8pm, and more cases move forward
So what happens after violence against women becomes prime-time news? Prosecutors prosecute more. In the week following news coverage of crimes against women, prosecutors are less likely to dismiss sexual and intimate partner violence cases (Figure 2). In my data, this translates into a 2.3% increase in the prosecution rate over the next seven days. Scaled up, the estimates imply roughly 260 additional cases prosecuted per year, on average.
Figure 2: Effect of TV news coverage of crimes against women on the dismissal rate of sexual and intimate partner violence cases
Judges, however, do not change their behavior. I find no effect of news coverage of crimes against women on judges’ decisions regarding conviction and sentence lengths in the short run. This occurs in a context where prosecutors largely determine case outcomes: while they dismiss nearly 80% of cases at the outset, most prosecuted cases ultimately result in conviction (Figure 1).
Consistent with this pattern, cases prosecuted following news coverage of violence against women appear just as likely to result in conviction as other prosecuted cases, with an increase in the long-run conviction rate equivalent to the increase in the prosecution rate. This is reassuring, because one natural concern is that heightened media attention might push prosecutors to bring weaker cases to court, which would then be more likely to fail at trial. Instead, the results suggest that visibility helps additional legally viable cases clear the prosecutorial threshold that they otherwise would not.
Why might this happen? I cannot conclusively identify a single mechanism, but the heterogeneity patterns are suggestive. They are consistent with a mix of short-term, unconscious cognitive or emotional reactions to increased salience (priming) and strategic responses to temporary increases in public scrutiny and accountability (reputational concerns).
So, is this good news?
Media visibility of violence against women increases prosecution rates for sexual and intimate partner violence cases in the short run, without detectable changes in judges’ conviction or sentencing decisions. Importantly, the additional cases brought forward appear just as likely to result in conviction as those prosecuted otherwise, suggesting that some viable cases might have been dismissed prematurely in the absence of this salience shock.
A natural policy concern is whether this comes at the expense of other priorities. Reassuringly, I find no evidence of crowding out other serious violent crimes against persons in the short run.
What this means for policy
Beyond violence against women, the results highlight a central feature of criminal justice systems: prosecutorial decisions are a key chokepoint, and they are malleable. These decisions do not respond only to the level of evidence in a case, but can also react to broader contextual factors, such as media visibility and public attention.
Two lessons follow. First, continued discussion of violence against women in the public sphere may matter, since greater salience may help additional legally viable cases reach trial while leaving judges’ decisions unchanged. Second, a more persistent effect on prosecution rates may rely on channels other than media reporting. Potential channels to consider could include clearer prosecutorial guidelines, more systematic review practices for dismissals, and targeted training for legal actors on potential unconscious biases related to these crimes, so that legally viable cases are prosecuted more consistently over time.
My job market paper was awarded the 2025 Young Researcher Prize from the French Association of Law and Economics and Institut Robert Badinter
About the Author
Maëlle Stricot is a PhD candidate in Economics at the Paris School of Economics and Paris 1 Panthéon Sorbonne University.
She is interested in crime, gender, and labor economics. To learn more about her research, visit: https://sites.google.com/view/maelle-stricot/home
Social Media Handles:
