1. Introduction
Traditionally, comparatism is about comparing what is not evidently comparable, by dissecting the conditions that allow comparability. Conversely, comparatism is also about questioning and probing what is too quickly compared. Comparatism involves making comparisons with an awareness of hidden hypotheses and a commitment to bring these hypotheses forth, in pursuit of science and falsifiability. The scope of this article is to illustrate one specific instance of such circumstance and develop the due diligence of comparatism enunciated above.
Comparatism in the administration of justice is a developed and institutionalized field of study within the rule of law. The European Commission for the Efficiency of Justice (CEPEJ) is an example of one such field.1 The CEPEJ supports the Council of Europe in issuing its evaluation on the state of the rule of law in the 48 countries that compose ‘geographical’ Europe. The CEPEJ sets the standard for the quality of judicial systems. Among other activities, the CEPEJ collects and compares a vast number of variables about the European criminal justice systems. With regard to the outcomes of the criminal trial proceedings, the CEPEJ collects data on Disposition Time, which is currently the standard measure for international comparison and regulation.
One important circumstance to utilize comparatism within the criminal justice systems of Europe – and beyond – happens with a variable for which the CEPEJ does not collect data: conviction rates. Attention is called to such circumstance by the mere observation of criminal conviction rates that are quantitatively very different across countries.
The focus of this article is to compare conviction rates across selected countries, something that has not been done before. Different conviction rates, raise the question: why are they so different? The immediate answer is because the legal systems are different. And yet, paraphrasing American poet Gertrude Stein’s ‘a rose is a rose is a rose’, a crime is a crime is a crime: reality is hardly different, and differences in conviction rates should be explainable through analysis of institutionalized judicial activities.
The research question and the aim of this article is whether it is meaningful and valid to compare conviction rates across different countries by placing them side by side in a table and whether these rates are actually measuring the same thing in each country, so that they can be fairly compared.
This approach is relatively original. Rasmussen et al. used a similar approach, for different states within the U.S.. This article is in that same strand, with an original touch on international comparison, which makes for fewer and simpler data, but for a very interesting question, seeing the big differences among those data.
We will explore the different conviction rates in the criminal justice systems across the United States of America (U.S.) (federal), England and Wales, and Italy. The motivation behind these selections vary. Firstly, the subject treated here started with scrutiny from the European Commission2 on judicial reform in Italy, which was followed up by a reform of criminal justice procedure.3 Italy has a relatively low conviction rate4 and comparison with U.S. federal system is pursued as the U.S. represents the opposite extreme in the system’s performance. England and Wales represents a bridging case between the two extremes. On the one hand, the U.S. and England and Wales judicial systems have fundamental similarities. On the other hand, the U.S. federal system only deals with a subset of criminal law throughout the country. Whereas, England and Wales has a population very similar to Italy and deals with all criminal cases, just like the Italian system.5
The methodological approach of this article is comparative and quali-quantitative. This article examines qualitative and quantitative differences in procedure and in practice, making use of simple comparative statistics. The paradigm is regulatory rather than radical; it focuses on possible reform of procedure which does not question the foundations of the judicial organization and the roles of judges and prosecutors.
The structure of the article is as follows: Section 2 presents empirical evidence and recalls the relevant literature in the domain of comparatism and organizational behavior. Section 3 illustrates legal systems and criminal procedure in a comparative perspective; Section 4 presents the analysis and findings; Section 5 discusses impact and possible remedies; Section 6 discusses limitations and future studies; and Section 7 concludes on the findings of this article.
2. Empirical evidence and relevant literature
2.1 Conviction rates
We show the basic data that has triggered this study: the conviction rates. Table 1 (first column) shows a conviction rate of approximately 92% for the U.S. Federal proceedings.6 The conviction rate is the sum of 90% guilty pleas7 and 2% of individuals who had pled not guilty went to trial and were found guilty, thus generating a conviction rate of 92%. More broadly, in the U.S., “about 90 percent of the federal defendants and 75 percent of the defendants in the most populous counties were found guilty.”8 “The conviction rate in England and Wales was 83.6% in Magistrates Court and 77.9% in the Crown Court, as of the first quarter of 2022.”9 The Italian conviction rate was 52.2% for the solar year ending June 2022.10
Table 1
Comparative conviction rates 2022, first degree, percentage and absolute values.
| JURISDICTION | CONVICTION RATE (PERCENTAGE) | TOTAL CASES |
|---|---|---|
| U.S. Federal | ||
| Guilty pleading | 90.0 | 63,073 |
| Pleading not guilty | 2.0 | |
| Terminations | 92.0 | 70,223 |
| England and Wales | ||
| Magistrates Court | 83.6 | |
| Crown Court | 77.9 | 96,000 |
| of which guilty pleas | 66.0 | |
| Italy | ||
| Terminations (court decisions) | 52.2 | 262,766 |
| of which guilty pleas | 6.9 | 18,041 |
Table 1, second column, shows the number of cases in each country. In the U.S. federal system, between 2021–2022, there were 63,073 cases of a guilty plea over 70,223 total defendant terminations (court decisions).11 The percentage of cases of a guilty plea over total defendant terminations were 90%. In England and Wales, 66% pled guilty at Crown Court in the year ending June 2023.12 That same Crown Court disposed of 96,000 cases.13 In Italy, in year ending June 2022 there were 18,041 cases of guilty plea over total defendant terminations of 262,766 cases.14 The percentage was 6.9%.
If this is the underlying data, the central question becomes: what are the behaviors and the principles of the legal systems that generate this data? Put differently: what are the principles – explicit or implicit – that this data reveals about the legal systems from which it comes? First, we need to ask the question if anybody studied this before, namely: a) previous literature, composed of two different areas: a.1) comparative legal systems: CEPEJ; and a.2) comparative conviction rates: Rasmusen et al. Then, we can move on to our subject matter, composed of two areas: b.1) description and analysis of the different legal systems; and b.2) description and analysis of comparative criminal procedure.
2.2 The European Commission for the Efficiency of Justice – CEPEJ
The major and institutionalized source of comparative work on justice systems – on the Western side of the hemisphere – is the CEPEJ.15 CEPEJ collects data on the justice systems of 48 country systems. The UK is represented by three of these 48 systems: England and Wales, Scotland, and Northern Ireland. CEPEJ publishes a report called ‘European judicial systems: Efficiency and quality of justice’. Such report has all kinds of data about justice organizations and their functioning such as resources to measures of output, like Disposition Time. CEPEJ not only collects and distributes data but also advises the Council of Europe in issuing their evaluation of judicial systems and making recommendations which countries are expected to abide by.. However, CEPEJ does not collect data about conviction rates. Additionally, CEPEJ does not include data from the U.S. Despite these omissions, CEPEJ remains a major source of credibility for European judicial systems.
2.3 Literature on comparative conviction rates
There are instances of grey literature about conviction rates or criminal procedure in general.16 However, a unique academic source on comparative conviction rates is Rasmusen et al. Rasmusen et al. performed a regression analysis on cases in the U.S. and concluded that the U.S. prosecutorial reality ‘tradeoffs theoretically in two models, one of a benevolent social planner and one of a prosecutor who values not just the number of convictions but the conviction rate and unrelated personal goals.’17 In fact, ‘higher budgets are associated both with higher number of convictions and with higher conviction rates’.18 Rasmusen et al. thus confirms a basic model of the prosecutor’s ‘production function’ in which conviction rates and cases sent to court are in a negative relationship with each other.
2.4 Literature on organizational behavior
It must be remarked that the literature (Rasmusen et al.) does not imply at all that pursuit of ‘unrelated personal goals’ consists of criminal activities. The goals meant in the literature might be the study of law or any other perfectly legal activity. To better corroborate the notion of ‘unrelated personal goals’, let us introduce the concept from the general organizational behavior perspective. Jurists have appreciated a theory of organizational behavior that qualifies conservative decision-making as ‘defensive administration’. ‘Defensive decision-making occurs when one ranks an option as the best for the organization yet deliberately chooses a second-best option that protects him or herself against negative consequences.’19
Artinger et al. developed an empirical study through interviews of hundreds of public employees, which can be qualified as a ‘boots on the ground’ approach to the study of public decision-making. They went out and asked public employees themselves what they thought of their own behavior, in their own awareness. Artinger et al. came to the conclusion that ‘a major cause [of defensive decision-making was] a team’s approach to failure, that is, whether the reaction to failure is to seek someone to blame as opposed to identifying the underlying causes in order to learn how to prevent similar failures in the future.’20 In such circumstances, there is little incentive in ‘identifying the underlying causes [of failure] in order to learn how to prevent similar failures in the future’.
3. Description and analysis of the three different living legal systems
We have argued that comparison of different legal systems is part of routine institutional work – for example, by CEPEJ – and that it is therefore possible. However, the fact that CEPEJ does not collect data about conviction rates tells us that the story is more complicated. On the other hand, Rasmusen et al. work in a single country and one legal system, the U.S. Their regression data panel is about diverse jurisdictions which apply the same rules and work with the same laws. Our ambition is to compare conviction rates across countries and legal systems. So, in order for us to argue the comparability of conviction rates, before we even try to explain their differences, we need to get a little deeper into the different legal systems. So, after having laid out our quantitative data, we may begin our qualitative analysis of the different legal systems.
3.1 Two legal families, three legal systems
The research questions we deal with in this article take place against the backdrop of two different legal families: civil and common law systems. For sake of completeness then, it is important that we ask ourselves what these systems are and in what ways they may explain or hinder the research questions we are applying to our three living legal systems: U.S., England and Wales, and Italy.
The two legal systems are common law, present in the U.S. and England and Wales, and civil law, present in Continental Europe. The most significant difference between common law and civil law lies in the structure of the proceedings, and this directly affects the role of the defense. Such structure can take two basic shapes: adversarial system (Common Law) and inquisitorial system (Continental Europe).
Adversarial system (Common Law) has the following characteristics:
Party-driven process: The prosecution and the defense are equal opponents in an adversarial setting.
Judge as referee: the judge does not conduct the investigation but rather ensures procedural fairness and rules on the admissibility of evidence.
Evidence presented by the parties: the defense plays an active role in the presentation and testing of evidence (e.g. through cross-examination, preparation of witnesses).
The principles of orality and immediacy dominate the proceedings.
Truth is sought through the contest between the parties.
The defense has a central, combative role – not merely protecting rights but actively shaping the search for truth.
Inquisitorial system (Continental Europe)
Investigative principle: The proceedings are state led, often by the public prosecutor or investigating judge.
Active role of the court in establishing the facts.
The judge seeks the truth: judges take responsibility for evidence gathering and evaluation.
The defense is more reactive: it observes the proceedings, submits motions, and ensures rights are respected, but has less influence on the investigation.
The written procedure is more prominent.
The defense has a more passive, protective function, guarding the rights of the accused within a process largely steered by state authorities.
Before we proceed to examining other implications of the two different legal systems, it is important to note that Italy appears to have embraced an adversarial system, too. The Italian Constitution – that came into force on January 1, 1948 – at Article 27, establishes the principle of the presumption of innocence, stating that “the accused is not considered guilty until a final conviction.” This principle implies that throughout the criminal process, the accused person is considered innocent until a final conviction. Fuller implementation of the adversarial system came with the Vassalli Reform in 1988 (after the then minister of justice Giuliano Vassalli). Before then, an inquisitorial system from the fascist era (the so-called Rocco Code, after the then minister of justice Alfredo Rocco, 1930) applied.
As an empirical point, supporting the idea of an adversarial system, one might expect that an inquisitorial system would ex ante lead to expecting higher conviction rates than an adversarial system, as, in an inquisitorial system, the defendant has to prove their innocence. Keeping in mind the 52.2% conviction rate for Italy (See Table 1 above), if the system were inquisitorial, one would expect ‘Chinese’ conviction rates of over 99% rather than ‘Indian’ conviction rates in the neighborhood of 54%.21 Such observation leads us to hypothesize that possible differences in the role of the defense do not significantly affect conviction rates as we have observed them above, in the introductory quantitative paragraphs. Italy has lower conviction rates than the U.S. and England and Wales.
Once we have outlined the extremes of the spectrum in the structure of the proceedings – adversarial system (Common Law) and inquisitorial system (Civil Law) – let us move on to two other possible differences: the sources of law and the criminal principles that inform the adversarial or the inquisitorial systems.
In regard to the sources of the law: common law vs. civil law. However, common law systems have also integrated their precedent sources within laws and codes; hence, the two legal systems tend to converge.
Under criminal law principles, the adversarial system implements in practice some principles that are shared across the three legal systems we are examining here: 1) Reasonable time; 2) Presumption of innocence; and 3) Immediacy, orality, and concentration. Reasonable time derives – in Europe – from Article 6 of the European Convention on Human Rights (ECHR), which guarantees the right to a fair trial within a ‘reasonable time’. Reasonable time is also in the U.S. Constitution (the sixth amendment). The presumption of innocence is also enshrined in all three systems. Specifically, presumption of innocence leads to ‘reasonable cause’ and ‘reasonable probability of conviction’, and – in the case of Italy – it was also sanctioned by the Cartabia Reform of 2021.22 The principles of immediacy, orality, and concentration are derived from the characteristics of the adversarial system outlined above. Immediacy emphasizes the direct presentation of evidence to the decision-maker, ideally through oral testimony at a hearing. Orality refers to the reliance on spoken rather than written communication in the legal process. Concentration dictates that proceedings should be conducted with minimal interruption and focus on the relevant issues.23 According to the principles of immediacy and orality, a judgment may only be based on what has been orally presented to the court by the parties at the main hearing. It is generally considered that the principles of immediacy, orality and concentration facilitate the application of the court’s discretion in terms of assessing the value of evidence.
Finally, one more element of characterization of the different legal systems is the position of the prosecutor within the constitutional framework. Different from the U.S. and the England and Wales, the Italian prosecutor is not part of the executive branch, but it is a member of the judicial branch. They have the same status as judges. Italian prosecutors and judges are under the control of a judicial self-government body. The self-government body is a Constitutional mandate. Differently, in the England and Wales the prosecutor appears to be positioned at arm’s length vis-à-vis the executive branch. Whereas the U.S. prosecutor is positioned within the executive branch.
So, the two legal systems may seem far apart, but the common ground between England and Wales, U.S. and Italy is not that small, and we can conclude that the three legal systems are more comparable than it would appear prima facie from a top-down point of view.
3.2 Comparative criminal procedure
Coming now to comparative criminal procedure.24 In this section, we examine the procedural circumstances that have been foreshadowed above, whereby judges and prosecutors operate and make their own decisions: burden of proof, statute of limitations, plea dealing, planning policy. In the following we provide an overview of criminal procedure in the U.S., England and Wales and Italy. We closely look at the decision-making chain of intermediate proceedings between the investigation and the beginning of trial.25 The criteria to select the topics analyzed are to understand the points that may hamper the action of the system and make one system produce fewer convictions than the other.
In the examined countries the burden of proof is on the prosecution, and criminal proceedings are divided into three stages: investigation, intermediate proceedings, and trial. The length of investigation appears to be limited in the U.S. only by statute of limitations depending on the crime whereas in Italy it appears to be mostly limited to eighteen months, also depending on the crime.26 England and Wales do not have limits for criminal cases. The prosecutor decides when the investigation is complete and decides whether to discontinue the proceedings or to press charges on the suspect.
In the intermediate proceedings a court assesses the charge and decides whether prosecution should proceed. In the U.S. a grand jury determines whether there is enough probable cause to go to trial.27 In the U.S. the pre-trial is called ‘preliminary hearing’ and it is a ‘mini trial’.28 In England and Wales, the Magistrates Court assesses the seriousness of a crime and sends ‘indictable only’ offences to the Crown Court. The case of ‘Either-way’ offences is interesting as they can be heard in Magistrates’ court or the Crown Court, and the defendant can make a choice. Such proceedings are called pre-trial court hearings.29 In Italy, the court that assesses the charge and decides whether the prosecution should proceed is the ‘judge for preliminary hearing’.30 In Italian practice, only pre-trial motions are entertained in the pre-trial hearings, and the pre-trial court does not check the available evidence.
When the investigation is finished, in the U.S., the prosecutor presses charges in front of the grand jury. If the prosecutor is successful, indictment follows, and arraignment of the defendant takes place in front of a judge. Such process leads to the arrest of the defendant due to the charges against them. Arrest in England and Wales appears to be warranted only under specific circumstances: need for immediate questioning, and probability of escape or reoffence.31 As with the other countries, arrest of the defendant in Italy is warranted only in specific circumstances defined by the constitution: peril of escape, manipulation of proof, or reoffence.32 If the defendant is arrested, there is no pre-trial hearing, and the case is sent to court for trial by the ‘judge for preliminary investigation.’33
Plea deal and plea bargaining are key moments in the U.S. procedure. At arraignment, the prosecutor may offer the defendant a plea deal. The defendant may either accept the plea and plead guilty or reject the plea and plead not guilty. If the defendant pleads not guilty, the defendant refuses the deal and exercises their right to a trial. Then, the case proceeds to trial. If the defendant pleads guilty, then a negotiation between the prosecutor and the defendant takes place regarding the charges and penalty. Such a negotiation is ratified by the judge. Negotiations take place instead of a trial.34 ‘Bargaining is not officially part of the system in England and Wales, except in complex fraud cases, but the judicial sentencing guidelines suggests those who plead guilty at the earliest hearing over other crimes may be given a reduction of up to a third of their sentence’.35 Such procedure of England and Wales is available in Italy as well, with the same benefit of one third of the sentence36 applied albeit less used in practice. Further, in the U.S., a plea deal appears to be more flexible than in England and Wales, and Italy. Of course, when a defendant pleads guilty, this counts as conviction in the statistics.
Quantitative evidence on plea dealing is consistent with the above qualitative description of the procedure. As shown in section 2 on empirical evidence, in the U.S. federal system, between 2021–2022, there were 63,073 cases of guilty plea over 70,223 total defendant terminations.37 The percentage of cases of guilty plea over total defendant terminations were 90%. In England and Wales, 66% pled guilty at Crown Court in the year ending June 2023.38 That same Crown Court disposed of 96,000 cases.39 In Italy, in year ending June 2022 there were 18,041 cases of guilty plea over total defendant terminations 262,766.40 The percentage was 6.9%.
A ‘silent’41 element of procedure is the planning policy of trial hearings. In fact, a key difference between the U.S. and the England and Wales systems vis-à-vis Italy appears to be the practice of planning trial hearings. In the U.S. and England and Wales trial hearings are planned in series or sequentially: one hearing of a trial is planned next to the previous hearing of the same trial and one trial is planned after the previous trial is finished. Whereas, in Italy trials are planned in parallel: hearings of one trial are intertwined over time with the hearings of many other trials, possibly rotating the whole caseload of a court bench between one hearing and the next hearing of the same trial.42 More specifically, any agenda setting implies some parallel activities, but in sequential planning there comes a point where all trial activities are carried out in one single hearing (as in the Magistrates Court) or in sequential hearings (as in the Crown Court), day in and day out, or one day after another.
Wrapping up this excursus on the procedure, judges act as ‘pacemakers’ of the trial, because it is the judges that set the hearings planning policy. Prosecutors can be thought of as the “gate keepers of criminal justice.”43 In fact, according to the United Nations Office on Drugs and Crime (UNODC): ‘Public prosecutors perform a crucial role in society. They are the ‘gate keepers’ of criminal justice, insofar as without their initiative there cannot be prosecution and repression of crimes. Prosecution services are, in fact, society’s principal means of pursuing criminal behavior and its interface with the adjudicative power. Only cases that are brought to courts by public prosecutors can be processed and adjudicated by judges. Other forms of prosecution (like private prosecution) exist only in a few countries and apply in limited circumstances.’44
Another key element affecting criminal procedure is constituted by the incentives that the system places on individual prosecutors. In the U.S. federal judicial system, prosecutors are part of the executive branch and are evaluated on their performance,45 England and Wales Crown Prosecution Service also monitors conviction rates and other parameters.46 Whereas Italian prosecutors are part of the judicial branch of government and share the same status and the same career with judges. In the Italian procedure, prosecutors do not have any personal incentive to pursue the conviction of defendants, because the outcome of trials are not considered in the evaluation of prosecutors or judges for advancement in their careers. Rather, career advancement is based on seniority.47
We can convene at this point that all the actors involved in the case (judges, prosecutors, defense lawyers) are responsible for the end result of the process. Specifically, what comes after the ‘gate’ of entrance into the justice system, may influence what is acted on before the gate. This seems to suggest that the prosecutor’s behavior in sending cases to court may very well factor in and anticipate the practice of what will happen next More specifics and comments on the comparative procedure will be warranted by the following quantitative data.
4. Analysis and findings
U.S. Federal data show a conviction rate of 92%.48 The total conviction rate is the sum of 90% guilty pleadings49 and 2% of those who had pled not guilty, went to trial, and were found guilty; thus, this generates a conviction rate of 92%. More broadly, in the U.S., ‘about 90 percent of the federal defendants and 75 percent of the defendants in the most populous counties were found guilty.’50 ‘The conviction rate in England and Wales was 83.6% in Magistrates Courts and 77.9% in the Crown Court, as of the first quarter of 2022.’51 Italian data for the solar year ending June 2022, conviction rate was 52.2%.52
A prima facie explanation for the U.S. conviction rate (92%: higher than England and Wales’s and much higher than Italy’s 52.2%) can be attempted through the more stringent organizational circumstances that U.S. trial actors are faced with vis-à-vis their Italian counterparts. In fact, U.S. judges and prosecutors appear to be immersed in a procedural environment leading them to produce a higher conviction rate. Recalling the elements of comparative criminal procedure outlined above, let us specify such differences in detail:
U.S. prosecutors need to make their case in front of a grand jury which appears as a broader and substantive evaluation than the evaluation that is done in Italy by the ‘judge for preliminary hearings’;
U.S. judges and prosecutors abide a criminal procedure that is rather intimidating towards the defendant: the experience of being arrested and being handcuffed may make a person willing to get out of the procedure at all costs;53
U.S. judges and prosecutors have more discretion in negotiating with defendants who plead guilty, whereas England and Wales, and Italian judges and prosecutors have less flexibility to negotiate with a defendant;
U.S. and England and Wales prosecutors are evaluated in their performance. Italian prosecutors are not evaluated on their performance;
U.S. and England and Wales sequential planning practice may very well feed back into prosecutors’ decisions to press charges. The sequential planning of trials may make prosecutors conservative about their procedure: pursuing a smaller and safer number of cases, also making more explicit the discretion they may apply on plea deals. Under sequential planning, prosecutors must prepare for trial, which is a significant amount of work. Such circumstance creates an incentive to offer plea deals rather than having each case go to trial. Adversely, parallel planning of trials may lead to formulate the hypothesis that Italian judges and prosecutors do not need to prepare for trial all at once, as trial hearings will be diluted over time; lawyers, judges and prosecutors do not need to study each case in detail and upfront, weighing the ‘likelihood of conviction’54 in their decision-making as they will have time to work through the case and will see the end result of their action only in the long run;
finally, a prosecutor depending on whether they work for the executive branch, as is the case in the U.S. and (albeit at arm’s length) in England and Wales, may pursue only ‘safe’ cases; therefore, consistent with ex ante expectations, this results in higher conviction rates.
We are likely observing two sides of the same coin, the coin being the organizational circumstances. Such different organizational circumstances manifest under the form of high conviction rates in the case of the U.S. and low conviction rates in the case of Italy. In the U.S., organizational circumstances (legislation and procedure) lead to a high conviction rate with a high proportion of plea deals and possibly a not very high number of cases. Above, we provided comparative data about conviction rates and plea deals. Also, Rasmusen et al. suggest the number of U.S. cases might be not very high. In the Italian case, organizational circumstances lead to cases sent for trial so that more information is collected to arrive at a decision, as it is shown by the lower conviction rates. In both cases, our findings are in accord with the literature whereby lawyers, judges, and prosecutors pursue their own utility function alongside55 the mission of the public organization they are part of. We may have found the same phenomenon taking place in apparently opposite situations and directions.
In the U.S. prosecutors need to present the grand jury with substantive evidence rather than investing them with procedural matters whereas in the Italian case the preliminary hearings are mostly about procedural matters. So, the U.S. constitutional and legislative context is demanding on judges and prosecutors and – at the same time – it provides them with powerful instruments. Conversely, U.S. prosecutors act in a tighter environment whereby they have procedural dominance and are at the same time controlled by departmental hierarchy on their performance. Therefore, they tend to pursue cases they can win.
The Italian system, in contrast, leaves a lot of room for lawyers, judges, and prosecutors to define their own ways to proceed. In Italy, prosecutors are more independent and are less supervised. Data provided here – on low conviction rates – suggests that public prosecutors send cases to court to check and probe the evidence available prior to trial. Their decisions are not discriminating. It should be noted that the correspondingly high rate of acquittals (47.8 percent) demonstrates independence of judges vis-à-vis prosecutors, an element of quality in the organization of justice and it constitutes a further positive characteristic of the adversarial system taking place in Italy as well, besides the U.S. and England and Wales. In fact, a low conviction rate shows that we cannot prove judges are dependent on the prosecutors; If they were dependent on the prosecutors, they would convict at a higher rate than 50% approximately. Low conviction rate is a sufficient, but not necessary, condition for evaluating independence of judges from the prosecutors.
In England and Wales, prosecutors seem to occupy a middle ground between the U.S. and Italy. England and Wales prosecutors have rather high conviction and plea dealing rates notwithstanding that their discretionary power in relation to guilty pleas is rather limited, as it is in Italy. However, their courts adopt sequential planning of trials, which may encourage them to pursue only those cases in which they feel confident.
5. Impact and remedies
In practice, it appears we are confronted – in the U.S. case – with a more effective judicial system than the Italian one. At the same time, the impression is that some procedural and legislative measures make U.S. prosecutors a little too powerful vis-à-vis defendants. The England and Wales system seems more equitable. The U.S. system could likely benefit from loosening at least some of those procedural measures.
The Italian procedure appears to provide many theoretical guaranties for defendants, albeit it imposes enormous costs on the criminal justice system (for instance, congestion of courts56), and on defendants (for instance, financial and human cost, such as job loss and social shaming). Other short run incentives must also take into account additional stakeholders, like police investigators – who are the operational arm of the Executive Branch. They too are subject to short run incentives. For instance, they receive a ‘career praise’57 when they make an arrest or gain media visibility. The media, on the other hand, gain the public’s attention by presenting the investigators’ statements verbatim, in the indicative mode, as established facts rather than investigative hypotheses. Social implications of the Italian procedure consist of an implicit cost to society, due to the public perception of ineffectiveness and low deterrence effect of criminal justice, which in turn feeds back into defendants’ behavior, discouraging them from engaging in plea deal proceedings.
Considering impact, we need to elicit the notion of discretion. The current Italian parallel planning policy tends to mitigate the (implicit and necessary) use of discretion on the part of judges and prosecutors, whereas discretion is accepted in the U.S. and England and Wales systems.58 It must be underlined here that once we consider the practice, it becomes evident that there is a ‘creeping’ discretion that comes in also in systems that try to avoid discretion, as it is the case in the Italian parallel planning policy. Once again, such discretion need not be conscious, however it is implicit in the piecemeal decisions that are made over the long run in the Italian system of parallel planning of hearings.
About possible remedies, in Italy, sequential planning of trials would probably be a key change to make the system more effective.59 More remedies could include keeping explicit statistics on conviction rates. Judicial reform in Italy states that the judge for the preliminary hearings should send to court only in the case of a ‘reasonable prognosis of conviction’.60 However, simply making the statement, without any specific measuring action referring to the ‘reasonable prognosis’ criterion, does not appear to change the incentives for actors in the process; therefore, it will probably not change the outcome very much.61
6. Research limitations and future studies
Turning to limitations of the present study, we need to be aware that this was only a first attempt – a preliminary investigation – into cross country statistics on conviction rates, trying to overcome preliminary issues of comparability. Future research would benefit from a number of aspects that could be further investigated. One of the developments could pursue a deeper analysis of the U.S. judge and prosecutor’s leeway in plea deals, in order to ascertain the limits of deal making in the U.S. Also nuances of the U.S. procedure should be examined, for instance, knowledge on the part of the defense lawyers of all information available to the prosecutor at the time of deciding about plea.
Another aspect that could broaden the present analysis is expanding the study to more European countries, with their likely diverse landscape of practice: from sequential planning in the Northwest of the Continent to parallel planning in the Southeast. Such an analysis could test the hypothesis of a positive relationship between low conviction rates and parallel planning of trials. Alternatively, research can consider – from a global perspective - extending the analysis to other jurisdictions such as China’s high and India’s low conviction rates.62 Such further studies could also take stock of the hypothesis developed in this article and delve deeper into judicial statistics, adding for example indicators such as caseload per professional.
7. Conclusion
The quest for a ‘just’ conviction rate is rather elusive, and the answer likely resides in a continuous search. Nonetheless, such comparative work is rewarding as we have come to identify segments of the procedure that appear to be neglected at least from an analytical point of view: on the one hand the U.S. judge and prosecutor’s leeway in negotiating guilty pleas and, on the other hand, the Italian operational parallel planning of trials vis-à-vis sequential planning in the U.S. and England and Wales case.
This article has dealt with very different conviction rates in selected countries, and it has made a first attempt at establishing comparability of such different data across different legal systems. We started a quantitative path of cross-countries comparison between different legal systems, with specific reference to conviction rates, where only qualitative cross-country comparisons had been previously made (Caianiello et al.). The conclusion may sound somewhat like a jury verdict: it is ‘not unreasonable’ to make such comparison. Though the causes of the differences are complex, different planning policies appear to be a plausible causal element within such complexity. The explanation is as follows: parallel planning policy diffuses responsibility among the stakeholders and over time; so, the trial becomes an extended investigation.
We have also proposed new terminology for comparative procedures where terminology seemed to be lacking: the notion of sequential vs. parallel hearings planning policy was introduced.
We have shown facts, norms and principles. It will be for future work to decide the extent to which the three legal systems illustrated above implement the shared principles of presumption of innocence (with its implications of reasonable cause and reasonable probability of conviction), reasonable time and immediacy, orality and concentration. In non-technical terms, it will be for future readers to decide the extent to which the systems studied here reflect that pietas – respect, mercy, compassion – of ‘human animals’ for one another that poet Ugo Foscolo praised in tribunals and justice in his 1806 poem on Sepulchres63:
Marriage, tribunals and altars
Gave to the human animal pietas
Both for himself and others
Notes
[1] Council of Europe, “Special File: CEPEJ,” CEPEJ, accessed April 11, 2026, https://www.coe.int/en/web/cepej/special-file.
[2] European Commission, EU Justice Scoreboard 2021: EU Justice Scoreboard Report for the European Semester (2021), 3. (“In the context of the European Semester, the Council, on the Commission’s proposal, addressed country-specific recommendations relating to their justice system to seven Member States in 2019 (HR, IT, CY, HU, MT, PT and SK) and eight Member States in 2020 (HR, IT, CY, HU, MT, PL, PT and SK).”) 3 note 16. Also: European Commission for the Efficiency of Justice (CEPEJ), European Judicial Systems: Efficiency and Quality of Justice, CEPEJ Studies No. 23 (2016), 231. ‘in all these jurisdictions (with the exception of Italy) the Disposition Time can be considered acceptable’ (emphasis added).
[3] Italy, Law 27 September 2021, no. 134.
[4] Pietro Curzio, Relazione sull’amministrazione della giustizia nell’anno 2022, Corte Suprema di Cassazione (January 2023), 60.
[5] UK Office for National Statistics, “Population of England and Wales,” accessed November 6, 2023, https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates.
Population of England and Wales: ‘On Census Day, 21 March 2021, the size of the usual resident population in England and Wales was 59,597,542 (56,490,048 in England and 3,107,494 in Wales); this was the largest population ever recorded through a census in England and Wales’ Population of Italy: ‘Italy’s total population was 58.96 million in January 2023’ in DataReportal, “Italy Population Data,” accessed December 1, 2023, https://datareportal.com/reports/digital-2023-italy.Accessed 1st December 2023.
[6] U.S. Administrative Office of the U.S. Courts, “Federal Judicial Caseload Statistics 2022,” Table D-4, accessed November 6, 2023, https://www.uscourts.gov/judicial-caseload-indicators-federal-judicial-caseload-statistics-2022.U.S. District Courts–Criminal Defendants Disposed of, by Type of Disposition and Offense, During the 12-Month Period Ending March 31, 2022, author’s percentage calculations https://www.uscourts.gov/judicial-caseload-indicators-federal-judicial-caseload-statistics-2022 accessed 6 November 2023.
[8] PBS, Research the System (2002), https://www.pbs.org/kqed/presumedguilty/3.2.1.html accessed January 31, 2020.
[9] D. Clark, “Conviction Rates for Magistrates’ Courts and Crown Court in England and Wales,” Statista, October 11, 2023, https://www.statista.com/statistics/1100769/conviction-rate-in-england-and-wales/.
https://www.statista.com/statistics/1100769/conviction-rate-in-england-and-wales/#:~:text=The%20conviction%20rate%20in%20England,the%20third%20quarter%20of%202021 Accessed 6 November 2023.
[10] Pietro Curzio, Relazione sull’amministrazione della giustizia nell’anno 2022, Corte Suprema di Cassazione (January 2023), 60.
[Report on the administration of justice in the year 2022, Supreme Court of Cassation] author’s calculation.
[12] UK Government, “Criminal Court Statistics Quarterly: April to June 2023,” published September 28, 2023, https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-april-to-june-2023.
https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-april-to-june-2023/criminal-court-statistics-quarterly-april-to-june-2023 accessed 6 november 2023.
[13] UK House of Commons Library, Georgina Sturge, “Court Statistics for England and Wales,” CBP-8372 (31 January 2023), 8.
[14] Pietro Curzio, Relazione sull’amministrazione della giustizia nell’anno 2022, Corte Suprema di Cassazione, (Jan. 2023) [Report on the administration of justice in the year 2022, Supreme Court of Cassation] 60.
[15] Council of Europe, “Special File: CEPEJ,” CEPEJ, accessed April 11, 2026, https://www.coe.int/en/web/cepej/special-file accessed July 1, 2025.
[16] Andrew Birrell et al., The Trial Penalty: The Sixth Amendment Right to Trial on the Verge of Extinction and How to Save It (National Association of Criminal Defense Lawyers, 2018). www.nacdl.org/trialpenaltyreport accessed 6 November 2023, this work emphasizes a racial discriminatory element in U.S. criminal justice;
Rod Morgan, Summary Justice: Fast, but Fair? (Centre for Crime and Justice Studies, King’s College London, 2008), www.crimeandjustice.org.uk accessed 6 November 2023.
[17] Eric Rasmusen, Manu Raghav, and J. Mark Ramseyer, “Convictions versus Conviction Rates: The Prosecutor’s Choice,” American Law and Economics Review 11 (2009): 47–70. https://hls.harvard.edu/bibliography/convictions-versus-conviction-rates-the-prosecutors-choice/ accessed 6 November 2023.
[19] F. M. Artinger, S. Artinger, and G. Gigerenzer, “C.Y.A.: Frequency and Causes of Defensive Decisions in Public Administration,” Business Research 12 (2019): 9–25, https://doi.org/10.1007/s40685-018-0074-2. The author is grateful to professor Elisabetta Morlino for pointing out literature on defensive administration.
[21] Dui Hua Human Rights Journal, “China’s Acquittal Rate Lowest in Two Decades,” September 12, 2023. states that ‘The conviction rate in 2022 was 99.95% according to statistics in the China Law Yearbook.’ Whereas India Court Conviction Rate data was reported as 54.20% in 2022. https://www.ceicdata.com/en/india/crime-statistics/court-conviction-rate#:~:text=India%20Court%20Conviction%20Rate%20data,to%202022%2C%20with%2023%20observations accessed 2 January 2025.
[22] Italy, Law 27 September 2021, no. 134.
‘Reasonable likelihood of conviction’ This law was dubbed as the ‘Cartabia Reform.’ It was put in place by the Draghi Cabinet Minister of Justice, Marta Cartabia.
[23] Lexology, “Principles of Immediacy and Evidence Evaluation,” accessed August 1, 2025. https://www.lexology.com/library/detail.aspx?g=ac1ba85f-5e7f-4013-9e38-bdc912cd7572#:~:text=According%20to%20the%20principles%20of%20immediacy%20and,by%20the%20parties%20at%20the%20main%20hearing.&text=It%20is%20generally%20considered%20that%20the%20principles,terms%20of%20assessing%20the%20value%20of%20evidence.
[24] Michele Caianiello and Jacqueline Hodgson, eds., Discretionary Criminal Justice in a Comparative Context (Durham, NC: Carolina Academic Press, 2015). Covers prosecutorial discretion, plea agreements, and exclusionary rules in the People’s Republic of China, Italy, Spain, and Switzerland. Table of contents available at https://cap-press.com/pdf/9781611637731.pdf accessed 9 November 2023.
[25] U.S. Office of the United States Attorneys, Steps in the Federal Criminal Procedure, U.S. Department of Justice (2022), https://www.justice.gov/usao/justice-101/steps-federal-criminal-process Accessed 30 November 2023.
[26] Italian Code of Criminal Procedure, art. 407.
[29] UK Government, Criminal Procedure Rules and Practice Directions 2020 (2023), https://www.gov.uk/guidance/rules-and-practice-directions-2020.
https://www.gov.uk/guidance/rules-and-practice-directions-2020#criminal-court-procedure.
‘Crown Courts deal with more serious offences and require a jury, while Magistrates Courts deal with less serious offences, and require no jury.’ D. Clark, “Conviction Rates for Magistrates’ Courts and Crown Court in England and Wales,” Statista, October 11, 2023, https://www.statista.com/statistics/1100769/conviction-rate-in-england-and-wales/. https://www.statista.com/statistics/1100769/conviction-rate-in-england-and-wales/#:~:text=The%20conviction%20rate%20in%20England,the%20third%20quarter%20of%202021 Accessed 6 November 2023.
[31] University of the West of England The Bar Training Course 2024–205, ‘Criminal Litigation & Sentencing An Overview of the Procedure in the Criminal Courts’, 2-3. For a historical perspective on the UK system, see generally U.S. Department of Justice Office of Justice Programs, “Criminal Prosecutions in England and Wales,” Justice System Journal 3, no. 1 (1977): 38–49. https://www.ojp.gov/ncjrs/virtual-library/abstracts/criminal-prosecutions-england-and-wales.
[32] Italian Code of Criminal Procedure, art. 274.
[33] In Italian: Giudice per le indagini preliminari – GIP. Of the 262,766 terminated cases of 2022, only 24,294 had seen their defendants arrested or 9% (Curzio, 2023, 60).
[34] The process may not always be as smooth as it sounds: Emilio C. Viano, “Plea Bargaining in the United States: A Perversion of Justice,” Revue internationale de droit pénal 83 (2012): 109–145. https://www.cairn.info/revue-internationale-de-droit-penal-2012-1-page-109.htm#:~:text=About%2097%20percent%20of%20the,summary%20proceeding%20before%20a%20judge.
[35] Incentives to plea bargaining need to be taken into account, including what one could consider ‘perverse’ effects, as reported by Daniel Boffey, “Rise of Plea-Bargaining Coerces Young Defendants into Guilty Pleas,” The Guardian, October 6, 2022. https://www.theguardian.com/law/2022/oct/06/rise-of-plea-bargaining-coerces-young-defendants-into-guilty-pleas-says-report accessed 14 November 2023 citing fairtrials.org report ‘Young minds big decisions’, October 2022.
See also UK Sentencing Council, “Going to Court: Information for Defendants,” accessed November 7, 2023. https://www.sentencingcouncil.org.uk/going-to-court/information-for-defendants/. Daniele Alge, “Plea Bargaining in England and Wales: Some Comparisons with the USA,” paper presented 2014. In: Academy of Criminal Justice Sciences Annual Meeting, 18–22 Feb 2014, Philadelphia, USA. (Unpublished) https://repository.uwl.ac.uk/id/eprint/2365/. Also, Daniele Alge, “Pressures to Plead Guilty or Playing the System? An Exploration of the Causes of Cracked Trials,” PhD thesis, University of Manchester, 2009.
[36] Italian Code of Criminal Procedure, art. 444. In Italian: patteggiamento.
[38] UK Office for National Statistics – Criminal court statistics quarterly: April to June 2023 – Published 28 September 2023. https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-april-to-june-2023/criminal-court-statistics-quarterly-april-to-june-2023 accessed 6 November 2023.
[39] UK House of Commons Library, Georgina Sturge, ‘Court statistics for England and Wales’ 31 Jan 2023, CBP-8372, 8.
[40] Pietro Curzio, Relazione sull’amministrazione della giustizia nell’anno 2022, Corte Suprema di Cassazione, (Jan. 2023) [Report on the administration of justice in the year 2022, Supreme Court of Cassation] 60.
[41] We call ‘silent’ this element of the procedure because it is hardly found in the literature. For instance, the CEPEJ does not collect information on such element of procedure. ‘European judicial systems: Efficiency and quality of justice’, CEPEJ studies No. 23, edition 2016 (2014 data) section 5.4 pp 217–237.
[42] In Italy, it is not uncommon that a trial last years. Whereas in the U.S. and England and Wales the length of one trial is likely to be measured in weeks or months since its inception. Comparative data are available for Disposition Time, which is a ratio of measures of the number of cases that in the denominator includes dismissed cases and it is therefore optimistic, because it is dampened by the large number of dismissed cases that have very short length (in Italy there is about one million cases per year of which about 600.000 are dismissed (Curzio, 2023)). Disposition Time in 2014 was 82 days in England and Wales and 386 days in Italy (CEPEJ, European Judicial Systems: Efficiency and Quality of Justice (2016), 230.
In the last 2024 CEPEJ report, Italy lowered its Disposition Time, even though it remains high (lower by 143 days, since the previous cycle. Looking at data out of context, and with little experience of judicial statistics, i.e. a general media point of view, Disposition Time of 386 (or 243) days may appear reasonable, as it is about one year. However, comparison of data is much more telling: Disposition Time in Italy was five (three) times what it was in England and Wales, a staggering difference. Let us also underline here that ‘parallel planning’ of trials is only a matter of praxis in the Italian case. In fact, as in the other countries, other Italian law mandates ‘serial’ planning : the Italian code of criminal procedure, art. 477, ‘duration of debate’, par. 1, says: ‘When it is absolutely impossible to finish the debate in only one hearing, the president [of the court] sets that the debate is resumed the following non holiday day’. In fact, there have been trials that have been held through sequential planning of hearings. The ‘Mondo di mezzo’ trial held 240 hearings in one and half years. Web News ‘Mafia capitale’ [Mafia in the capitol city] 27 June 2017 accessed 3 November 2023. https://www.interris.it/roma-caput-mundi/mafia-capitale-il-legale-di-carminati-quello-del-mondo-di-mezzo-resta-un-processetto/. It is also worth mentioning that parallel planning makes each trial longer over time and thus subject to incidents such as, for instance, changes in the composition of the three-judge bench, which carries delaying implications. Also, parallel planning makes the length of the trial a central variable towards statute of limitations. Italy had a statute of limitations rate of 17.2 % with 45,143 cases (Curzio 2023, 57) over a total 262,766 cases (Curzio, 2023, 60).
[43] UNODC, “Public Prosecutors as the Gatekeepers of Criminal Justice,” 2020, https://www.unodc.org, https://www.unodc.org/e4j/en/crime-prevention-criminal-justice/module-14/key-issues/2--general-issues--public-prosecutors-as-the-gate-keepers-of-criminal-justice.html.
[45] U.S. Government Accountability Office, GAO-04-422 (2004), 25.May 2004 U.S. Attorneys Performance-Based Initiatives Are Evolving GAO-04-422 p. 35 of document or p. 41 of 83-page pdf.
[46] UK Government Crown Prosecution Service, “Key Measures,” Q1-Q3-2018-19-1. https://www.cps.gov.uk/publication/key-measures.
UK Government Crown Prosecution Service, “Data Summary Quarter 1 2023–2024.” https://www.cps.gov.uk/publication/cps-data-summary-quarter-1-2023-2024.
[47] It may be worth mentioning that when Minister of Justice, Marta Cartabia, in 2022, proposed an evaluation system for the Italian magistrates, the trade union of the magistrates objected that ‘it could generate anxiety.’ See Valentina Stella, “Pagelle ai magistrati,” Il Dubbio, March 2, 2022., (translation: Scoreboards for magistrates, the magistrates’ association disagrees: “Scoreboards would make us anxious”’), https://www.ildubbio.news/2022/03/01/pagelle-ai-magistrati-lanm-non-ci-sta-mettono-troppa-ansia/.
[50] P.B.S. Research the System, P.B.S. (2002) https://www.pbs.org/kqed/presumedguilty/3.2.1.html accessed 2020.
[51] D. Clark, Statista.com ‘Conviction rates for Magistrates Courts and Crown Court in England and Wales from 2nd quarter 2013 to 1st quarter 2023’ Oct 11, 2023. https://www.statista.com/statistics/1100769/conviction-rate-in-england-and-wales/#:~:text=The%20conviction%20rate%20in%20England,the%20third%20quarter%20of%202021 Accessed 6 November 2023.
[52] Pietro Curzio, Relazione sull’amministrazione della giustizia nell’anno 2022, Corte Suprema di Cassazione, (Jan. 2023) [Report on the administration of justice in the year 2022, Supreme Court of Cassation] 60, author’s calculation.
[53] From a theoretical point of view, taking into account the high generalized cost of a trial to a defendant (cash, emotional, social, and time per se), there is always a non-zero cost that a rational and innocent defendant would be willing to pay to get out of prosecution through a guilty plea and deal. The matter then appears only one of quantitative calculation: whether the deal that is offered by the prosecutor is lower enough compared to the expected generalized cost of a trial.
[54] ‘Reasonable likelihood of conviction’ was only introduced in Italy by law 27 Sept 2021, n. 134 art. 1 par. 9 a), m) par 12, d). This law was dubbed as the ‘Cartabia Reform.’ It was put in place by the Draghi Cabinet Minister of Justice, Marta Cartabia.
[55] Rasmusen et al., supra note 17, underline: ‘a prosecutor who values not just the number of convictions but the conviction rate and unrelated personal goals’ (emphasis added). It needs being remarked – once again – that the literature does not imply at all that pursuit of ‘unrelated personal goals’ consists of criminal activities. The goals the literature referred to might be the study of law or any other perfectly legal activity.
[56] See generally, Stephen Holmes and Cass R. Sunstein The Cost of Rights: Why Liberty Depends on Taxes (2000). Holmes and Sunstein argue all rights are positive – according to the Isaiah Berlin definition – because their actionability requires resources, including Isaiah Berlin’s negative rights. In the same vein, this article also argues in favour of an appreciation of implementation and time transients, i.e. we should also take into account costs and benefits of what happens over time while the rule of law makes its way to its steady state. That is what we call ‘a culture of implementation’.
[59] It would be objected that in Italy criminal action by prosecutors is constitutionally mandated. However, such mandate does not imply a mandate to send to court each case. And in practice such mandate is ‘managed’ over time through statute of limitations and implicit discretion. Compulsory criminal action is different from compulsory prosecution. See generally Joachim Herrmann ‘The Rule of Compulsory Prosecution and the Scope of Prosecutorial Discretion in Germany’ The University of Chicago Law Review Vol. 41, No. 3 (Spring, 1974), 468–505. Kenneth Culp Davis ‘Discretionary Justice’ Journal of Legal Education Vol. 23, No. 1 (1970), 56–62. About criminal action, it has to be noted here that dismissal of a case requires the prosecutor to write a verdict, subject to evaluation and appeal, whereas sending the case to court does not require such a formalized procedure. This appear to be a short run incentive on the prosecutor to send cases to court.
[60] In Italian ragionevole previsione di condanna: Italian law 27 Sept 2021, n. 134 art. 1 par. 9 a), m) par 12, d).
[61] ‘Covenants without the sword are but words’, Thomas Hobbes, Leviathan (1651), ch. 17.
[62] Dui Hua Human Rights Journal, “China’s Acquittal Rate Lowest in Two Decades,” September 12, 2023.
[63] Ugo Foscolo, Sepulchres (trans. J. G. Nichols, 2015), vv. 91–93.
Acknowledgements
The author is grateful to the anonymous reviewers and the editors of the IJCA for their attention and time.
Author Information
He was a publisher of guidebooks to public administration and authored Unknown Values and Stakeholders: The Pro-Business Outcome and the Role of Competition, Palgrave 2017 2nd ed.