Microsoft Word - PDF_Issue_14_1_Rebessi.docx Italian Political Science, VOLUME 14 ISSUE 1, MARCH 2019 © 2019 Italian Political Science. ISSN 2420-8434. Volume 14, Issue 1, 38-49. Contact Author: Elisa Rebessi, University of Milan. E-mail address: elisa.rebessi@unimi.it Unexpected consequences of the implementation of the Code of Public Contracts: judicial decisions on awarding procedures in three Italian cities Elisa Rebessi UNIVERSITY OF MILAN Abstract By building on Spiller (2008) and Moszoro and Spiller (2012), this paper analyses the implementation of the Italian Code of Public Contracts. Contracting authorities are expected to design tendering procedures strategi- cally, using less discretion the higher the risk of third-party challenges, and more discretion the lower this risk. The analysis considers three cases: the municipalities of Milan, Turin and Rome. After assessing the potential for third-party challenges with measures of competitiveness of the economic environment where the municipal- ities are located, the paper sheds light on how the municipalities apply the Code, analysing judicial decisions that accept the reasons of the applicants on the awarding of public contracts issued by Regional Administrative Courts (2006-2018). Results confirm expectations on the strategic use of administrative discretion, and also reveal that the Code is generating litigation when the contracting authorities rigidly assess the moral and con- tributory requirements of the economic operators. 1. Introduction ollowing New Public Management reforms of public administration, the awarding of public contracts is nowadays a cornerstone in the delivery of services and public works. Italy, affected in the 1990s by administrative reforms inspired by the prin- ciples of New Public Management and at the beginning of the 2000s by a constitutional reform that considerably strengthened local governments, provides a context in which local government plays an important role in the awarding of services and public works, including a considerable degree of discretion in shaping decisions and implementation (Marcou 2016). Decisions on the awarding of public contracts actually feed local econo- mies and open opportunities for shaping public service implementation, with specific reference to the role of the Administrative Courts. Nevertheless, at the present time there is no empirical evidence on the role of courts and their interactions with the other actors of local government in shaping the implementation of public awarding. This paper aims to fill this gap by focusing on the implementation in three Italian cities of a key regulatory instrument: the Italian Code of Public Contracts. The Italian Code of Public Contracts aims to limit discretionary actions of the contracting authorities through a formalized, process-oriented legal framework (OECD 2016). This paper analyses the implementation of the Code, building on Spiller (2008) and Moszoro and Spiller (2012) who provide an institutional theory of public contracts and a rationale for the use of discretion in public contracting. In particular, by analysing the strategies that local contracting authorities use in designing public contracts, we show that the potential for third-party challenges F ELISA REBESSI 39 (i.e. legal challenges from competitors of the contractors) affects the use of administra- tive discretion in public contracts more than by-law prescriptions. Scholars in positive political theory and rational choice have highlighted the role of third parties in supervising and controlling public contracts in democratic societies (McCubbins and Schwartz 1984; McCubbins, Noll and Weingast 1987, 1989). These scholars have also extensively analysed rational anticipation of courts’ decisions by legis- lators (among others: Ferejhon and Shipan 1990, Ferejhon and Weingast 1992, Steunenberg 1997, Santoni and Zucchini 2006, Brouard and Honnige 2017). This kind of theoretically driven and empirically grounded approach has not been applied to the study of local public procurement in Europe, and in particular in the case of Italy. By analysing the strategies that local administrations use in designing public contracts, the present contribution shows how rational anticipation of litigation affects the implementation of the Code. Hence, the present contribution enriches the literature in two ways. First, it in- vestigates the empirics of implementation in public procurement in Italy by considering actors such as the decision-makers, the courts, and the third applicants. Second, it applies rational choice theories to the behaviour of these actors, considering also the rationality of the Municipality, thus enriching the academic debate on their role in public procure- ment (Wollmann et al. 2016). The paper addresses the following research questions: how are local government ad- ministrations implementing the Code of Public Contracts and, in particular, how are they using administrative discretion in tendering procedures? What type of litigation is foster- ing the implementation of the Code before the administrative judge? Is litigation revealing unexpected consequences of legislation, once implemented by local governments? The paper is based on a comparative analysis of three cases: the Italian municipali- ties of Milan, Turin and Rome. The use of administrative discretion in the awarding procedures is assessed through a diachronic analysis of the decisions on the awarding of public contracts issued by Regional Administrative Courts in the three municipalities be- tween 2006 and 2018. The potential for third-party challenges, which is supposed to affect the use of administrative discretion, is approximated by measures of the competi- tiveness of the economic environment where the municipalities are located. Results confirm that the administrations tend to use more discretional procedures in less competitive economic contexts, where the potential for third-party challenges is lower, and more rigid procedures in more competitive economic contexts, where the risk of potential challenges is higher. Surprisingly, the implementation of the Code is gener- ating litigation when the contracting authorities rigidly assess the moral and contributory requirements of the economic operators. This outcome possibly reflects a controversial aspect of the Code, which gives responsibility for the control of the participatory require- ments to the same contracting authorities, and not to an independent body not involved in the awarding procedures. The paper is structured as follows: section one presents the theoretical framework, according to which one hypothesis is generated. Section two presents the cases, the de- pendent and independent variables, and briefly outlines the regulatory framework of the Code of Public Contracts. Section three provides the analysis of litigation for the awarding of public contracts. Section four concludes the paper. Unexpected consequences of the implementation of the Code of Public Contracts 40 2. Institutional theories of public contracts and their implications for the use of administrative discretion Following New Public Management reforms, public administrations have differentiated the provision of public works and services, opening up the traditional hierarchical pro- duction and delivery systems to contract-like relationships (Pollit 2001). In this context, the awarding of public contracts has become the main instrument for local government administrations to purchase goods and services to achieve their institutional goals.The fundamental features of a contract depend on the nature of contracting hazards (William- son 1979) and specific forms of opportunism can arise in public contracting: opportunism from public agent (‘governmental opportunism’) and opportunism from interested third parties (‘third-party opportunism’) (Spiller 2008). Governmental opportunism consists in the government’s power to adopt new legislation to obtain part of the quasi-rents of the contracting partner. This might be the case of the adoption of a secondary legislation act that denies a tariff increase to a public utility. Governmental opportunism however finds institutional limits, since the potential for the opportunistic use of legislative powers de- pends on the control of the executive over the legislature and on the presence of an independent judiciary. Third-party opportunism relates instead to the incentives that competitors, interest groups and political opponents may have in challenging a public contract. The legitimacy of a public contract can be contested both in court and informally through the media. Interested third parties play a fundamental role in democratic socie- ties in supervising and controlling public contracts (McCubbins and Schwartz, 1984; McCubbins, Noll and Weingast 1987, 1989); however, they can also challenge the probity of a public agent for the benefits they may receive in political and economic terms. A successful challenge in political terms entails the replacement of the public agent with an agent whose preferences are closer to those of the interested third party. A suc- cessful challenge in economic terms causes the replacement of the private contractor, or the terms of the contract are changed and become more favourable to the interested third party. Moszoro and Spiller (2012) claim that the higher level of contract specificity and rigidity in public contracting can be interpreted as a risk adaptation by public agents. More precisely, by limiting their discretionary actions and increasing contract formali- ties and rigidities, public agents would reduce their exposure to the risk of third-party challenges. Marshall, Meurer and Richard (1994) similarly noted that agreements that can be contested by excluded operators tend to be more carefully delimited and governed by formal features. The strategies of the economic operators are affected in turn by the level of contract specificity and rigidity: the more discretion they observe in contract terms, the more room they have to challenge the contract while the less discretion they observe in the contract terms, the less room they have to challenge the contract. Spiller’s (2008) institutional theory of public contracts and Moszoro and Spiller’s (2012) account of public contract rigidity can be used to analyse the strategies adopted by contracting authorities in local public procurement, and in particular to explain the room contracting authorities leave for their own discretionary evaluations in adjudication pro- cedures. The presence of more rigid/less discretionary assessments in the adjudication process reflects the higher risk of third-party challenges. Third-party challenges are ex- pected to come from economic operators that would benefit from a private contractor’s replacement or from changing the rules of adjudication. Contracting authorities are ELISA REBESSI 41 expected to perceive a higher potential for third-party challenges when local productive systems are more competitive, and firms are more numerous and competitive enough to enter the procurement market and eventually appeal against the adjudication decision before an administrative court. According to the argument just illustrated, I put forward the following hypothesis: H1. I expect tendering procedures to be more rigid (less discretional), the higher the potential for third-party challenges. 3. Data and variables The analysis is based on three cases, the municipalities of Milan, Turin and Rome, three of the biggest cities in Italy, all characterised by advanced urban infrastructures and a value- added service industry, sectors at the centre of the public procurement market. The dependent variable, the use of discretion in tendering procedures, is assessed through the analysis of the decisions that accept the reasons of the applicants on the awarding of public contracts (‘decisioni in materia di aggiudicazione’) issued by Regional Administrative Courts against the municipalities of Milan, Turin and Rome between 2006 and 2018. In decisions that accept the reasons of the applicants (from now on ‘accepting decisions’), the Administrative Court states that the reasons of the applicant who is chal- lenging the decision of a Municipality on the awarding of a contract are founded. For instance, the Court can recognize that the procedures applied by the administration in a specific tender procedure, such as the use of weighting criteria to evaluate the offers, were not coherent with the tender notice or with specific provisions of the Code. If the Court states that the appeal is founded and assesses that the administration did not correctly ap- ply the law or the procedures of the tender, the judicial decision considers the acts illegitimate and therefore it nullifies them. The administration can decide to present an appeal to the second instance Court, the Council of State; otherwise the awarding proce- dure is nullified. The dependent variable, which describes the use of discretion in accepting decisions, is assessed through the judges’ description of two fundamental as- pects for the award of the contract: 1. the rigidity of the administration in applying the Code in the selection of participants in the tender, and 2. the coherence of the acts of the administrations with the procedures declared in the tenders. Judicial decisions on the awarding of public works and services are collected from the Juridical Database Leggi d’Italia. Applicants are unsuccessful bidders: firms, consortia, cooperatives, ventures, associations and, less frequently, public institutions that appeal for the annulment of awarding measures and against measures of exclusion. The appeals are against provisions adopted by the municipalities, represented by their mayor pro tempore. Decisions against provisions adopted by subsidiaries of the municipalities (e.g. Expo Mi- lano 2015 S.p.a. and other public owned companies) were not included. Judicial decisions are distributed as follows: 165 decisions against the municipality of Rome (45 of which are accepting decisions), 135 decisions against the municipality of Milan (21 of which are ac- cepting decisions), and 49 decisions against the municipality of Turin (5 of which are accepting decisions). Unexpected consequences of the implementation of the Code of Public Contracts 42 The Italian Code of Public Contracts1 aims to prevent corruption, promote transpar- ency and achieve the opening-up of public procurement to competition in coherence with European Directives.2 To this end, firms that intend to participate in a procurement pro- cedure must meet a list of requirements, which include general morality requirements, financial capacities, absence of contributory negligence and technical skills. In order to certify their status, firms must produce a series of certificates, such as the DURC ‘Docu- mento Unico di Regolarità Contributiva’ which attests regular payments of social security and insurance contributions, the subscription to official Registers (‘albi’), and other certif- icates of technical skills and self-declarations. The same contracting authorities are responsible for assessing firms’ requirements and are asked to verify the information pro- vided by the applicants. The Code of 2016 charged the Italian National Anti-Corruption Authority (ANAC) with the task of explaining through guidelines the evidences appropri- ate to justify operators’ exclusion. The Code regulates procurement both above and below specific financial thresholds, established and periodically updated by the European Commission (art. 35 of the Code). Procedural rules of the Code must be applied to contracts above the EU thresholds and al- low for any European firms or institution to participate in the procedures. Ordinary awarding procedures include open and restricted tender procedures. Open and restricted ordinary procedures are characterized by a limited use of discretionary powers in the se- lection of the bidders, who submit non-renegotiable offers. Special awarding procedures, which can be applied to procurements below the EU financial thresholds, include compet- itive procedure with negotiation, negotiated procedure without previous publication of the call for tender and direct purchasing (‘affidamento diretto’) permitted only for low value tenders (art. 36 of the Code). Negotiated procedures are characterized by significant dis- cretionary powers for the contracting authority, which consults potential suppliers and negotiates contract conditions with them (Baltrunaite et al. 2018). Unsuccessful bidders can apply to the Regional Administrative Courts for any measure adopted during the awarding procedures (ICLG 2019). This paper focuses precisely on the decisions issued by Regional Administrative Courts in three cities (Rome, Milan, Turin), in order to better de- scribe the implementation of the Code. In particular, we first describe whether the Courts recognize that the complaints of the applicants are founded, thus we focus on the judges’ evaluation of tendering procedures and on the reasons why the applicants were successful before the Courts (see Figure 1). The municipality with both the largest number of disputes before the administra- tive judge and the highest rate of acceptance decisions is Rome. Indeed, the municipality of Rome presents 165 judicial decisions, followed by Milan (135) and Turin (49). In all municipalities, rejection decisions are higher than acceptance decisions. In the city of Turin, the rejection rate is particularly high and amounts to 84%. In Milan, the rejection rate amounts to 68% and is close to Rome’s 60%. The highest rate of acceptance decisions is achieved by Rome, with 29% of accepted cases, followed by Milan, with 16% of total 1 Legislative Decree no 163/2006, then abrogated by Legislative Decree no. 50/2016, as amended by Leg- islative Decree no. 56/2017 and Law 96/2017 2 The Public Procurement Code in its first version of 2006 implements the EU Directives 2004/18/EC and 2004/17/EC; the second and present version of the Code, of 2016, transposes the EU Directives 2014/23/UE, 2014/24/UE and 2014/25/UE. ELISA REBESSI 43 decisions, and Turin, with 10% of acceptance decisions. Partially accepted cases, namely cases in which one instance is accepted, while the others are rejected or declared not ad- missible, amount to 16% in Milan, 10% in Rome and 6% in Turin. Two interesting features that emerge from data are the apparently low number of litigations in the city of Turin, compared to Rome and Milan, and the relatively high rate of acceptance decisions in the municipality of Rome. Figure 1. Distributions of judicial decisions on the awarding of public contracts for type of outcome, by municipality (2006-2018). To compare the scope of litigation, the number of administrative disputes is related to the amount of resources that the municipalities planned to spend on public works be- tween 2006 and 2018. The Code of Public Contracts requires the municipalities to publish the tri-annual plan for the commitment of public works with an estimated value equal to or greater than 10 thousand euro (from 2006).3 The total funds that the three cities planned to spend on public works between 2006 and 2018 are represented in Fig- ure 2. Data show that the three cities present different levels of investment. According to the data provided by the municipalities, the city of Turin presents a considerably lower level of planned investment in public works throughout the period considered: the yearly average value of planned funds for public works amounts to 161 million euro and the highest value observed, in 2009, does not reach half a billion euro. On the contrary, the total volume of funds is comparable in Milan and Rome, where, ex- cept for a few peaks, it is rarely lower than 1 billion and presents a yearly average value of 1.3 billion in the city of Milan and of 1.5 billion in the city of Rome respectively. It is 3 Information on total planned funds for the commitment of public works are available on the official websites of the three municipalities in the section: ‘amministrazione trasparente’, ‘calls for tenders and contracts’, and on Osservatorio regionale contratti pubblici of Lombardy Region and Osservatorio dei con- tratti pubblici of Lazio Region. Unexpected consequences of the implementation of the Code of Public Contracts 44 reasonable to suppose that, in absolute terms, the lower number of litigations observed in Turin can be due to the lower resources tendered for public works. Hence, if we take into account the total planned resources, the number of decisions in Turin is not lower than in the other municipalities. Figure 2. Total planned resources for public works (billion euro) in the three municipalities, by year Source: personal elaboration of data from Osservatorio contratti pubblici Regione Lombardia; Osservatorio dei contratti pubblici Regione Lazio (SITARL); Official websites of the Municipalities of Milan, Rome and Turin, sections ‘Amministrazione trasparente’, ‘Calls for tenders and contracts’ (‘Bandi di gara e contratti’). 4. Competitiveness of the economic environment The market of public procurement has a consistent local dimension and local firms are often the largest participants in the auctions (Moretti and Valbonesi 2015; Coviello and Gagliarducci 2010). The independent variable, the potential for third-party challenges, is approximated by measures of numerosity and competitiveness of the firms that operate in the main sectors affected by the public procurement market (manufacturing industry and service sector) in the regions to which the municipalities belong (regions of Lom- bardy, Piedmont and Lazio). In addition, more comprehensive indicators of competitiveness of the firms, together with measures of the macro-economic context, are considered for the provinces to which the municipalities belong (provinces of Milan, Turin and Rome). Regional data come from the Italian national statistics office (ISTAT) database on firms’ economic indicators. Data include information about firms’ local units, turnover, value added at factor cost, personnel costs, wages and salaries, gross investment in ELISA REBESSI 45 tangible goods, persons employed, and number of employees4. All economic indicators show that firms in the manufacturing sector are more numerous and competitive in the Lombardy Region. Considering the most relevant indicators, the average number of lo- cal units in Lombardy between 2008-2015 (equal to 94,070) is almost three times the number of local units in Piedmont (35,787) and almost four times the number of local units in Lazio (24,038). Firms in Lombardy are significantly more numerous also in the service sector5. Manufacturing firms seem more competitive in Piedmont than in Lazio, while the same does not apply in the service sector, where the situation is reversed. In both the manufacturing and services sectors, Lombardy presents more numerous and competitive firms. Provincial data on competitivity of the firms and on macro-economic contexts are provided by Ciccarelli (2006) who ranks the Italian provinces according to a series of in- dicators6. The province of Milan is in first place, followed by Rome in 26th place and Turin in 54th place as regards the competitiveness of local firms. The province of Milan is in 6th place, followed by Rome in 39th place and Turin in 41st place as regards the competitive- ness of the macro-economic context. All measures highlight that the municipality of Milan belongs to a more competitive economic environment, in which firms are more numerous and efficient compared to Turin and Rome. Since there is a larger number of economic operators in the market of public pro- curement and the risk of potential challenges increases, I expect that the Municipality of Milan uses less discretion in the awarding procedures in order to prevent litigation. On the contrary, given the lower number of economic operators in the market of public procurement and the lower risk of potential challenges, I expect that the administrations of Turin and Rome tend to use more discretion in their procedures. 5. Analysis of the accepting decisions for the awarding of public contracts The analysis of the economic environment outlines two contexts: one more competitive, in which the municipality of Milan is embedded, and where the public procurement market potentially involves more economic operators; the other less competitive, which concerns the municipalities of Turin and Rome, and that potentially involves fewer 4 Firms are classified according to the economic classification ATECO. Letter C defines the manufactur- ing industry, letters I, J, N, M, Q, R, and S define the Service sector. Data on the manufacturing industry are available between 2008 and 2015; data on the service sector are available between 2008 and 2014. 5 Lombardy has on average more than twice the number of local units in the accommodation and food services sector (Letter I, ATECO) (46,393) compared to Piedmont (21,620), and Lombardy’s firms are rather more numerous than Lazio’s firms in the sector (29,730). The same applies to information and communication services (Letter J, ATECO), where Lombardy presents 25,971 local units, against 14,882 units in Lazio and 8,242 units in Piedmont. In healthcare and social assistance services (Letter Q, ATECO), local units are on average 50,745 in Lombardy, with 35,490 in Lazio and 20,371 in Piedmont. 6 The competitiveness of local firms is measured by indicators of the firms’ natality/mortality, presence of advanced manufacturing, general industry productivity, average dimension of local units, rate of cor- porate firms, failure rates, and cross-territorial dissemination. The competitiveness of macro-economic contexts is measured with indicators of value added per capita, number of employees out of total popula- tion and working age population, information on trade credit usage, solvency ratio of firms, firms’ propensity to export. Unexpected consequences of the implementation of the Code of Public Contracts 46 economic operators. I expect that the judicial decisions reveal a more rigid (less discre- tional) use of the tendering procedures in the first context, where challengers of the contracts are potentially more, and a more discretional use of tendering procedures in the second context, where challengers are potentially fewer. By reviewing the content of the judicial decisions, I expect to find that there is a less discretional and more rigid use of tendering procedures in the first context, where challengers of the contracts are po- tentially more numerous, and thus where competitiveness over public procurement is higher. Conversely, I expect that the analysis of the judicial decisions in the less compet- itive contexts will show that the use of discretion in tendering procedures is higher. The analysis of judicial decisions for the awarding of public contracts in the three municipalities seems to confirm such expectations. The analysis of the 21 accepting decisions against the municipality of Milan reveals a rather homogeneous content of litigations. Indeed, 19 decisions concern cases where the applicants question their own exclusion, while the legitimacy of the choice of the con- tractor is contested in two decisions. Of those 19 decisions in which applicants question their own exclusion, in 14 cases the applicants claim that the contracting authority has implemented the Code too rigidly, in particular regarding the assessment of their moral and contributory requirements. This is the case for decisions 1314/2012, 1578/2012 and 2985/2012 where the applicants claim they were unfairly excluded from the tender, since the administration acknowledged irregularities in their tax and wage compliance certificate (DURC), but without further assessing the real gravity of their social security infringements. In decisions 4842/2009 and 4843/2009, applicants claim that the ad- ministration decided on their exclusion because of their false declarations made in previous tender procedures, but without further investigating their position through the consultation of the register of Public Works Observatory. In decisions 249/2018 and 250/2018, applicants claim that the administration interpreted too rigidly the provision of the Code that provides for the exclusion of economic operators who have been con- victed by definitive judgment of certain offences (in this case, the violations of norms on safety and health at work), since the offences were committed by a subject with a mar- ginal role in the company (a member of the board of auditors) and in the capacity of legal representative of another firm7. Five decisions contest the assessment by the admin- istration of cartel formation: the applicants contend that the mere presence of a family relationship is unfit to signal cartel formation (289/2013, 444/2013, 502/2013), or, by referring to the jurisprudence of the Council of State, that the elements considered by the contracting authority are insufficient to signal cartel formation (decisions 844/2015, 845/2015). A common feature emerging from decisions against the administration of Milan is the unwillingness of the contracting authority to assess deeper, contextual evaluations of the cases when certain irregularities have occurred. In particular, when there are sig- nals that violations of the Code such as lack of moral requirements, or contributory negligence have taken place, the contracting authority reacts by limiting its own discre- tionary power and does not take responsibility for further investigations into the 7 On similar cases, the National Anti-Corruption Authority (ANAC) and the supreme administrative court, the Council of State, have expressed contrasting opinions. In the above-mentioned decisions, the TAR of Lombardy confirmed Council of State jurisprudence against the interpretation of ANAC. ELISA REBESSI 47 operators. Quite unexpectedly, the rigid assessment of moral and contributory require- ments is generating litigation and is often the cause of the Milan administration’s defeat. This outcome reveals how the assessment of participatory requirements presents several problems in practice. In particular, the analysis of accepting decisions shows that the contracting authorities can receive documents which are not fully updated and may fail to assess the most recent changes in the status of the economic operators. In this re- spect, a potential weakness of the Code is that it assigns the responsibility for assessing the economic operators’ requirements not to an independent authority as is common in other regulatory frameworks (Gilardi 2005), but to the same contracting authorities. Conversely, the analysis of the accepting decisions in the municipalities of Rome (45 decisions) and Turin (5 decisions) reveals a more heterogeneous litigation, quite dif- ferent from that against the administration of Milan. In the municipality of Rome, a first group of 8 decisions reflects administrative inertia and other administrative errors. This is, for instance, the case of decision 12400/2015, where the administration did not final- ize the tendering procedure, nor did they conclude the agreement, and decisions 5336/2006 and 5337/2006, where the call did not clearly state the deadlines for partici- pating in the procurement procedure. A second group of 29 decisions in the municipality of Rome and the 4 decisions in the municipality of Turin reflect an improper use of dis- cretionary powers. This is, for instance, the case of: decision 2469/2008, where the administration of Rome introduced new weighting criteria after the bids were already opened; decisions 7326/2014 against the municipality of Rome, where the bids were not opened in public; decisions 2573/2011, 10313/2015 against the municipality of Rome and 2300/2006 against the administration of Turin where the weighting criteria used by the administration were judged unclear or applied inadequately; decisions 5607/2009, 122/2016, 1429/2017 against the municipality of Rome and 155/2016 against the munic- ipality of Turin, which outline that the administrations did not comply with the principles of transparency and public access required by the tender-opening sessions; decisions 12348/2008, 7716/2009, 3766/2011, 2217/2014 against the administration of Rome and decisions 1119/2014 and 1226/2017 against the municipality of Turin, which ascertain misuse of powers and incoherence of the administrative action, since the ten- ders were awarded to operators that lacked the requirements; decisions 1969/2016, 1505/2018 against the municipality of Rome and 160/2008 against the municipality of Turin, which state that abnormally low bids were wrongly included. Remaining deci- sions against the municipality of Rome concern claims for damages (1), assessment of cartel formation (2), real estate procurement (2), execution of the judicial decisions (2), and revocation decision (2). Most of the decisions against the municipality of Rome con- cern negotiated procedures. For instance, in decision 5607/2009 the administration states that precisely because it was following a procedure below the EU thresholds, it did not expect that the formal procedures of the Code, such as opening the bids in public, had to be applied. Similarly, in decision 6839/2013, where an economic operator contested the admission of another participant in the tender, the administration declared that the contract was divided into two lots precisely so as not to incur in the above threshold leg- islation. In decision n. 1873/2016, the administration, which opted for a negotiated procedure for urgency reasons, interpreted the procedure too creatively, imposing exces- sively short time-limits on participants in order to apply. The analysis of litigation shows Unexpected consequences of the implementation of the Code of Public Contracts 48 that in implementing public contract legislation, the administrations of Rome and Turin tend to use more discretional procedures and often the use of discretion is considered illegitimate by the administrative judge. 4. Conclusions This paper analyses the implementation of the Code of Public Contracts in three Italian cities: Milan, Turin and Rome, adopting a theoretical framework which states that pub- lic contracts are designed strategically, by rational actors concerned with reducing the risk of litigation before the Administrative Courts. The local administrations may design public contract awarding procedures using more or less discretion, according to the like- lihood of receiving legal challenges from third parties. By capturing potential for third- party challenges using measures of competitiveness of the economic environment in which municipalities are located, the paper identifies two contexts: one more competi- tive, where potential challenges from the economic competitors are expected to be higher, the other less competitive, where the potential for third-party challenges is ex- pected to be lower. The municipality of Milan belongs to the first context; the municipalities of Rome and Turin to the second. The analysis of judicial proceedings for the awarding of public contracts confirms that the administration of Milan, which be- longs to the first context, tends to use more rigid (less discretional) procedures. Conversely, the administrations of Rome and Turin, which belong to the second context, tend to use more discretional procedures, and often the use of discretion is judged illegit- imate by the administrative judge. In this respect, this paper shows that the Italian municipalities also act strategically when they implement policies and, as positive polit- ical theory suggests, the interactions with other actors and interests can generate unexpected consequences. Surprisingly, the analysis of litigation reveals that the rigid implementation of the Code can also be the cause of defeat for the administrations. This outcome is, however, related to a specific and potentially controversial aspect of the Code, which is the attribution of the responsibility for the control of moral and contrib- utory requirements of the firms to the same local contracting authorities. It is reasonable to argue that such responsibility should be delegated to an independent body, not directly involved in the awarding procedures. References Baltrunaite, Audinga, Giorgiantonio, Cristina, Mocetti, Sauro and Orlando, Tommaso (2018) Discretion and Supplier Selection in Public Procurement. Bank of Italy Temi di Discus- sione (Working Paper) No. 1178. Available at SSRN: https://ssrn.com/abstract=3210748. Brouard, Sylvain and Christoph Hönnige (2017) ‘Constitutional courts as veto players: Les- sons from the United States, France and Germany.’ European Journal of Political Research 56(3): 529-552 Ciccarelli, Andrea (2006) L'articolazione della competitività a livello territoriale, in Del Colle, Enrico Tecnopoli. L'articolazione territoriale della competitività in Italia, FrancoAngeli, Milano, 2006 Coviello, Decio and Stefano Gagliarducci (2010) Building political collusion: Evidence from procurement auctions, IZA Discussion Papers, No. 4939, Institute for the Study of Labor (IZA), Bonn ELISA REBESSI 49 Ferejohn, John and Charles Shipan (1990) Congressional Influence on Bureaucracy, in Jour- nal of Law, Economics & Organization, 6 (Special Issue): 1-20 Ferejohn, John and Weingast Barry R. (1992) A Positive Theory of Statutory Interpretation, in International Review of Law and Economics, 12: 263-279 Gilardi, Fabrizio (2005) The institutional foundations of regulatory capitalism: the diffusion of independent regulatory agencies in Western Europe. The Annals of the American Acad- emy of Political and Social Science 598(1): 84-101 ICLG (2019) Italy: Public Procurement 2019, accessed online on March 24th at https://iclg.com/practice-areas/public-procurement-laws-and-regulations/italy Marcou Gérard (2016) The Impact of EU Law on Local Public Service Provision: Competition and Public Service. In Wollmann Hellmut, Koprić Ivan, and Marcou Gerard (eds.) Public and Social Services in Europe - From Public and Municipal to Private Sector Provision, London: Palgrave Macmillan: 13-26 Marshall, Robert C., Meurer, Michael J. and Jean-François Richard (1994) Curbing agency problems in the procurement process by protest oversight, RAND Journal of Economics 25: 297-318 McCubbins, Matthew and Thomas Schwartz (1984) Congressional Oversight Overlooked: Po- lice Patrols versus Fire Alarms, American Journal of Political Science 28: 165-179 McCubbins, Matthew, Roger, Noll and Barry Weingast (1987) Administrative Procedures as Instruments of Political Control, Journal of Law, Economics and Organization 3: 243-277 McCubbins, Matthew, Roger, Noll and Barry Weingast (1989) Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, Virginia Law Review 75: 431-482 Moretti, Luigi and Paola Valbonesi (2015) Firms’ Qualifications and Subcontracting in Public Procurement: An Empirical Investigation, Journal of Law, Economics, and Organization 31(3): 568-598 Moszoro, Marian and Pablo T. Spiller (2012) Third-party opportunism and the nature of pub- lic contracts. Working Paper 18636, National Bureau of Economic Research OECD (2016) Public procurement – a study on administrative capacity in the EU, accessed online on March 24th at https://ec.europa.eu/regional_policy/en/policy/how/improv- ing-investment/public-procurement/study/#12 Pollit, Christopher (2001) Clarifying convergence: striking similarities and durable differ- ences in public management reform, Public Management Review 3(4): 471-492 Santoni, Michele and Zucchini Francesco (2006) ‘Legislative output and the Constitutional Court in Italy.’ Constitutional Political Economy 17: 165-187 Spiller, Pablo T. (2008) An institutional theory of public contracts: Regulatory implications, NBER Working Paper 14152, National Bureau of Economic Research Steunenberg, Bernard (1997) Courts, Cabinet and Coalition Parties: The Politics of Euthana- sia in a Parliamentary Setting, in British Journal of Political Science, 27: 551-571 Williamson, Oliver E. (1979) Transaction-Cost Economics: The Governance of Contractual Relations, Journal of Law and Economics, 22: 233-261 Wollmann Hellmut, Koprić Ivan, and Marcou Gerard (eds.) (2016) Public and Social Services in Europe - From Public and Municipal to Private Sector Provision, London: Palgrave Macmillan