Journal Issue

Italy: Selected Issues

International Monetary Fund. European Dept.
Published Date:
September 2014
  • ShareShare
Show Summary Details


Judicial Reforms for Growth

The chapter investigates options for improving the efficiency of the Italian judicial system and closing the regional performance gap. Better courts would bring about macroeconomic benefits, including increased employment opportunities, and higher productivity, investment, and R&D. Reforms should focus on court management, rationalization of the appeal system, reduction of the backlog of pending cases, and wider use of out-of-court mediation.

Future Challenges Facing Italy’s Financial Sector

The Italian financial system faces a number of challenges in order to restore profitability under weak growth conditions and to adapt to a changing global environment. This chapter explores ways of improving profitability and the challenges of shifting from a bank-based financial system, common in EU countries, to a more “market-based” (“arm’s length”) system. Along with this shift comes a diversification of financing sources, led by further development of capital markets.

Improving Public Spending Allocation and Performance in Italy: An Efficiency Analysis

Budget allocation in Italy will need to increasingly rely on efficiency analysis to find savings and improve performance. The analysis in this chapter finds that large social spending in Italy, particularly current pensions, will need to be tackled to generate sizable expenditure savings. In education and non-pension social protection there is scope for improving outcomes with current resources. In other areas, reducing cross-regional variation in spending efficiency could also lead to savings.

The Use of Performance Information in Resource Allocation

The role of performance information in budget management is growing in advanced economies. In Italy, performance information will be critical for making the Spending Review a permanent part of the budget process. Italy generates significant performance information and more can be done to actively use this information in budget decision making. High quality performance information needs to be used more proactively to evaluate public policy and inform resource allocation decisions

Judicial Reforms for Growth1

There is ample room to improve the efficiency of the Italian judicial system and close the regional performance gap. Better court performance would bring about macroeconomic benefits, including increased employment opportunities, and higher productivity, investment, and R&D. Reforms should focus on court management, reduction of the backlog of pending cases, wider use of out-of-court mediation, and rationalization of the appeal system.

A. The Macro-Judicial Linkages—A Regional Perspective

1. The Italian judicial system performs significantly worse than the average OECD country in terms of court times and case backlog. As discussed in Esposito, Lanau, and Pompe (2013), it takes an average of 1,185 days to enforce a contract in Italy according to the Doing Business survey (more than twice the OECD average). Italy also has the highest number of violations of the reasonable time requirement enshrined in Article 6 of the European Convention of Human Rights. The backlog of pending civil cases has fallen by about 11 percent since 2009 but still stood at 5.2 million cases in mid 2013.

Days to Enforce a Contract

Source: Doing Business, 2014

2. Court performance differs across regions, with the north generally faring better than the south. Figure 1 maps regional court efficiency under the assumption that the average duration of civil proceedings and the per capita stock of pending civil cases are reasonable proxies for judicial efficiency. These statistics abstract from factors such as the predictability of court decisions but do proxy for the economic opportunity cost of drawn-out trials. The indicators show that the gap between the north and the south is as wide as two years for the average length of civil proceeding and four times the number of per capita backlog cases. A similar picture emerges if the proxy is the regional public sector efficiency measure in Giordano and Tommasino (2011), which relates the number of judges per 1,000 new trials to the average length of trials. Performance heterogeneity across judicial districts within regions also exists but it is not explored in this paper. Since the procedural rules and the structure of the judicial system are set at the national level and are the same throughout the country, factors such as managerial practices in each court, individual judge effort, degree of specialization, court size, productivity of court clerks, availability and use of IT tools, and degree of regional informality may explain the large interregional differences in court performance.

Figure 1.Average Length of Civil Proceedings

(in days, 2006–12)

3. Regional macro outcomes such as productivity and unemployment are associated with the performance of the justice system (Figure 3). A statistically significant correlation exists between improved judicial efficiency (proxied by the duration of court proceedings) and better regional macro outcomes. The correlations span a number of important areas for potential growth and highlight a north-south gap in most performance indicators:

  • GDP per capita and productivity. Productivity in industry is 44 percent higher in the three regions with the most efficient courts than in the bottom three. The correlation between GDP per capita and judicial efficiency is also present at the provincial level (Table 1). For a given initial GDP per capita, growth in the province where the judicial system is the most efficient is 1.9 percentage points higher than in the province with the most inefficient judicial system.

  • Firm size and R&D. Italian firms are small from a cross-country perspective, and are even smaller in the regions where judicial efficiency is low (30 large firms per thousand, compared to more than 70 in regions where courts are more efficient). Private R&D expenditure is ½ percent of GDP higher in the three regions with the most efficient courts than in the bottom three.

  • Employment. The average duration of labor court proceedings is significantly correlated with inactivity and unemployment rates. The unemployment rate is 10 percentage points higher in the three regions with the most inefficient courts than in the three best.

Figure 2.Backlog of Pending Civil Cases

(per thousand inhabitants, 2012)

Figure 3.Italy: Regional Judicial Efficiency and Macro Outcomes

Source: Istat, Ministry of Justice

Table 1.Provincial Growth and Judicial Efficiency 1/
Dependent variable: Provincial GDP/capita growth 2006-11
Judicial efficiency0.020**
Initial GDP/capita (log)−0.018***

Judicial efficiency is the 2006 civil justice public sector efficiency score in Giordano and Tommasino (2011). ** statistically significant at the 5 percent level. *** statistically significant at the 1 percent level.

Judicial efficiency is the 2006 civil justice public sector efficiency score in Giordano and Tommasino (2011). ** statistically significant at the 5 percent level. *** statistically significant at the 1 percent level.

4. These regional patterns are consistent with the findings of a growing literature on the macro impact of judicial efficiency (see Esposito et al. 2013 for a full literature survey). Regional factors beyond judicial efficiency may also correlate with macro outcomes, opening up the possibility that some of the relationships in Figure 4 are not causal. That said, the literature has shown that judicial efficiency does cause certain macro outcomes. For instance, Giacomelli and Menon (2012) have established a causal link between judicial efficiency and firm size exploiting performance differences across Italian judicial districts. Moreover, business surveys suggest that the justice system is indeed an important determinant of the business environment and macro outcomes.

Figure 4.Average Duration of Ordinary Labor Court Proceedings

(in days, 2006–12)

B. A Deeper Look at Court Efficiency and Employment

5. Labor courts are no exception to the regional heterogeneity in court performance. It takes less than 300 days to resolve a labor dispute in Piemonte or Trentino, but more than a thousand in Puglia or Sicilia (Figure 4). Since labor laws are set at the national level, local court management and individual judge effort likely play a role in explaining the differences. Factors such as the complexity of cases in certain regions may also be relevant.

6. Long labor court proceedings increase the expected cost of dismissals and may result in less job creation. Inefficiencies in other parts of the judicial system could have an indirect impact on employment to the extent that they result in reduced firm size, limited access to credit, or reduced incentives to invest. Using Italian data, Gianfreda and Vallanti (2013a, b) have linked labor court efficiency to the composition of employment, labor market participation, and job relocation rates and find that difference between best and worst courts in Italy results in a 6 percentage point gap in relocation rates.

7. Micro data allow for a closer examination of the correlation between court efficiency and employment outcomes. Table 2 provides descriptive statistics of the dataset grouped by macro-region.

  • A probit model is estimated for the probability of an individual being employed, based on a cross-section of 12,430 individuals in 19 Italian regions from the 2012 Survey on Household Income and Wealth by the Bank of Italy.2 An individual is considered to be employed if they worked full-time for at least 7 months in 2012.

  • The key independent variable is a measure of regional judicial efficiency. In a first set of regressions, the log average duration of court proceedings for the period 2005–12 is the proxy for judicial efficiency. Three versions of the variable are used (in separate regressions): duration of ordinary civil proceedings, duration of ordinary labor court proceedings, and combined duration of ordinary and appeal labor court proceedings.3 In an alternative specification, the log stock of pending civil cases per thousand inhabitants is the judicial efficiency proxy.

  • The control variables include individual characteristics such as gender, education, and age, and two regional macroeconomic variables that likely affect an individual’s employment prospects: GDP growth and employment growth. Other variables such as GDP per capita, the degree of labor market informality, and the composition of output display significant variation across regions and could affect the labor market. Unfortunately, these variables are highly collinear with judicial efficiency measures and cannot be included in the regressions. The implications of this limitation are discussed in more detail below.

Table 2.Summary Statistics(percent of sample)
College education14.314.810.7

8. Halving the duration of civil proceedings would be associated with a 7.7 percentage point increase in the probability of individual employment. The regression results show a statistically significant correlation between employment probabilities and the four measures of judicial efficiency (Table 3):

  • The marginal effects of a one percent reduction in the duration of court proceedings on the probability of employment range from 6 to 13 percent depending on the type of court considered. The magnitude of the correlation is the largest for civil courts, suggesting that the indirect effects of court efficiency on employment discussed in ¶6 may be important. A one percent reduction in the backlog of pending cases is associated with a 7 percentage point increase in employment probabilities. It is worth emphasizing that since the judicial efficiency measures are very collinear with other factors such as GDP per capita and labor market informality, there is a chance that the regression coefficients capture some phenomena not strictly linked to the judicial system.

  • The regressions can be used out of sample to trace the change in employment probabilities from halving the duration of court proceedings holding other control variables constant. In the case of labor courts, halving the duration of proceedings is associated with an average 3.8 percentage point increase in the probability of employment. The equivalent figure for civil proceedings is 7.7 percentage points. There is some regional variation in the estimated probabilities but it is relatively moderate and does not follow a clear north-south pattern (for example, the increase is employment probabilities in Abruzzo is 1.2 percentage points higher than in Campania).

  • The two regional macroeconomic controls are generally insignificant. The individual controls have the expected signs and are highly significant in line with the literature. For instance, all else equal, a female is 6 percent less likely to be employed than a male.

Table 3.Probit Models for Probability of Employment
Dependent variable = 1 if person is employed, zero otherwise
Marginal effects
Duration civil proceedings−0.13***
Duration labor court proceedings−0.07***
Duration combined labor proceedings−0.06***
Pending cases / capita−0.07***
Head of household0.41***0.41***0.41***0.41***
Log(family income)0.02***0.02***0.02***0.02***
High school education0.10***0.10***0.10***0.10***
College education0.14***0.14***0.14***0.14***
Postgraduate education0.10*0.10*0.10*0.10**
Regional employment growth0.970.910.851.53*
Regional GDP growth−0.53−1.26−0.64−1.14
* statistically significant at the 10 percent level; ** statistically significant at the 5 percent level; *** statistically significant at the 1 percent level. Young equals one if the person is 16–24 years old. Old equals one if the person is 60 or older. Head of household and female are self-explanatory dummies. Family income is the income of other members of the household. High school, college, and postgrad are education dummies, with elementary schooling or less being the omitted category. Marginal effects for dummy variables are for a discrete change from 0 to 1. Regional employment growth and regional GDP growth are the change in log of total employment and log GDP in the region where the person lives. Robust standard errors clustered by region.
* statistically significant at the 10 percent level; ** statistically significant at the 5 percent level; *** statistically significant at the 1 percent level. Young equals one if the person is 16–24 years old. Old equals one if the person is 60 or older. Head of household and female are self-explanatory dummies. Family income is the income of other members of the household. High school, college, and postgrad are education dummies, with elementary schooling or less being the omitted category. Marginal effects for dummy variables are for a discrete change from 0 to 1. Regional employment growth and regional GDP growth are the change in log of total employment and log GDP in the region where the person lives. Robust standard errors clustered by region.

9. A causal interpretation of the results is not straightforward. As discussed above, the correlation between judicial efficiency and other slow-moving regional factors is very high and the possibility that these factors cause both employment outcomes and court efficiency cannot be ruled out. Moreover the 25 percent R2 indicates a risk of omitting in the regression other relevant factors to explain employment outcomes. Another econometric issue is relevant when considering a causal interpretation of the results: the most capable individuals, and thus more likely to be employed, may migrate to the regions where institutions and courts work better. Under these circumstances, the probit could overestimate the relationship between labor courts and employment outcomes. Finally, there is the possibility that the length of labor court proceedings could be endogenous since courts may be busier and slower at times when employment is falling sharply. The potential endogeneity problem should be attenuated by the fact that the duration variable is averaged over 2005–12.

C. Judicial Reforms for Growth

10. Judicial reform could help increase growth, create jobs, and improve the business environment. The research discussed above and related findings in the literature suggest that increased judicial efficiency could help lift potential growth by increasing productivity and R&D investment, and reducing unemployment.

11. Improving the efficiency of the Italian justice system has been a central concern for the governments of the past few years… (see Esposito et al. (2013) for a full survey of the measures adopted). The measures taken have yielded some results, including an 11 percent reduction in the backlog of pending civil cases and an average reduction in disposition times of 3.7 percent. However, with over 5 million civil cases still pending and more than a thousand days to enforce a contract, faster progress is needed. A comprehensive reform and a fundamental simplification of civil justice are needed to restore judicial efficiency.

Pending Civil Cases


Reduction in Disposal Time, 2011-13


….and the Renzi government is no exception. On June 30, 2014, PM Renzi announced a series of measures aiming at (i) reducing the length of civil trials (one year in first instance); (ii) reducing the backlog of civil cases by a half; (iii) creating “fast-track procedures” for company and family law cases; and (iv) reforming judges’ careers progression and their civil liability.4 At this time, the specific measures to deliver on these objectives and their timing have not been unveiled.

12. The online civil trial (“processo civile telematico”) introduced on June 30, 2014 is an important measure that could deliver efficiency gains.5

  • Under the new process, most judicial acts by court professionals (e.g., lodging of a lawsuit by a lawyer, issuance of an injunction order, or delivery of a judgment by judges) should gradually be carried out electronically. The data are collected on a web platform and available to judges, lawyers, and court clerks. The online civil trial is expected to produce economies of scale and reduce the length of civil trials. It is also accompanied by a 15 percent increase in the so-called unified court fees (“contributo unificato”).

  • While it is too early to assess the effectiveness of the online civil trial, it is certainly a welcomed development. IT solutions are not objectives in themselves, but tools which will need to be complemented with the measures (including those discussed below) to improve the performance of the justice system fundamentally. The increase in fees primarily offsets revenue losses and it thus remains to be seen whether it will play any role in deterring the inflow of (especially ill-founded) cases.

13. The 2013 Article IV identified options to improve the efficiency of the judicial system (Esposito et al. 2013). This paper further develops some of these options taking into account the policy developments since the last Article IV. The measures identified by Esposito et al. (2013) are: the development of a performance and accountability framework for courts; measures to limit review by the Supreme Court of Cassation6 (SCC henceforth); action to incentivize the use of mediation; and backlog reduction measures.

Court Performance and Accountability

14. Developing court performance indicators would help speed up court processes and increase accountability. These productivity indicators include the number of cases and the length of procedures amongst other variables. A key feature of any performance management framework is the setting of time management for judicial proceedings. The Council of Europe’s Commission for the Efficiency of Justice (CEPEJ, 2006) has developed best practices in this area which should be implemented. In a nutshell, these include: (i) setting realistic and measurable timeframes; (ii) enforcing the timeframe; (iii) monitoring and dissemination of data; (iv) procedural and case management policies and practices; and (v) caseload and workload policies. Some of best practices in this area are being followed in Italy and disseminating them across the country may help reduce the performance gap. For example, Box 1 describes a successful Italian scheme, the so-called “Strasbourg Program” of the Turin Court. Since the approach was successful in Turin, there is no objective reason why it cannot be institutionalized and made operational throughout the country.

15. Performance evaluation must respect judicial independence. The latter relates to judges’ decision-making and to the non-interference by the executive and the legislative branches, but not to the judges’ performance and accountability. Owing to the potentially complex relationship between judicial independence and accountability, the performance framework should ensure that any court evaluation system upholds judicial independence.

Review by the Supreme Court of Cassation

16. Measures to reduce the number of pending cases before the Supreme Court of Cassation and their processing time are needed. There are nearly 100,000 pending cases with an average time for resolution of 1,200 days (the respective figures in France are about 24,000 cases and 395 days; in Germany, there are about 8,000 cases pending before the Supreme Court). Moreover, a high number of lawyers can plead before the SCC (about 55,000 compared to 105 in France). This situation calls for reforms to make sure that the SCC, to use the words of its First President, Mr. Lupo, preserves its role of “jurisdictional guidance […] and of guarantee for the jus constitutionis, rather than the jus litigatoris.”7 Reforms could occur at two levels:

  • Role of lawyers. Relatively non-binding filters to regulate lawyers’ access to the SCC exist but their efficacy could be improved. Consideration could be given, as in other countries, to creating a special category of lawyers allowed to plead only before the SCC, thus increasing their specialization and reducing the number of lawyers with access to the SCC.

  • Screening. Similarly to other Supreme Courts in Europe, filters should be instituted to reduce the inflow of cases, including through summary dismissals and pre-selection (Esposito et al. 2013). This would allow the SCC to focus on those cases that are complex and of general interest, as opposed to simple cases of limited relevance (e.g., decisions by the Peace Court could be appealed to the Court of Appeal). In addition, the very limited types of cases that could be allowed before the SSC could be set by law, effectively establishing two levels of jurisdiction to review the substance of most cases (first instance and appeal).

Out-of-court mediation

17. The participation rate in the mediation process is low, although the success rate is high when the parties engage. Mediation should not be seen as a “pre-trial room” or a necessary routine step before going to the first instance court. It should instead be a device for parties to genuinely attempt to reach a “win-win” solution to their dispute. The measures discussed Esposito et al. (2013), such as allowing mediation without the compulsory presence of lawyers, would help strengthen mediation.

Backlog of pending cases

18. The authorities announced their intention to reduce the civil case backlog by half. This is a welcomed policy objective but detailed information about how and when such a target will be achieved is not yet available. Developing specific targeted measures to reduce the backlog is necessary to make future civil justice reforms more effective. Full implementation of the measures in the Decreto del Fare (e.g., auxiliary judges in appeal courts) would also help reduce the backlog.

19. Measures to reduce the backlog of pending cases need not be the same as those aimed at improving the general efficiency of the court system. Ad-hoc measures, including measures similar to those taken elsewhere in Europe, could help reduce the backlog of pending cases:

  • Creation of special staff units composed of so-called “judicial advisors” working under the supervision of a few judges to clear the backlog of civil cases (as in the Netherlands in the early 2000s);

  • Issuance of pilot judgments to address a large number of similar cases simultaneously;

  • Monitoring and labeling backlogged cases, as in the Court of Turin (Box 1);

  • Allowing parties in cases lodged, for instance, prior to 2012, to resort to a sort of “arbitration chamber” to resolve their dispute, in exchange of the elimination of all legal fees, tax deductibility of the costs incurred so far, and a commitment by the parties to not request compensation for excessive length of the procedure (under the so-called “Pinto Law”).

Labor court procedures

20. Since 2012, the Italian labor law has a new fast procedure for deciding upon dismissal disputes… (Law 92/2012, so-called “Fornero Law”). This is a specific, rapid summary procedure to ascertain the legitimacy (or otherwise) of a dismissal decision. Proposals for the reform of this new procedure, two years after its entry into force, have been voiced in some quarters.8 Without entering at this stage into the merit of these reform proposals, it would be important to assess the experience with this new procedure and to discuss possible enhancements, if needed.

21. ….but the interaction of this new procedure with the existing “ordinary” procedure should be clarified. Title IV of the Italian Code of Civil Procedure contains provisions on labor-related disputes in court. Within this title, Article 414 provides for the ordinary procedure governing the submission of labor-related disputes to courts. This provision remained after the introduction of the fast procedure in 2012. The relationship between fast and ordinary procedures is somewhat unclear. It could be clarified whether the former replaces the latter, is an alternative, or cumulative. Staff recommends the authorities clarify this issue when assessing the overall effectiveness of the new labor court procedure.

Box 1.The Strasbourg Program of the Turin Court

The “Strasbourg Program” is among the first experiences of successful case management in Italy. It resulted in a significant reduction of the backlog (about 27 percent in five years) and a substantial increase in speed of civil cases. Born in 2001 (as an idea of the then President of the First Instance Court of Turin, Mr. Barbuto), the program proved very successful and obtained the Crystal Scale of Justice Prize of the Council of Europe. Starting from the assumption that cases lasting more than three years would be against the “reasonable time” requirement of the European Convention on Human Rights (Article 6), the program entailed a number of sequenced steps:

  • Monitoring of the backlog: a regular six-month survey of the backlog was carried out with a view to categorizing cases by length and type.

  • Issuance of a “Decalogue”: the President of the Court issued a “Decalogue” aimed at ensuring a uniform practice in all civil sections of the Court, for the speedy and targeted treatment of cases. This “Decalogue” included recommendations for judges to: (i) make sure they play an active role in the proceeding; (ii) prevent and address dilatory tactics and avoid adjourning cases, if possible (if not, such an adjournment should not exceed 40-50 days and be granted only on the basis of specific reasons); (iii) promote a mediated solution to the dispute; (iv) make an effective use of the rules relating to evidence and witnesses; (v) ensure compliance with deadlines by experts; and (vi) strive for concise motivations of judgments.

  • Special tags: Cases were marked with special tags of different colors depending on whether they were pending before the court for (i) more than six months; (ii) between six months and two and a half years; and (iii) more than two and a half years, to allow priority treatment by judges.

Securing buy-in by stakeholders: the “Decalogue” was shared with the local Bar Association to ensure their necessary buy-in.


22. There is scope for significant improvements in the efficiency of the Italian judicial system, with potentially important macroeconomic effects. The reform should be structural, comprehensive and should have the necessary institutional support. It should also have the buy-in of all relevant stakeholders, notably judges and lawyers. The strategy should follow a four-pronged approach: (i) reducing the backlog; (ii) promoting wider use of alternative disputes resolution proceedings, such as mediation; (iii) rationalizing the appeal system, including review by the Supreme Court of Cassation and the role of lawyers in this context; and (iv) focusing on courts’ management and accountability. Such a comprehensive reform package, if taken together and effectively implemented, could help reduce unemployment and lift potential growth by increasing investment.


Prepared by Gianluca Esposito (LEG) and Sergi Lanau (EUR).

Only persons age 16-64 are considered. Results are similar for previous vintages of the survey and are not reported.

Combined duration is defined as duration of ordinary proceedings + appeal rate*duration of appeal proceedings.

Proposed measures in the criminal area include the criminalization of the false accounting offense and of selflaundering, as well as the reform of the limitation period.

See Law Decree 90/2014 for more details.

The Supreme Court of Cassation is the highest court in the Italian judicial system

See Opening of the Judicial Year (2014) by the First President of the Italian Supreme Court of Cassation, which is also the source for some of the statistics in this paragraph.

See, for instance, the reform proposal by the National Judges Association and the Labor Law Lawyers Association.

Other Resources Citing This Publication