1. Introduction

Throughout the twenty-first century, the rule of law has been struggling to assimilate and keep up with the dizzying and dynamic evolution of societies, states, businesses and citizens all over the world. Judicial systems seem to have been the victims of a major economic problem that had already been pointed out decades ago by such well-known intellectual figures as Barnard and Hayek: the inability to adapt quickly to changing circumstances (Barnard, 1971; Hayek, 1973, 1978, 1979).

In the recent past, the Portuguese judicial system was no exception to this rule. This was also due to the diagnosis made in 2011 by the European Commission, the European Central Bank and the International Monetary Fund. Signed in May of that year between Portugal and these three institutions, the Memorandum of Understanding on Specific Economic Policy (Portugal, 2011), aimed to induce a broad range of improvements in the Portuguese legal sector, namely by ensuring the execution of contracts in a timely and effective manner (a factor of noteworthy relevance for a healthy economy), and increased efficiency through reduction of delays (achieved through restructuring of the judicial system, new models of court management and streamlining and simplification of extrajudicial procedures). During the implementation phase of the assistance program, compliance with the measures contained in the MoU was judicious and complete. The positive results gradually began to become undeniable (Correia and Videira, 2015, 2016).

This paper expands the statistically objective analysis of the Troika experience in the Ministry of Justice of Portugal, presented by Correia and Videira (2015, 2016). The authors hope that the expanded analysis will continue to stimulate discussion of the academic and judicial system on these issues and continue to inspire and foster further theoretical and empirical research.

2. Framework and Objectives

Family and commercial over-indebtedness, low levels of savings, easy credit and low levels of economic growth, prosperity and productivity could easily characterize Portuguese society at the end of the first decade of the new millennium Correia and Videira (2016). This situation has led to high levels of non-compliance with financial obligations and, in turn, has led to an increase in disputes associated with debt collection, including pronounced increases in the number of civil enforcement actions processed by the court system (Direção-Geral da Política de Justiça, 2017).

The structure of this component of the judicial system was visibly complex and often counterproductive and burdensome. It was not equipped with adequate infrastructure to deal with this increased demand, particularly civil enforcement actions. A reform was required. The reform of the Portuguese justice system has taken as one of its main priorities the substantial reduction of delays and pending civil enforcement actions. To achieve this objective, a series of measures were implemented, including (Correia and Videira, 2016): a) improvements in the role of law enforcement officers, b) reinforcement of supervision and disciplinary powers of the regulatory body, c) restructuring of the judicial map, d) improvements to the alternative dispute resolution system, e) improvement and streamlining of procedures in the civil procedural area^[4] (including the elimination of unnecessary formalities).

Progressively, after an implementation period that covered approximately one year and six months, a decrease in the number of pending civil enforcement actions became evident. These results were considered particularly significant, since they put an end in more than 20 years of constant increases in the number of pending of this type of actions.

More than three years after the end of the financial and economic assistance program, and after the most critical phase has passed, the following question remains: the Portuguese judicial system took advantage of the rare contextual, political and social opportunity to sustain the improvements achieved during the Troika? Was this initial trend confirmed and continued?

The approach used in this article closely follows that presented by Correia and Videira (2015, 2016).

Despite the existence of relevant literature on the subject matter, no theoretical perspective was favored by the authors in order to encourage readers to refrain from any theoretical bias^[5]. Thus, it is important to emphasize that the purpose of this article is to foster dissemination and sharing, making available a theoretically neutral and statistically sound analysis of the experience of the Ministry of Justice with the Troika after their departure from the country (with emphasis on the later period to the adjustment program). In doing so, future theoretical and empirical work may be inspired by this unbiased approach.

3. Metodology

Empirically, this study analyses civil enforcement actions in the courts of first instance in Portugal in the perspective of the behavioral temporal evolution of this type of case, based on a sample of 126 months, between January 2007 and June 2017 (covering a period of ten and a half years and representing a 15-month increase, or 13.5% over the 111-month period used by Correia and Videira (2016) and an increase of 48 months, or 61.5% over the 78-month period used by Correia and Videira (2015)). Of the 126 months, 53 months (January 2007 to May 2011) precede the arrival of the Troika in Portugal (henceforth referred to as the pre-Troika period); 34 months (from June 2011 to March 2014) correspond to the stay of the Troika in Portugal (henceforth referred to as the Troika period); and the remaining 39 months (from April 2014 to March 2016) correspond to the period after the troika's departure from the country (hereafter referred to as the post-Troika period). The number of executive actions filed, finished^[6] and pending^[7] in the Portuguese courts of first instance constitute the raw data used in this investigation^[8].

These statistical data^[9] were made available to the public, by the Directorate-General for Justice Policy, a governmental body belonging to the Ministry of Justice, which is responsible for statistical data in the justice sector. The three original variables were complemented by three additional composite indicators: procedural balance^[10], procedural resolution rate^[11] and disposal time^[12]. The formulas for calculating these indicators can be found, for example, in Correia and Videira (2015, 2016):

Correia y Videira (2015, 2016)

In the absence of Gaussian distributions for the monthly data in each category (pre-Troika, Troika and post-Troika groups), the application of the parametric ANOVA test was set aside^[13] in favor of the non-parametric Kruskal-Wallis (Kruskal and Wallis, 1952), in order to determine the validity of the following scenarios:

H0: The Pre-Troika period, the Troika period, and the post-Troika dataset have equal medians.

H1: The Pre-Troika period, the Troika period, and the post-Troika dataset do not have equal medians^[14].

4. Results

This section is divided into two distinct subsections, in order to follow the structure proposed by Correia and Videira (2016), Subsection 1) presents a descriptive statistical analysis which establishes strong evidence of the prevalence of positive results not only in the Troika period, but also in the post-Troika period. Subsection 2), on the other hand, makes use of statistical tests to obtain an unbiased confirmation of the results suggested by the descriptive statistical analysis.

1) Descriptive statistics- Dissipation of Doubts about Success, Continuation and Confirmation of Positive Results

Due to the usual period of judicial holidays, with special emphasis on the month of August, seasonality is an inherent characteristic of the data presented in figure 1. This seasonal nature, which is more intense at the end of the process, is particularly visible between 2007 and 2012. Figure 1 shows the evolution, between January 2007 and June 2017, of the number of civil enforcement actions (entries and terminations) in the Portuguese courts.

Figure 1
Civil enforcement action entries and terminations, January 2007–June 2017
Own authorship.

To reduce these effects and obtain a less biased reading, the data was adjusted to compensate for the seasonal effects^[15]. Figure 2 shows the monthly numbers of the civil enforcement actions filed and completed, between January 2007 and June 2017, adjusted to compensate for seasonality. The calculations of the procedural balance, procedural resolution rate and disposal time (figures 3 to 5) are based on the number of processes filed and ended shown in figure 2^[16].

Figure 2
Civil enforcement action entries and terminations, adjusted according to seasonality, January 2007–June 2017
Own authorship.

The compensated procedural balance for seasonality is, for the period in question, shown in figure 3. The change in trends reported by Correia and Videira (2015, 2016), started about a year and a half after the start of the adjustment program, remains very visible in the post-Troika period. Of the 126 months considered in the analysis, 61 presented favorable procedural balances (number of cases ended greater than the number of cases filed, resulting in a decrease in the pendency equivalent to the balance). Of those 61 months, 6 (or 9.8%) were registered in the pre-Troika period, 18 (or 29.5%) were registered in the Troika period and 37 (or 60.7%) were registered in the post-Troika period. However, in the 53 months that included the pre-Troika, only 11.3% (6 months) presented favorable procedural balances, in the 34 months covering the Troika period, 52.9% (18 months) presented favorable procedural balances and in the 39 months covering the post-Troika period, a surprising 94.9% (37 months) presented favorable procedural balances.

Figure 3
Procedural balance of civil enforcement actions, adjusted according to seasonality, January 2007-June 2017
Own authorship. White bars represent favorable values ​​(negative procedural balances) and dark bars represent unfavorable values ​​(positive procedural balances).

The procedural resolution rate adjusted for seasonality is, in turn, and for the same period, shown in Figure 4^[17]. At the procedural resolution rate adjusted for seasonality, from mid-2013 and extending well beyond the end of the adjustment program. Of the 126 months considered in the analysis, 61 presented favorable procedural resolution rates (i.e. over 100%, resulting in a decrease in pendency). Of those 61 months, 6 (or 9.8%) were registered in the pre-Troika period, 18 (or 29.5%) were registered in the Troika period and 37 (or 60.7%) were registered in the post-Troika period. It should also be noted that in the 53 months that included the pre-Troika period, only 11.3% (6 months) had favorable procedural resolution rates, in the 34 months covering the Troika period, 52.9% (18 months) had favorable procedural termination rates, and the 39 months covering the post-Troika period had a surprising 94.9% (37 months) of favorable procedural termination rates^[18].

Figure 4
Procedural resolution rate of civil enforcement actions, adjusted according to seasonality, January 2007-June 2017
Own authorship. White bars represent favorable procedural resolution rates (over 100%) and dark bars represent unfavorable procedural resolution rates (below 100%).

In turn, for the period under review, Figure 5 shows the seasonally adjusted disposal time. As with the indicators presented earlier, a change in trend is observable, especially from January 2013 onwards. Of the 126 months considered in the analysis, 53 presented disposal times of less than 1,500 days^[19]. Of those 53 months, 8 (or 15.1%) were registered in the pre-Troika period, 17 (or 32.1%) were registered in the Troika period and 28 (or 52.8%) were registered in the post-Troika period. However, of greater emphasis is the fact that in the 53 months that included the pre-Troika period, only 15.1% (8 months) had disposal times of less than 1,500 days in the 34 months covering the Troika period, 50.0% (17 months) presented the disposal times below 1,500 days and in the 39 months that cover the post-Troika period 71.8% (28 months) presented disposal times less than 1,500 days.

Figure 5
Disposal time of civil enforcement actions, adjusted according to seasonality, January 2007-June 2017
Own authorship. White bars represent disposal times below 1,500 days and dark bars represent disposal times over 1,500 days

Figure 6 shows the evolution, between January 2007 and June 2017, of the number of pending civil enforcement actions, which is a consequence of the results presented previously. A detailed analysis of figures 2 to 6 suggests that the initial stabilization and decline of pending cases during the Troika period and the subsequent continued decline in the post-Troika period were not achieved by chance or by random fluctuations, but rather by the measures implemented by Portugal in the context of the objectives defined in the MoU. If we consider the maximum reached in August 2012 as a benchmark in June 2017, a reduction of more than 505,000 pending cases has been achieved (including a reduction of more than 290,000 pending cases in the first 39 months of the post-Troika period). The measures, even after the departure of the Portuguese Troika, appear to have proved to be effective.

Figure 6
Pending civil enforcement actions, January 2007–June 2017
Own authorship.

The data allows us once again to ask an interesting and relevant question: do the values ​​of the indicators examined above (cases filed, ended and pending, procedural balance, procedural resolution rate and disposal time) for the three periods considered (pre-Troika, Troika and Post-Troika), have different statistical properties? If so, these differences cannot be merely attributed to the common fluctuations of the phenomena under study and should be considered as consequences of the continued action of the Portuguese judicial system to address the challenges identified in the 2011 MoU.

2) Statistical Evidence - A final confirmation

Due to the absence of Gaussian distributions for the various data groups, the Kruskal and Wallis (1952) test was applied to determine if the three datasets come from the same population (H0) or, alternatively, if they come from different populations (H1). Table 1 presents the results of the Kruskal-Wallis test for the six variables included in this study.

* Values adjusted ​​for seasonality.

In all variables considered, H0 is rejected (p-values ​​= 0.000 < 0.05), and H1 is validated, that is, the three datasets do not come from the same population.

The logical question arising from the Kruskal-Wallis test results is, according to Correia and Videira (2016): if the three datasets do not come from the same population, should their results (for each of the six indicators) be considered statistically different or, instead, can they be statistically grouped?

To answer this question, Table 2 presents a stepwise step-down comparison ^[20] for the statistical similarity of the medians for the pre-Troika period, the Troika period, and the post-Troika period^[21].

Understanding the contents of Table 2 complements and confirms the results reported in previous studies (Correia and Videira, 2015, 2016). On the one hand, with regard to the median number of filed cases adjusted for seasonality, it is possible to conclude that there is no statistically significant difference in relation to the pre-Troika and Troika periods, but the difference of approximately 8,000 processes filed between these two periods and the post-Troika period is statistically significant and indicates a decrease in society's demand for the resolution of this type of case (median of 20,913 cases filed per month before the Troika arrived in Portugal, 21,158 cases filed during the Troika's stay in Portugal and 12,986 cases filed per month after the departure of the Portuguese troika).

On the other hand, as regards the cases ended, the three periods are statistically different: median^[22] of 16,045 cases ended per month before the troika arrived in Portugal, of 23,938 cases ended per month during the Troika's stay in Portugal and 20,386 cases ended per month after the departure of the Troika from Portugal. Despite the average decline of more than 3,500 cases ended per month after the departure of the Portuguese Troika, compared to the previous period, the median value per month of the cases for the post-Troika period is still 4,300 units higher than in the pre-Troika period. This important and sustained improvement in the number of finished processes, in turn, strongly influences the results of the composite indicators used in this analysis.

The results obtained in terms of pendency are also statistically different for the three periods under analysis: median of 1,009,556 cases pending before the Troika arrived in Portugal, 1,232,333 cases pending during the Troika's stay in Portugal and 942,311 cases pending after the departure of the Troika from Portugal. Despite the median increase in more than 222,000 pending cases during the Troika period compared to the previous period, the median value of pending cases for the post-Troika period is more than 67,000 units below the value for the pre-Troika period and more than 290,000 units below the value for the Troika period. This desirable and sustained decline in the median number of pending cases began in 2013 and remains in progress more than four and a half years later. It is also noteworthy that the post-Troika period is the only one with a monthly median of less than 1,000,000 pending cases.

With regard to the seasonally adjusted procedural balance, it can be concluded that there is no statistically significant difference in relation to the Troika and post-Troika periods (both with favorable, that is to say, negative, procedural balance values), but the approximate difference between -5,000 and -12,000 units between these two periods and the pre-Troika period statistically indicate a substantial improvement in the procedural balance for civil enforcement actions (median of +4,422 cases per month, a positive median procedural balance, therefore, unfavorable) before the Troika arrived in Portugal, of -415 cases per month (a negative median negative procedural balance, therefore, favorable) during the Troika's stay in Portugal and -7,616 cases per month (a negative median procedural balance, therefore, favorable) after the departure of the Troika from Portugal. It should also be noted that the difference in the "signal" constitutes a strong qualitative improvement^[23].

On the other hand, the results obtained in terms of the procedural resolution rate adjusted for seasonality are statistically different for the three periods under analysis: monthly median of 78.1% before the arrival of the Troika to Portugal, of 102.1% during stay of the Troika in Portugal and of 160.4% after the departure of the Troika from Portugal.

There was therefore an increase in the median by 24.0 percentage points during the Troika period when compared with the previous period and the median value of the procedural resolution rate for the post-Troika period is 58.3 percentage points above the median value for the Troika period and more than 82 percentage points above the median value for the pre-Troika period. It should also be noted that the increase above the 100% mark for the procedural resolution rate also constitutes a strong qualitative improvement, directly related to the reduction in the number of pending cases.

Finally, the results obtained in terms of the disposal time adjusted for seasonality are also statistically different for the three periods under analysis: monthly median of 1,966 days before the arrival of the Troika to Portugal, of 1,512 days during the Troika's stay in Portugal and of 1,239 days after the departure of the Troika from Portugal. There was therefore a decrease in the median by 454 days during the Troika period when compared to the previous period, and the median disposal time for the post-Troika period is 273 days below the median value for the Troika period and 727 days below the median value for the pre-Troika period. It should be noted that the reduction in disposal time indicates the existence of important fast gains in the processing of civil enforcement actions in the Portuguese judicial system (in fact, the reduction of 727 days in disposal time between the pre-Troika period and the post-Troika period corresponds to a reduction of almost two years in this indicator).

5. Discussions and conclusions

When the Portuguese authorities began pursuing the objectives set out in the Memorandum of Understanding of the Justice Sector in May 2011, the deterioration of the socio-economic and financial environment in the country contributed to a general increase in the number of pending civil enforcement actions^[24].

In March 2014, the situation was greatly improved: the judicial system improved significantly, and this was reflected in the values ​​achieved in indicators such as the procedural balance, the procedural resolution rate or the disposal time. As a result, the number of pending civil enforcement actions has begun a steep decline^[25] and, as is evident from the data presented in this study, continues today. The next logical and essential step in this process was to assess whether the public policies implemented would produce only short-term results or, alternatively, whether those positive outcomes would persist.

After the Troika's departure from March 2014 to June 2017, the statistical response is clear: the community’s demand for civil enforcement actions has diminished (probably reflecting the improvement of the socio-economic and financial environment); the judicial system supported, on a sustained basis, most of the improvements implemented during the Troika period; and the number of pending civil enforcement actions continues to decline considerably. All these results converge in highly improved values ​​for indicators such as the procedural balance, the procedural resolution rate or the disposal time. It should be noted that these indicators improved not only when compared to the period prior to the Troika but also when compared to the Troika period, so it can be said that the good results were not only continued but even deepened and strengthened.

Therefore, it is acceptable to conclude, based on the empirical study presented here, that the success of the implementation of public policies at a civil level, reported by Correia and Videira (2015, 2016), was not limited to the Troika period, but extends in addition, with observable statistically significant results, 39 months (more than three years) after the departure of the Troika from the country. Doubts are thus dispelled, and the continuation of good results is, as such, empirically confirmed.

As for Correia and Videira (2016), we leave the task of analyzing to what extent these results are evidence of a successful strategy of the International Monetary Fund, the European Commission and the European Central Bank for future research for countries that benefited from financial assistance. A caveat is due in this regard and caution is advised in the assessment of the benefits of the Troika.

The intention of this document is not to present a more optimistic view of the facts. The conclusions are restricted to civil enforcement actions (see Correia and Jesus (2016) for an instance of other Portuguese lawsuits that do not follow the same behavior) and the variables used should not be extrapolated carelessly. It should be noted that even for civil enforcement actions, factors such as the duration of pending cases were not considered. The information presented is factual and the interpretation assumes a ceteris paribus scenario. As in all sciences, the conclusions should be considered valid until a better explanation emerges for the same data.

Future studies may consider pursuing a few different paths. The first is to carry out similar empirical studies on other types of actions specifically provided for in the Memorandum of Understanding (e.g. bankruptcy, insolvency and corporate reorganization proceedings, or special revitalization processes). This will enable a more comprehensive assessment of the tangible achievements of the adjustment program for the Portuguese judicial system.

A second path is related to the pursuit of empirical studies of results monitoring, at a civil enforcement action level, resulting from objectives included in the MoU. This will allow for the continuation of the construction of the rich and detailed historical reference that has been documented, which will, in turn, determine whether the successful history resulting from the experience in the justice sector in Portugal and now confirmed in this article is limited to a well-defined period of time, and is destined to reverse sooner or later, or whether, on the contrary, the fruitful results achieved will be continued in the medium to long term.

[4] It is important to highlight the main legislative measures, namely Decree-Law nº 4/2013 of 11 January, which approved a set of provisional measures to combat pending and delays, which later was absorbed by the new Code of Civil Procedure. (Error 1: El enlace externo [4] debe ser una URL) (Error 2: La URL [4] no esta bien escrita)

[5] As the perspective of new public management (Frederickson, Smith, Larimer and Licari, 2012; Gomes, 2007; Hill and Hupe, 2014; Lane, 2000), the perspective of judicial governance theories (Guimarães, Correia, Bilhim, and Akutsu, 2015; Frederickson et al., 2012), the perspective of the permanent crisis of judicial systems (Campbell, 2013), the perspective of erosion of judicial legitimacy and concern for the separation of powers (Langbroek, 2008; Stephenson, 2004), the perspective of the productivity of human resources and demand pressure (Walsh, 2008), or the perspective of the macro-model approach of the judicial system (Ambach and Rackwitz, 2013; Bell, 2006).

[6] A precise definition of "process completed" is provided by the Directorate-General for Justice Policy in several of its official data (see, for example, the Direção-Geral da Política de Justiça (2016)).

[7] A precise definition of "pending case" is provided by the Directorate-General for Justice Policy in several of its official data (see, for example, the Direção-Geral da Política de Justiça (2017)).

[8] The cases that were transferred, merged, incorporated or integrated into other processes and those sent to another entity were removed from the initial data because they did not correspond to new cases in the courts, but rather to internal transfers within the Portuguese judicial system, not reflecting movements of supply or demand.

[9] The statistical data was retrieved from http://www.siej.dgpj.mj.pt

[10] According to Correia and Videira (2015), "the negative values ​​correspond to a favourable procedural balance (more processes ended than processes filed and therefore a decrease in pending) and the positive values ​​correspond to an unfavourable procedural balance (more processes that have been ended, and thus an increase in pending)."

[11] According to Correia and Videira (2015), "the values above 100% ​​correspond to a favourable procedural resolution rate (more processes ended than processes filed and therefore a decrease in pending) and the values below 100% ​​correspond to an unfavourable procedural balance (more processes that have been ended, and thus an increase in pending)."

[12] Or disposal time. According to Correia and Videira (2015), "the smaller the value, the more favourable it is".

[13]The use of the ANOVA test requires the existence of Gaussian and homoscedasticity distributions (equality of variance). If one (or both) of these prerequisites fails, the ANOVA test should not be applied to that dataset.

[14]The reader is also recommended to consult the work of Catarino and Correia (2016), Correia and Catarino (2016), Correia, Carrapato, and Bilhim (2016), Correia and Dias (2018) and Correia, Pinto, Garcia, and Dias (2015).

[15] Following the procedure adopted by Correia and Videira (2015, 2016), several measures were taken: removal of seasonality, removal of any possible linear trend of the data and realisation of a stabilising variance transformation.

[16] Note that the number of pending enforcement actions is not affected by seasonality, and so the adjustment of this dataset is unnecessary (figure 6).

[17] Note that in August 2007, the seasonally adjusted procedural resolution rate was exceptionally high at 849.4% (not fully visible in Figure 4). However, this figure is obtained in extremely low volumes of cases filed and ended (530 and 2,705, respectively) and results from a relatively high number of cases settled in a month of judicial holidays. Despite its magnitude, the very low volume of cases indicates that it is not a particularly relevant month for reducing the number of pending cases.

[18] It should be noted that the procedural resolution rate indicator "has the advantage, in relation to the procedural balance indicator, of being a measure based on relative rather than absolute values, allowing prolonged periods of time to be better compared, even if the demand and supply conditions of the judicial system change significantly” (Correia and Videira, 2015).

[19] While it is important to stress that 1,500 days (over four years) is a long period of time for a court case to be finished (a time lapse that few people or companies are willing to support), the authors chose this threshold for reasons of comparison with the literature of previous works, namely Correia and Videira (2015, 2016).

[20] The stepwise step-down comparison is a complement to the Kruskal-Wallis test that allows a refinement of analysis: the identification of similarities or statistical dissimilarities for the various combinations of categories.

[21]These three periods correspond to the three original datasets described in the methodology section. The groups in table 2 were created for each variable, using the stepwise step-down comparison. The periods (pre-Troika, Troika and post-Troika) within the same group can be considered to have statistically equivalent medians for this variable. The periods (pre-Troika, Troika and post-Troika) in different groups can be considered to have statistically different medians for this variable.

[22] It should be remembered that, as indicated in the section on methodology, the median is used rather than averages for all variables, due to the absence of Gaussian distributions.

[23] A change from a period of unfavourable procedural balances (+ sign) and an increase in the number of pending cases for a period of favourable procedural balances (- sign) and a decrease in the number of pending cases.

[24] For further information on this subject, see European Commission for the Efficiency of Justice (2016).

References

Ambach, P., & Rackwitz, K. (2013). A Model of International Judicial Administration? The Evolution of Managerial Practices at the International Criminal Court. Law and Contemporary Problems, 76 (3-4), 119-161. Retrieved from: https://scholarship.law.duke.edu/cgi/viewcontent.cgi?article=4373&context=lcp

Barnard, C. (1971). The Functions of the Executive: 30th Anniversary Edition. Cambridge: Harvard University Press.

Bell, J. (2006). Judiciaries within Europe: A Comparative Review. Cambridge: Cambridge University Press.

Campbell, D. (2013). The Sky is Falling (Again): Evaluating the Current Crisis in the Judiciary. New England Law Review,47, 571-603. Retrieved from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2515994

Catarino, J., & Correia, P. (2016). Receitas Fiscais e Tributação Geral sobre o Consumo em Portugal: Um Estudo sobre Eventuais Assimetrias do Comportamento dos Sujeitos Passivos do Imposto no Final da Primeira Década do Século XXI. Revista da FAE, 19 (1), 6-17. Retrieved from: https://www.researchgate.net/publication/315772893_Receitas_Fiscais_e_Tributacao_Geral_sobre_o_Consumo_em_Portugal_Um_Estudo_sobre_Eventuais_Assimetrias_do_Comportamento_dos_Sujeitos_Passivos_do_Imposto_no_Final_da_Primeira_Decada_do_Seculo_XXI

Commission for the Efficiency of Justice (2016). European Judicial Systems – Edition 2016 (2014 data): Efficiency and Quality of Justice. Retrieved from Conselho da Europa:

Correia, P. M., Carrapato, P., & Bilhim, J. (2016). Administração Hospitalar em Portugal: Relação entre Antiguidade e Envolvimento Laboral, e Implicações para o Risco de Saída. Jornal Brasileiro de Economia da Saúde, 8 (2), 73-79. doi: 10.21115/JBES

Correia, P. M., & Catarino, J. (2016). Ingreso Bruto Tributable del IVA: Evidencia de Diferenciación de los Municipios de la Costa Portuguesa. Revista del CLAD Reforma y Democracia, 64, 225-246. Retrieved from: http://old.clad.org/portal/publicaciones-del-clad/revista-clad-reforma-democracia/articulos/064-febrero-2016/gross-revenue-for-vat-tax-evidence-of-differentiation-at-the-portuguese-littoral-municipalities

Correia, P. M., & Dias, M. T. (2018). O poder detido pelos dirigentes máximos das organizações: evidências estatísticas de homogeneidade ao nível das perceções, em função da idade e antiguidade dos trabalhadores, em Portugal. Pensamiento Americano, 11(20), 80-92. doi: http://dx.doi.org/10.21803%2Fpenamer.11.20.497

Correia, P. M., & Jesus, I. (2016). Combate às Transferências Bancárias Ilegítimas pela Internet no Direito Português: Entre as Experiências Domésticas e Políticas Globais Concertadas. Revista Direito GV, 12 (2), 542-563. doi: 10.1590/2317-6172201622

Correia, P. M., & Videira, S. (2015). Troika’s Portuguese Ministry of Justice Experiment: An Empirical Study on the Success Story of the Civil Enforcement Actions. International Journal for Court Administration, 7 (1), 37-50. doi: 10.18352/ijca.180

Correia, P. M., & Videira, S. (2016). Troika’s Portuguese Ministry of Justice Experiment, Part II: Continued Positive Results for Civil Enforcement Actions in Troika's Aftermath. International Journal for Court Administration, 8 (1), 20-31. doi: 10.18352/ijca.215

Correia, P. M., Pinto, R., Garcia, B., & Dias, M. (2015). Teoria Institucional e o Poder dos Dirigentes dos Hospitais em Portugal, Parte III: Comparação com Outros Setores de Atividade. Jornal Brasileiro de Economia da Saúde, 7(3), 135-141. Retrieved from: https://www.researchgate.net/publication/309410556_Teoria_Institucional_e_o_Poder_dos_Dirigentes_dos_Hospitais_em_Portugal_Parte_III_Comparacao_com_Outros_Setores_de_Atividade

Direção-Geral da Política de Justiça. (2016). Estatísticas da Justiça – Alguns Indicadores Estatísticos Sobre os Processos nos Tribunais Judiciais de 1ª Instância, 2007-2015. Retrieved from Ministério da Justiça, Portugal: http://www.dgpj.mj.pt/sections/siej_pt/destaques4485/sections/siej_pt/destaques4485/algunsindicadores_2/downloadFile/file/Resultados%202015_20161028.pdf?nocache=1477659998.74

Direção-Geral da Política de Justiça. (2017). Estatísticas Trimestrais Sobre Ações Executivas Cíveis (2007-2017) – Trimestre 2. Retrieved from Ministério da Justiça, Portugal: http://www.dgpj.mj.pt/sections/siej_pt/destaques4485/estatisticas-trimestrais8196/downloadFile/file/A%C3%A7%C3%B5esC%C3%ADveis_2%C2%BATrimestre_2017.pdf?nocache=1510740566.21

Gomes, C. (2007). The Transformation of the Portuguese Judicial Organization: Between Efficiency and Democracy. Utrecht Law Review, 3 (1), 101-111. Retrieved from: https://www.utrechtlawreview.org/articles/abstract/10.18352/ulr.39/

Guimarães, T., Correia, P., Bilhim, J., & Akutsu, L. (2015). Governança em Tribunais de Justiça de Portugal: Adaptação e Validação de Escala de Medida. Ciências e Políticas Públicas, 1 (1): 25-46. Retrieved from: http://capp.iscsp.ulisboa.pt/images/Ejournal/CPP%20Journal%20CAPP%20v1%20n1.pdf

Frederickson, H., Smith, K., Larimer, C., & Licari, M. (2012). The Public Administration Theory Primer. Boulder, Colorado: Westview Press.

Hayek, F. (1973). Law, Legislation and Liberty – Volume I: Rules and Order. Chicago: The University of Chicago Press.

Hayek, F. (1978). Law, Legislation and Liberty – Volume II. The Mirage of Social Justice. Chicago: The University of Chicago Press.

Hayek, F. (1979). Law, Legislation and Liberty – Volume III: The Political Order of a Free People. Chicago: The University of Chicago Press.

Hill, M., & Hupe, P. (2014). Implementing Public Policy. London: Sage.

Lane, J. (2000). New Public Management. London: Routledge.

Kruskal, W., & Wallis, W. (1952). Use of Ranks in One-Criterion Variance Analysis. Journal of the American Statistical Association, 47 (260), 583-621. doi: 10.1080/01621459.1952.10483441.

Langbroek, P. (2008). Entre Responsabilisation et Indépendance des Magistrats: La Réorganisation du Système Judiciaire des Pays-Bas. Revue Française D’administration Publique, 125, 67-79. doi: 10.3917/rfap.125.0067

Portugal. (2011). Memorandum of Understanding on Specific Economic Policy Conditionality. Retrieved from: http://ec.europa.eu/economy_finance/eu_borrower/mou/2011-05-18-mou-portugal_en.pdf

Stephenson, M. (2004). Court of Public Opinion: Government Accountability and Judicial Independence. The Journal of Law, Economics, and Organization, 20 (2), 379-399. doi: 10.1093/jleo/ewh038

Walsh, B. (2008). Judicial Productivity in India. International Journal for Court Administration, 1 (1), 23-30. doi: 10.18352/ijca.123