Chapter 2. Public-private partnerships for infrastructure at the subnational level of government: Opportunities and challenges in France

Stéphane Saussier

This chapter presents a case study of public-private partnerships (PPPs) for infrastructure development in France. The first part of this chapter presents briefly the context for PPPs in France. In the second part, the challenges associated with the development of subnational PPPs are explored by looking at the cases of the city of Caen and Paris. The last section highlights the main findings, challenges and recommendations for foming forward. In particular, it suggests the creation of a a national observatory in charge of collecting data, informing and training local public authorities not only for PPPs, but more broadly for public-private complex infrastructure contracts.



After a short-lived expansion in reaction to the 2008 financial crisis, public investment has experienced a substantial decline in a number of European countries (OECD, 2013). As a natural reaction to these budget cuts, more efficient use of increasingly limited resources is put in the limelight: “Doing more with less” has become the mantra of this decade. Those financial constraints have been translated by a changing role of the government itself. Moving from own production to delegation and externalisation, the public sector has shifted its focus from dealing with the internal bureaucracy to managing relations with external partners through public-private partnerships (PPPs). As a result of this transition, which began before the financial crisis, the typical job description of public managers and bureaucrats has significantly changed over the last decades.

This chapter addresses difficulties related to the decision to go through public-private partnerships. In what follows, we will refer to the OECD’s definition of PPPs as:

long-term contractual agreements between the government and a private partner whereby the latter typically finances and delivers public services using a capital asset (e.g. transport or energy infrastructure, hospital or school buildings). The private party may be tasked with the design, construction, financing, operation, management and delivery of the service for a pre-determined period of time, receiving its compensation from fixed unitary payments or tolls charged to users. (OECD, 2008)

While these new forms of providing public services are mainly chosen as a reaction to limited resources and with the aim of accessing private sector expertise to overcome limited public sector capabilities, this shift created a set of new capacity challenges for the public sector: partial contracting out of government services requires specific types of contracting capacity (Brown and Potoski, 2003). In fact, such capacity might be a necessary condition for success and as such needs to be carefully studied. The question of administrative contracting capability as a determinant of performance has not received enough attention. In order to understand and reconcile the large variation in PPPs and contracting out performance with the theoretical predictions, it appears indispensable to put the spotlight on administrative skills. This is one of the objectives of this report. More specifically, through case studies of specific cities, we put forward several propositions in order to foster opportunities, overcome challenges and create supportive arrangements for PPPs at the subnational level in France.

The first part of this chapter presents briefly the context for PPPs in France. In the second part, the challenges associated with the development of sub-national PPPs are explored by looking at the cases of the city of Caen – a medium-sized city in France – and the city of Paris. As expected, the types of problems encountered by medium-sized and large cities are not the same. In the last section, the main results are presented. It is also highlighted that the new European public procurement and concession directives might deeply affect the rules of the game. These new directives, voted in early 2014, have been translated into national laws – the deadline to translate the Directives was April 2016. These directives, instead of pushing for more central or regional regulation, bet on a decentralised regulation based on more transparency (Saussier et Tirole, 2015). If this might be a good way to foster competition, it will not help reduce capability shortages for some cities to develop PPPs in a near future. One potential avenue to develop such capacities is to create a national observatory in charge of collecting data, informing and training local public authorities not only for PPPs, but more broadly for public-private complex infrastructure contracts.

Country context for subnational PPPs

Infrastructure needs in Europe and in France

Total public investment is a sizeable sum in France, with nearly EUR 80 billion invested in 2014. Most of public investment is done at the subnational level. In 2015, local public administration invested more than EUR 50 billion in infrastructure, accounting for more than 60% of total public investment. This illustrates that a substantial portion of public investments is made at the local level. There is a need to think carefully about the efficiency of local public investments and potential ways to foster it. This is particularly important at times, and in places, where there is a need for high-cost infrastructure investment, for example in transport, as well as to support the delivery of other public services.

Tools for investing in public infrastructure

There are three main tools available to local public authorities for investing in public infrastructure: 1) traditional public procurement; 2) concessions; and 3) availability-based contracts. For the purpose of this chapter, concession contracts and availability-based contracts will be the main focus.

Public procurement

In France, as in most of the world, traditional public procurement is the main tool used for infrastructure investment. According to the French Economic Observatory of Public Procurement of the French Ministry of Finance (OEAP), in 2013, procurement contracts accounted for around EUR 72 billion before tax in France (for some 96 500 contracts). OEAP data, however, only accounts for contracts above a threshold of EUR 90 000. With this in mind, public procurement contracts are believed to amount to about EUR 200 billion a year in France, or 10% of the gross domestic product (GDP), of which around 35% is dedicated to infrastructure.

Concession contracts

Concession contracts, a second tool for investing in infrastructure, have a long history in France, dating back to the 17th century. Through concession contracts, a public legal entity entrusts its investment needs and the management of a public service under its responsible to a public or private entity in return for a payment that depends on the results of the service operation. End users are usually (with the exception of shadow tolls) those who pay the concessionaire.

In France, such contracts are very common. They are employed for mass catering, water and sanitation, district heating, transport, sports facilities, etc. There is no concessions observatory in France (as it is the case for public procurement contracts with OEAP) to inventory ongoing concession projects. However, it is generally estimated that concessions in France generate a volume of business of over EUR 100 billion a year for operators alone (Institut de la Gestion Déléguée, 2011), or around 5% of the GDP (around half of which comes from transport initiatives).

Availability-based contracts

Launched in June 2004, “public-private partnership contracts” – very similar to private finance initiative (PFI) contracts in the United Kingdom – enable a public entity to entrust a company with a project as part of a long-term contract in return for a staggered payment from the public entity conditioned to key performance indicators (availability-based payment contract). It is used for major construction projects (educational establishments, train stations, etc.), urban infrastructure (street lighting, roads, etc.) and even sports and cultural facilities (theatres, stadiums, swimming pools, etc.).

The introduction of this new kind of PPP was primarily designed to help France catch up to other countries, such as the United Kingdom, which has used this type of tool since the early 1990s. While there have been an increasing number of availability-based contracts signed between 2005 and 2012, the total value of these contracts lags well behind the other two public procurement tools. Between 2004 and 2015 around 0.2% of the GDP and about EUR 14 billion in infrastructure investment have been financed through availability-based contracts in France. This amounts to less than 3% of public investment in infrastructure made over this period. The majority of contracts have been signed at the local level (city or urban areas). Contracts signed at the local level are of lower value (the average amount at local level is EUR 28 million compared to an average of EUR 220 million for contracts signed at the state level). According to the French PPP Task Force - Fin Infra, the EUR 3.5 billion spent at the local level over the 2004-15 period was allocated as follows:

  • 41% to sport and leisure infrastructure (mainly stadiums);

  • 23% to education infrastructure (mainly schools);

  • 13% to lighting, car parks, and waste management;

  • 11% to transportation;

  • 7% to information technology; and

  • 5% the renovation of buildings to achieve better energy performance.

The fact that the use of availability-based PPPs in France is largely developed but still small compared to traditional public procurement contracts should not be interpreted as the result of difficulties specific to this type of PPP. Traditional public procurement contracts are considered as the natural way to provide infrastructure, largely because of a French law of 1985, known as “the MOP law”, concerning “public contracting authorities and their relations with private contractors”. In addition, availability-based PPPs are subject to legal constraints (value- for- money reports) that do not exist yet for traditional public procurement contracts and which make them more difficult to use for a public authority.

However, distinguishing among the available tools for investing in public infrastructure is useful. First, because it highlights that PPPs refer to two different ways to finance infrastructure, namely concession and availability-based contracts. Second, because problems associated with PPPs are common to both types. However, availability-based and concession contracts also have also problems that are specific and that may explain why one form of PPPs will be preferred over the other at the subnational level.

Figure 2.1. Amount of investment in infrastructure: French availability-based contracts between 2004 and 2015 (October)

Source: Fin Infra, data updated 8 October 2015.

Figure 2.2. Type of infrastructures financed at the local level (by level of investments) (2015)

Source: Fin Infra

Challenges associated with PPPs in France

What does the theory suggest?

The economics of PPPs is now largely developed (Saussier and Tirole, 2015). Williamson (1976) highlights several fundamental problems associated with public-private partnerships: organising competition for the market is not easy; transaction costs make contracts necessarily incomplete; and switching costs make public authorities who enter contracts vulnerable to ex post contractual opportunism.

One problem local public authorities face is organising a competitive market in order to select the most efficient partner to provide a service (Chong, Huet and Saussier, 2006). This is challenging because the selection process itself may be complex, requiring the specification of a vector of prices for different types of customers, consuming at different times, and for different levels of quality. Moreover, if operators are selected according to price bids, then public authorities are vulnerable to the “winner’s curse”, since the best offer may come from the most “optimistic” operator who unintentionally underestimates production costs or overestimates future revenues. This problem is especially true for concession contracts but not for availability-based contracts because revenues for these do not come from future demand.

Alternatively, public authorities may also need to manage aggressive bidding situations. These arise when prospective operators strategically underestimate production costs or overestimate future revenues in order to win the deal and then provoke renegotiation with a “captive” local public authority in the future (low balling strategy) (Chong, Huet and Saussier, 2006). Significantly as well, public-private partnerships are not immune to collusion and corruption strategies, especially because they usually deal with high-value contracts. This makes such contracts particularly prone to political contestability issues (Spiller, 2009), explaining why they usually appear to be very rigid – unable to be adapted to an evolving environment (see for example the difficulties encountered by English schools in order to change menus after Jamie Oliver’s television show urging for a better food for children (Lawrence and Quarmby, 2005). Renegotiation may also occur because conditions change over the duration of a contract, necessitating an efficient adaptation mechanism for its terms (Estache and Saussier, 2014).

Public authorities also face switching costs in changing suppliers that induce them to stick with an inferior operator that was awarded a franchise. If a public authority switches suppliers, it could face political embarrassment and service interruption, reduced incentives for private parties to invest (fearing early contract termination), and would need to organise a new (costly) auction. On the flip side, these switching costs give firms incentive to renegotiate contracts to obtain higher prices, misrepresent costs, and provide low-quality service (to the extent that this behaviour is not monitored and/or that quality is not perfectly contractible). The problems that stem from imperfect and asymmetric information are even greater if the incumbent creates knowledge-specific capital that gives him/her a cost advantage.

Theory thus suggests that PPPs are plagued by difficulties that are not easily managed by (local) public authorities, who often require resources in order to manage the selection and contract execution processes. The capabilities required to manage PPPs are much more difficult to acquire for subnational governments than traditional procurement capacities.

What do we observe?

Public-private partnerships are not very popular in France – even if, paradoxically, concessions are widely used and have been for a long time. They are regularly criticised by French newspapers as being too expensive (see for example the advice given by the French Competition Authority concerning French concession contracts for highways – Avis 14-A-13, September 2014 – which was widely echoed and commented in the press1) and used in order to circumvent debt constraints.2 In addition, some PPPs have been hit by corruption scandals (see for example the Balardgone case, an availability-based PPP launched in order to build and manage Paris buildings that are welcoming civil servants from the Defence department).3 This is also the case for some public procurement contracts but PPPs are more frequently targeted by the news media. PPPs are also considered a political topic, their use being associated to greater or lesser degrees with political parties or political preferences, evidenced when looking at how availability-based contracts evolved in France (Box 2.1).

Box 2.1. The Caen Regional Multimedia Library: The influence of the state and regional levels

The Caen Regional Multimedia Library, financed through traditional procurement, opened in 2016 to replace the existing, outdated library in Caen. The new library was designed by Rem Koolhaas (who received the Golden Lion Award in the Venice Architecture Biennale in 2010) and is located in the city centre. The project started in 2010 and cost EUR 61 million, but the city will only have to pay EUR 40 million because the project benefits from EUR 21 million in subsidies from the national and regional levels of government. Many people have questioned the social value of such an investment and the choice made to locate the library in the city centre, especially considering recent technological evolutions that are reducing the need for people to physically be at the library to access books, newspapers, etc.

Source: Dagorn, Gary and Élise Delanoë (2014), “La future BMVR est-elle justifiée ?”, Mon Caen, 21 February.

These elements suggest that PPPs are not always well managed by the (local) public authorities. In what follows, we focus on two case studies of cities, one medium-sized and one large, to highlight the difficulties associated with PPPs at the local level, and the solutions that could be implemented to address them.

Subnational PPPs

The case of Caen

The use of PPPs in Caen

Caen is the prefecture of the Calvados department and the capital of the Normandy region. The city has 108 365 inhabitants (as of 2012 national census), while its functional urban area has 420 000 inhabitants, making Caen the largest city in Lower Normandy. The urban area of Caen, in turn, is the second largest in Normandy after that of Rouen, the 21st largest in France.

The city has adopted a mixed approach to the organisation of its public services. Some services are contracted out; the majority of them are publicly managed by the city. Infrastructure needs for the city are globally met. There is a multi-year strategic plan decided at the city level – which is an obligation for French cities. Only very few shortages were mentioned during interviews. This is because the city is mainly responsible for infrastructure investments. With the exception of national and department roads (that are not important for the city of Caen), the city decides unilaterally. When the investment is decided at the urban area level, the city of Caen is the main actor, deciding where and how to invest. There is no issue of mismatch nor overlap with other decision-making levels, nor are there laws or regulations that might interfere at different levels of government. The only interference comes from the fact that some regional investments are subsidised by the state, influencing investment decisions that are taken at the regional level. Some investments that are not a priority might be favoured as soon as the city or the urban area want to benefit from subsides (e.g. the Caen Regional Multimedia Library, Box 2.1.).

Infrastructure needs in Caen are mainly achieved through traditional public procurement, not PPPs (Table 2.1). When concession contracts are signed, they are often lease contracts involving few private investments (i.e. the private partner is mainly involved in the operation phase and may invest for maintenance, but not in the construction phase). According to our interviews, the main reasons why the city decided to develop PPPs were to decrease the delivery time for primary infrastructure and to increase the efficiency during the operation phase. Those motives are in line with previous studies on French PPP contracts. Saussier and Tran (2012) showed that 77% of availability-based contracts delivered infrastructure on time in France. Conditional payments to the private partner are a strong driver in achieving such a good result. In addition, another reason put forward is that when revenue budgets are under pressure, public authorities can decide to make short-term savings by cutting the cost of ongoing maintenance of infrastructure assets. This might lead to increased costs overall as the costs of delayed maintenance can greatly exceed regular, more timely, maintenance. It can also decrease the quality level of the provided services. By contrast, in a PPP, because the private partner is contractually obliged to maintain the infrastructure (and encouraged to do so in concession contracts), the procuring authority is contractually obliged to pay for it and the efficiency during the operation phase is increased.

Table 2.1. Services contracted out in Caen

Services contracted out

Type of contract


Private %

Decision level

Water distribution

Concession (Lease)

Private / Veolia



Peace Memorial

Concession (Lease)

Public-private local public company (LPE)



Equestrian Centre

Concession (Lease)





Concession (Lease)




Exhibition Hall

Concession (Lease)




Heating network


Private - Dalkia



Car parks


Private - Vinci



Zenith – Music Hall


Public-private LPE





Private – Chamber of Commerce and Industry


Urban area






Waste management

Concession (Lease)



Urban area

Court of Justice





IT network




Urban area



Private - Bouygues


Urban area


Concession (Lease)



Urban area

Vehicle pound

Concession (Lease)




Tourism Office

Concession (Lease)



City / Urban area

Source: Recollection by the authors from interviews made at the city level.

Caen’s experience with PPPs has been mixed. An availability-based contract signed by the urban area for the creation of the Caen Hospital Centre proved problematic at early stages. Inadequate public sector capabilities at the outset of the project hampered the implementation of the PPP. The resulting experience partially explains why such contracts have not been used more broadly by the city and highlights how a past experience may have long-lasting effect.

Problems with hospital PPPs are not unique to the urban area of Caen. In its 2014 report, the French Court of Auditors (Cour des Comptes)4 devotes a full chapter to public-private partnerships (PPPs) for hospitals. The Court wrote, “these proceedings were launched precipitously, the benefits that are attributed to PPPs were poorly exploited and financial issues have been insufficiently taken into account.” The report pointed out that 24 hospitals were conducted in PPPs for an investment of EUR 613 million and that there was no real choice made by public authorities to go through PPPs or other procedures. PPPs have been “deployed in haste without the accompanying legal tools and management institutions are sufficiently available.”

Concerning the choice of the PPP, the Court regrets that “no prior study has actually been required, neither on the financial implications of the project, nor in terms of comparison with the public project contracting procedure.” The Court believes that “the lack of sufficient preparation for this new procedure, inexperience and the negotiators largely oriented assistance for PPPs led to the selection of disparate operations and the signing of contracts holders of financial uncertainty.”

Concerning the contractual side, safeguard contractual provisions to ensure the continuity of public service have been incorporated into many contracts with a set of financial penalties applicable in case of unavailability of each function essential to the hospital. But the Court pointed out that “some are not sufficient,” and “contracts reviewed did not sufficiently cover the diversity of conflicts that can arise in a period of 18-30 years […] A balance must be struck between the stability of the contract that brings financial security and the necessary amendments to the evolution of medical practices. This is one source of the inherent complexity of hospital PPPs.” This balance is not easy to find for public authorities that lack contractual capabilities.

The difficulties encountered in availability-based contracts arising from a lack of contractual capabilities on the part of local authorities are also encountered in concession contracts, as the Caen tram network example illustrates (Box 2.2).

Box 2.2. The Caen guided light transit (Tram)

The Caen guided light transit (Tram) project involved 24 three-section articulated vehicles, guided by a central non-supporting rail, to provide service. The entire passenger line is guided, and in normal service, the vehicles are powered by electricity drawn from an overhead wire through a pantograph. After a construction period of three years, the system opened on 18 November 2002 with a total cost of EUR 227 million.

This project was organised through two 30-year concession contracts: one for the construction part of the project and one for the service part of the project. Many problems appeared during the execution phase of the contract because the two private partners took advantage of their informational advantage compared to the public authority. As soon as unexpected events/performance issues arose, it was impossible to clearly delineate which partner was responsible. With three major breakdowns per day, a rate 32 times higher than the contractual objective, above normal maintenance costs of EUR 1 million per year and penalties of EUR 1.5 million paid by both concessionaires for nine years, it is fair to say that the tram did not bring the expected results.

Due to its unreliability, the Caen urban area confirmed its plans to abandon the Caen TVR in favour of light rail by 2018. The light rail is set to take 18 months to construct and has an approximately EUR 170 million price tag. The conversion to light rail also means the termination of two concession contracts that Keolis and Bombardier-Spie Batignolles consortium STVR hold. Faced with possible appeals, Viacités – the transport union of the Caen urban area – will also apply to the administrative court for the appointment of an expert, in order to establish the level of responsibility of the STVR. In late 2014, the French government pledged EUR 23.3 million towards Caen’s light rail conversion project, which is now expected to cost approximately EUR 230 million.

Source: Leguel, P. and P. Ambrosi (2011), “Caen abandonne l’exploitation du tramway sur pneus avant son terme”, Les Echos, 19 December; Tramways and Urban Transit, January 2003; Ian Allan Publishing/Light Rail Transit Association.

What would facilitate the development of PPPs in Caen?

Despite being a dynamic French city, Caen, like other cities, has limited familiarity with PPPs. Having no specialised personnel for PPPs, the city is characterised by a lack of expertise and capabilities to manage them, from the award procedure to the end of the contract execution. Even if some examples of success in PPPs are cited during interviews - for example the case of the Caen Memorial Museum (Box 2.3) – those successes are not easy and necessitate a real human-capital investment from the city, in order to negotiate and follow the contract, sometimes renegotiating it in order to improve its efficiency like in the case of the Caen Memorial Museum (for which renegotiations were easy to manage because of the involved partner), cf. Table 2.1.

Box 2.3. The Caen Memorial Museum, Centre for History and Peace

In 1988, the Caen Memorial Museum, Centre for History and Peace, opened its doors. The infrastructure has been entirely built using public funds but the operating expenses (renewal of exploitation material and equipment) have been borne by a semi-public company (Société d’Economie Mixte, SEM) of which the city of Caen is the majority shareholder.

Because of a decline in the number of visitors, the city decided in 1997 to invest in the memorial extension to allow for a diversification of the themes addressed. However, anticipating that this new investment may generate generous extra revenue for the SEM, in 2002, the city has renegotiated the contract to compel the SEM to pay a rent to the city for occupying the place, just before the extension was open to the public.

In its 2014 annual report, the Cour des Comptes criticised the way the city of Caen subsidised the SEM. When looking at the financial reports from 1996 to 2001 (except in 1997), the SEM achieved a negative operating result each year, and was only positively balanced due to extraordinary results originating from the granted subsidies. There were two main sources of subsidies namely, a compensation for the gratuity of cultural services and discount rate for specific groups of individuals (e.g. students, seniors), and the renewal of equipment. Two main reproaches were made regarding the way the discount rates were compensated. First, it was based on the lower range of the expected number of visitors and therefore could leave some rents to the SEM. Second, the difference between the compensated tariff and the full-rate tariff was so low that the SEM did not get the proper incentives to make efforts to attract full-rate visitors, and it was also against the principle of reporting the risk of demand to the SEM. Also, the city subsidised spending that should have technically been imputed to the SEM (e.g. the national communication campaign, museum curator).

However, it is important to keep in mind that the Caen Memorial Museum is a cultural institution and is still an example of good management in the sector. Indeed, even though the amount of subsidies was important, it was relatively low compared to similar museums and its level was constantly declining before the extension opening. In addition, the revenue generated by the museum was also relatively high for the cultural sector. It is also worth noting that renegotiation could be favourable to the public entity, as was the case when the city of Caen decided to collect a rent for the usage of the installation after the extension of the museum was realised.

Source: Various sources, including interviews and Cour des Comptes (2014), Rapport public annuel de la Cour des comptes - 2014.

Capabilities needed by the city to run public services are different from those required to manage PPPs. Such capabilities are not easy to develop, especially in a short period. Public authorities can learn from accumulating experiences over time and through various situations. That is why what frequently comes up when discussing such issues with public authorities is the need for a national or regional observatory5 accumulating data and experience in order to diffuse information about what is working and what is not working well with PPPs. This observatory could also support cities in deciding whether or not to opt for PPPs, in the implementation of the call for bids, as well as in the contractual details. However, our interviews suggest that such an observatory should not be considered as a third party involved in the city’s decision to invest because cities are independent and feel strongly about their independence. That is why such third-party involvement could be more easily implemented during the execution stage (which is, theoretically, an important stage where renegotiations take place and might be very costly) and might be welcomed for complex infrastructure projects.

Interviews also suggest that more clear guidance from the national government and a stable regulatory environment regarding the use of PPPs are essential. More broadly, at the European level, discussions concerning the new European Directive on Concessions illustrate the divergence among European countries on this issue (Box 1.1).

The case of Paris

The use of PPPs in Paris

The case of Paris is interesting because, unlike Caen, the city is familiar with PPP contracts and has a dedicated staff unit for such PPPs. The city is most familiar with concession contracts, with 182 that are ongoing. In 2013, they represented a turnover of EUR 1.5 billion. Some of those contracts are signed with local public enterprises (LPEs). LPEs are mainly present for car parks, funerary services, district heating, cold energy, Eiffel Tower management as well as the management of the POPB (Paris-Bercy Stadium). Concessions managed by LPEs represent a global turnover of EUR 572 million (one-third of concessions total turnover). Figure 2.3. shows the repartition of activities for which concessions were used in Paris.

Figure 2.3. Concessions turnover in Paris City, 2013
EUR millions

Source: Based on Paris City document 2015-DFA-79.

The city of Paris’ experience with availability-based contracts is considerably more limited: only one has been signed by the city. This single contract is part of the city’s Climate Plan adopted in 2007. The city of Paris is committed to reducing the energy consumption and greenhouse gas emissions of its public buildings by 30% before 2020. This commitment involves the refurbishing of 600 schools in Paris. A first group of 100 schools was selected for renovation through an availability-based contract. With nearly EUR 50 million in investment, it is one of France’s largest energy renovation programmes. It was launched in 2011, and by 2014, performance goals were achieved for 45 of the 100 schools targeted for the first round of work performed, with a 33% reduction of energy consumption and 34% of greenhouse gas emissions. The results exceeded expected objectives (Khalid, 2014). However, the city decided to continue refurbishing the remaining 500 schools using traditional procurement contracts instead of availability-based PPPs. This is because of: 1) a new French legislation opening the path for global public procurement contract linking investments and service provision in a single contract (which was not allowed before 2011); and 2) the legislative complexity of using availability-based contracts. The difference between both types of contractual agreements is mainly derived from the fact that the payment is delayed in an availability-based contract and not allowed in a (global) public procurement contract.

Infrastructure needs and main reasons to use PPPs in Paris

The case of Paris is interesting because, contrary to that of Caen, the city is familiar with concession contracts and has staff dedicated to such activity. However, even with such experience, the city is going back and forth in some activities (e.g. water) and is experiencing difficulties regardless of the type of operator: private or semi-public (see Box 2.4, Box 2.5 and Box 2.6). This shows that concessions with private partners as well as with semi-public concessionaires have their problems. Some are common (e.g. difficulty in controlling the concessionaire and receiving data). Others seem more specific to semi-public concessionaires, such as the low level of effort placed in adapting to demand, as well as financial difficulties.

Box 2.4. The case of the Vélib’

The Vélib’ case is an interesting illustration of the pitfalls associated with public-private contracting highlighted by contract theories (Williamson, 1976; Hart, Shleifer and Vishny, 1997, Laffont and Tirole, 1993). In 2006, the city of Paris decided to provide a bicycle-sharing programme to its citizens – the Vélib’. In exchange for city billboards, private companies bid in order to provide the bicycle service. The call for bids was difficult to organise. In a first round, Clear Channel won, offering 14 000 bikes instead of the 7 500 bikes proposed by its competitor JC Decaux. However, after a judicial appeal, the city was obliged to reorganise a call for bids. The second round was won by JC Decaux, which offered 20 00 bikes. An investment of more than EUR 80 million was planned over the ten-year contract period.

During the execution stage of the contract, it became clear that the contract was not adapted and renegotiations occurred. Penalties for late deployment of the bikes were not applied, as is often the case for public-private contracts (Saussier and Tirole, 2015). Performance indicators were badly chosen: for example, the private company was responsible for monitoring the bike network to ensure that a user would find at least one bike at each station. One way to reach this performance level was for JC Decaux to leave a broken bike in each station (see CRC report, 2012). On top of this, JC Decaux had no financial incentive to provide a service of good quality because the user fees were going back to the city. Finally, vandalism was underestimated in the initial contract (61% of the bike fleet deteriorated during the first year of operation) leading the city to accept a renegotiation in 2007 and again 2009, paying EUR 400 per stolen bike over a threshold since 2009. An interesting detail illustrating asymmetric information issues is that the city found out, two years after accepting to renegotiate that more than 70% of the stolen bikes were found by the company and put back into the system, leading the city to pay several times for the same bike.

Renegotiations that occurred also changed the nature of the initial contract by providing incentives to JC Decaux to increase the number of users of the Vélib’. After 2009, the company kept 35% of the net turnover as soon as it went over EUR 14 million and 50% as soon as it reached more than EUR 17.5 million. The initial public procurement contract moved toward a concession contract where the private company was paid directly by the users.

The Vélib’ contract ended in 2017 and the city re-opened a call for tender. This time the city opted for a concession contract instead of a public procurement contract.

Source: Rapport de la Chambre Régionale des Comptes (CRC) d’Ile-de-France – June 2012; Inspection Générale Ville de Paris, Audit du contrat Velib, 2016; Les Echos, 12 April 2017.

Box 2.5. Car park concessions in Paris

Private operator

In 2010, the city of Paris conducted an audit of car parks allocated through concession contracts to the private operator Vinci Park. This is main concessionaire in Paris, with more than 90 parks under its responsibility for a total turnover of EUR 117 million per year (out of the EUR 175 million generated for all car parks organised through concessions in Paris). The report concludes that there is a lack of co-ordination between the city and Vinci Park; that information is asymmetric in favour of Vinci; and that concession contracts appear to be too rigid for the city to adapt to its strategy (i.e. the environmental strategy of the city would be strengthened by an increase of car park services but such an increase would favour Vinci Park too much as they were not anticipated in the long-term contracts). The main conclusion of the report is that the city, in order to recover its control over the public service, should revert to direct public management by not renewing the concession contracts when they expire.

The report stated, “The Audit recommends more exchanges between the City and the private operator to allow the City to ensure a better control of its parks and greater mastery of the pricing policy. The audit has indeed led to identify the existence in some parks entrusted to Vinci Park specific price schemes that deserved to be brought to the attention of the City.” This exemplifies the fact that concessions are incomplete contracts and that the private operator might decide some prices (toward specific clients) not envisioned by the city.

The report also stated the need to renegotiate “oldest car parks concession contracts in the inner city. Higher hourly rates can only take place at the initiative of the public authorities. They would represent, in the current situation, a financial advantage for the private operator that is difficult to justify. The new pricing policy implies the renegotiation of the financial terms of concession contracts of the oldest parks in the inner city.” This exemplifies the fact that concession contracts are incomplete but also too rigid, as they impede Paris from ensuring that car park prices evolve as the city wishes without costly renegotiations with the private operator.

Semi-private company

In 2013, the city of Paris audited car parks allocated through concession contracts to a local semi-public enterprise: the SAEMES. The SAEMES is in charge of 41 car parks in Paris, corresponding to 21% of parking places in the city. The report states that the profitability of the SAEMES is burdened by significant investments that are generating little revenues. Some 18 of 41 parks are structurally in loss, among which 6 have a negative gross operating profit. This contrasts with car parks operated by Vinci Park (above) for which 9 of 66 contracts appear structurally in loss including 5 contracts generating a negative gross operating profit. The report also states that SAEMES lacks the necessary tools to anticipate risks and to measure the impact of new management and financial decisions.

The report notes that the city would like to control the SAEMES’ investment renewal programme, in particular to know the number of failures recorded in the SAEMES’ car parks and their impact in terms of discomfort or unavailability. However SAEMES considers direct consultation of its computerised maintenance management software by the city as incompatible with the principle of a concession and refuses to let the city access it.

The report also found that in some parks the system that displays information did not work and that in general the company had little or limited equipment that was up to date technologically (plasma information screens, etc.). More broadly, few efforts were made to develop the brand name of the SAEMES and to respond to customer complaints.

Finally, since 2003 all maintenance projects have run behind the schedule set in the initial agreements. Neither the time nor the costs estimated during the bidding phase have been met. While the slippage of costs is only detrimental to SAEMES and its shareholders – at least in the short term – in the long run renegotiations may occur, passing costs to user fees - missing deadlines also penalises the users.

Sources: Audit des DSP des Parcs de stationnement confiées à la société Vinci – Inspection Générale de la Ville de Paris, 2010 ( -generale-183#audit-des-dsp-des-parcs-de-stationnement-confiees-a-la-societe-vinci); Audit du Train de vie de la SAEMES – Inspection Générale de la Ville de Paris, 2013 ( cipalite/l-hotel-de-ville/evaluation-inspection-generale-183#audit-du-train-de-vie-de-la-saemes).

Interestingly, even if the city of Paris has experience with respect to concessions, some difficulties still occur – as illustrated by the case of Ternes car park (Box 2.6).

Box 2.6. The case of the Ternes car park

In February 2002, the city decided to renew its concession contract for the “Ternes” car park. The winner, Omniparc, a society belonging to Eiffage, one of the major players in the field (together with Vinci), won the 12-year-long contract. The contract mentioned the need to renovate and extend the capacity of the car park (280 parking spaces in addition to the 1 336 existing ones), with a cost of more than EUR 13 million (EUR 8 million for the extension) for the new concessionaire.

Work was supposed to start within 18 months of the contract’s signature. The concessionaire began renovating the park and argued that it cost EUR 8 million instead of the EUR 5 million initially anticipated. The concessionaire refused to invest more in order to extend the capacity of the park contrary to the contractual agreement. The company asked for a renegotiation of the initial contract arguing that costs were underestimated and needed to be increase to EUR 15 million instead of the EUR 8 million initially anticipated. The reasons invoked were that new security legislation showed up and induced the need to amortise investment in a shorter period than was expected. The city refused to renegotiate and to pay more than initially agreed.

In October 2008, the city ordered the concessionaire to start the contractually agreed project. When the concessionaire did not do so, the city terminated the contract (in 2009), without any compensation for the concessionaire. The concessionaire sued the city, requesting that the termination be cancelled, arguing that investment had been underestimated and that the city should not have accepted to sign a contract with such an unreasonably low offer. On its side, the city asked for EUR 12 million as indemnities for damages due to bad quality service over the period the concessionaire operated the car park. The administrative court ruled in favour of the city in June 2011, but only required the concessionaire to pay EUR 27 000 in damages to the city. The car park is now operated through a concession contract with another concessionaire (Autocité – Spie batignolles).

Source: Saussier (2017) Public-private partnerships for infrastructure at the subnational level of government: Opportunities and challenges in France and

Several points are worth noting in the Ternes car park case. First, it illustrates the now well-known low-balling strategy used by concessionaires to bid very aggressively, anticipating that they will succeed in renegotiating the initial contract to their advantage ex post. Second, such conflicts are usually rare as the public authorities often accept to renegotiate or the concessionaire accepts not to renegotiate in order to preserve its reputation (sometimes by reducing costs and decreasing quality). This was not the case here, probably because of a change in Omniparc ownership in 2008, when it was purchased by Q-Park. This company was not at the origin of the offer and this might explain their reluctance to invest so much money in the park. Another explanation is also that the city of Paris is more experienced than other public authorities and can be credible when saying there will be no renegotiation (however the Vélib’case contradicts this view; see Box 2.4).

However, as is the case for the city of Caen, it appears that PPPs require that cities invest in and develop their capabilities to manage every step of the agreement, from the award procedure to the execution stage. Such difficulties, as well as the fact that PPPs are susceptible to challenge by political opponents, may explain why the city questions the value of PPPs, going back and forth sometimes – for water services for example. This implies the need to involve all the stakeholders at every step of the PPP process to avoid third-party opportunism(Spiller, 2009), which can take the form of political challenges that destabilise PPPs. The new governance for PPPs that was put in place in Paris at the end of 2014 can be considered an organisational innovation designed to cope with these issues (Box 2.7).

Box 2.7. A new governance for public-private partnerships in Paris

In September 2014, the city of Paris set up new governance bodies for concessions with high stakes: the Board of Concessions and the Commission of Elected People.

The Board of Concessions is a concession steering body responsible for defining the strategic objectives, upstream of award procedures but also at the renewal stage of emblematic or strategic city’s concessions. Its remit covers the concessions at large. It includes public procurement contracts with forgone revenues (e.g. Vélib’ or Paris’ kiosks1) and availability-based contracts. Its composition has a variable geometry but it comprises all stakeholders and decision makers concerned with the agenda: alongside the Secretary General of the Directorate of Finance and Procurement and Legal Affairs Department, the contract managers’ directions, sometimes some elected officials are invited to participate. The board thus constitutes a prior arbitration body and defines a strategy with a must at key stages of the contracts’ procedure.

The Commission of Elected People consists of nine elected permanent members. Members are carefully chosen, notably in order to represent the elected representatives of the political opposition. The commission meets to set up some guidelines in the case of procurement procedures or particular strategic or symbolic contract renewals. Meetings in this commission anticipate the ones of the Paris Council since the Commission considers well before awarding contracts, the project, the proposed management methods, terms of competition and the award criteria.

It is expected that these two bodies will address those strategic issues for contracts that are not necessarily with significant stakes in financial terms. The final objectives of these bodies are essentially to:

  • Increase transparency in the management of public services;

  • Modernise and secure the award procedures and implementation of public-private partnerships around a “standard city”, with the aim of drafting a guide for the main financial provisions being found in all concession contracts; and

  • Involve elected upstream of the decision and contract renewals.

The set-up of this new governance illustrates (political) difficulties encountered by the city concerning the development of PPPs. However, since this new governance system has been in place, little communication has been undertaken by the city concerning its decisions and its efficiency.

Note: Public procurement contracts with forgone revenues are contractual arrangements that are similar to concessions. For example, the Velib’ is a traditional procurement contract signed between the city and JC Decaux (see Box 2.4). In exchange for investing and operating the service, JC Decaux received the right to operate for free the city’s billboards, for which the city foregoes revenues.

Source: Saussier (2017) Public-private partnerships for infrastructure at the subnational level of government: Opportunities and challenges in France and


PPPs are long-term (incomplete) contractual agreements. As such, they are plagued by transaction costs. Those costs depend on several elements:

  • Contractual choices: Cities need to develop contractual capabilities in order to properly design their contracts, knowing that they face private companies that are used to contracts and that the devil is often in the detail (Brown and Potoski, 2003);

  • Governance choices: Cities need to think carefully about the kind of PPPs that is appropriate for their infrastructure and service needs. Availability-based and concession contracts are sometimes alternative solutions but should be evaluated before signature and confronted as well with the traditional procurement solution (this last point is often forgotten); and

  • Institutional frameworks: Cities need to make their choices in a secured and stable environment. Even better, this environment can provide help for them to develop or have access to relevant capabilities in order to sign successful PPPs.

PPPs are costly partnerships with a political dimension

The two cases presented here – Caen and Paris – offer several lessons. The first lesson is that PPPs are costly partnerships. PPPs can be associated with expectations that responsibilities/agreements will be met. However, cities must be aware that these expectations come with a cost. Because PPPs are long-term contracts, they are also characterised by transaction costs (Williamson, 1985) that the city should try to minimise. This is not an easy process. Asymmetric information exists between public and private parties, often to the advantage of private ones. In addition, and reinforcing this position of weakness, public authorities usually lack contractual capabilities because they differ from those needed to run public services through traditional procurement. Such difficulties are not insurmountable. As illustrated by the case of Aquanova (Box 2.8), even a small city can develop a new infrastructure through PPPs at the cost of investing a lot in the contractual process and details.

Box 2.8. The case of Aquanova: Challenges in the implementation of an availability-based contract in a small city

Aquanova is an aquatic centre financed through an availability-based contract at Saint-Dié-des-Vosges, a small city of 21 000 inhabitants. This project illustrates that even a small city can use PPPs but it has to be supported by strong political will and significant administrative investment. The EUR 18.5 million project (EUR 12 million investments) started in June 2012 and was completed in early January 2014. At the beginning, the city considered using a concession contract to finance the project, but after a year of discussions, it opted for an availability-based contract. The result was an aquatic centre offering multiple new services compared to the previous swimming pool but for the same price. It groups multiple services (e.g. swimming pool, diving pool, massages, fitness centre) with sustainable development objectives in a region where household revenues are low. Some 82% of the revenues generated by this project are subject to added value taxes, generating substantial revenue for the city. Thus far, the project has been considered a success.

Source: Interviews and Vosges Matin (2016), “Les succès d’AquaNova America”, 20 November,

Even when the city developed such capabilities, because PPPs are complex contracts, they come at a cost, opening the door to criticism. Such objections usually forget that alternative arrangements (i.e. traditional procurement) are not immune to other limits – not studied in this chapter – potentially even more costly for cities. Nevertheless, they might destabilise the willingness of the city to develop PPPs.

In addition, PPPs are characterised by the importance of the institutional environment. PPP legislation is not considered as secured by parties: availability-based contracts were launched in June 2004 and as such constitute an organisational innovation. Since then, a new law modified the legislation around those PPPs in July 2008 and the legislation evolved again with the translation into national law of the public procurement European Directive 2014/24/UE.6 These evolutions do not provide a stable institutional environment for contracting parties.

Lastly, PPPs are public-private agreements, characterised by the importance of the political dimension. As pointed out by Spiller (2009), “A fundamental difference between private and public contracts is that public contracts are in the public sphere, and thus, although politics is normally not necessary to understand private contracting, it becomes fundamental to understanding public contracting.” Hence, a large part of the control is done by third parties (e.g. stakeholders, consumer associations, political contesters) that are not necessarily interested in the success of the agreement and that might look to destabilise it (Beuve, Moszoro, and Saussier, 2015) and the political agenda is an issue that might destabilise PPPs, especially around election times (Le Squeren, 2016).

The need for public contracting capacity

One of the main avenues for developing PPPs is to provide cities with the knowledge needed to cope with PPP challenges. Because PPPs are considered for many cities as an “organisational innovation”, many of them are reluctant to develop them, waiting for others to start.

One way to provide cities with the needed knowledge and capability would be to put in place a national observatory in charge of collecting data on PPPs signed all over the country as well as traditional public procurement contracts. This observatory could advise cities in their choices of organisational arrangement for their infrastructures and could also be in charge of following PPPs’ performances once they are signed. In addition to strengthening capabilities, such an observatory could increase the level of transparency around PPPs and consequently the accountability of public decision makers.

In their note to the French government, Saussier and Tirole (2015) emphasised that a greater monitoring of public expenses is needed. They suggest that it could take the form, for high amount contracts whatever their form (i.e. traditional public procurement, concession or availability-based contracts), of a prior evaluation systematically performed, including the full cost and anticipated advantages for each project. These prior evaluations would make it possible to identify the most appropriate form of contract to meet the public need. They also suggest that such an observatory could also collect information about the contract execution stage, especially renegotiations. This point seems particularly important with the translation of European Directives in early 2016, allowing large renegotiations without being clear about how to monitor them. In addition to the relevant players being held accountable, such information would make it possible to compare the conditions under which contracts are executed, along with the conclusions drawn from studies performed at the pre-contractual stage and that have resulted in one method of procurement being chosen over another. Such evaluations should be carried out by a national observatory, which would also facilitate the centralisation of data and results and the possibility of comparing the best contractual practices.

Such a collection of data and information concerning PPPs and traditional public procurement contracts signed in France would also allow for an objective picture of PPPs’ performances, helping to reduce political contestability and to stabilise the institutional framework. Ideally, it should be extended to complex traditional public procurement contracts. The need for such an observatory is suggested by the European Directives on public procurement and concession contracts.

A PPP task force, the Mission d’appui aux PPP (MAPPP), was created in 2004 but had been limited to availability-based contracts. Since April 2016, this task force has been replaced by another, the Mission d’appui au Financement des Infrastructures (Fin Infra) with more responsibilities. In particular, its mission was extended to every kind of complex public private contracts. The objective is to standardise market practices by spreading the knowledge of adequate allocation of risks for bankable projects among public clients. For such an observatory to succeed, a considerable budget will be needed for these objectives to be fulfilled.


Beuve, Jean, Marian W. Moszoro and Stéphane Saussier (2015), “Political Contestability and Contract Rigidity: An Analysis of Procurement Contracts”, SSRN Scholarly Paper, Social Science Research Network, Rochester, NY,

Brown, Trevor L. and Matthew Potoski (2003), “Contract–management capacity in municipal and county governments”, Public Administration Review 63(2), pp. 153–164.

Chong, Huet and Stéphane Saussier (2006), “Auctions, ex post competition and prices: The efficiency of public private partnerships”, Annuals of Public and Cooperative Economics.

Cour des Comptes (2014), Rapport public annuel de la Cour des comptes - 2014.

Dagorn, Gary and Élise Delanoë (2014), “La future BMVR est-elle justifiée ?”, Mon Caen, 21 February.

Estache, Antonio and Stéphane Saussier (2014), “public-private partnerships and efficiency: A short assessment”, CESifo DICE Report 12(3), pp. 8-13.

Hart, Oliver, Andrei Shleifer and Robert W. Vishny (1997), ”The proper scope of government: Theory and an application to prisons”, Quarterly Journal of Economics 112(4), pp. 1 127–1 161.

Institu pour la Gestion Déléguée (2011), Aspects économiques et comptables des investissements dans les PPP.

Khalid, Nohmana (2014), “La rénovation énergétique des 500 écoles parisiennes ne se fera pas en PPP”,, 31 October,

Laffont, Jean-Jacques and Jean Tirole (1993), A Theory of Incentives in Procurement and Regulation, MIT Press, Cambridge, MA.

Lawrence, Felicity and Katharine Quarmby (2005), “Private deals block Jamie's school dinners”, The Guardian, 25 April,

Leguel, P. and P. Ambrosi (2011), “Caen abandonne l’exploitation du tramway sur pneus avant son terme”, Les Echos, 19 December.

Le Monde and AFP (2014), “Affaire du « Pentagone français » : deux hommes, dont un militaire, mis en examen”,

Le Squeren, Zoé (2016),”Politics and Public Administration: The Influence of Electoral Motives and Ideology on the Management of Local Public Services”, thesis.

OECD (2013), Investing Together: Working Effectively across Levels of Government, OECD Publishing, Paris,

OECD (2008) Public-Private Partnerships: In Pursuit of Risk Sharing and Value for Money, OECD Publishing, Paris,

Saussier, Stephane and Jean Tirole (2015), “Strengthening the Efficiency of Public Procurement”, Les notes du conseil d’analyse économique, No. 22, April,

Saussier, Stéphane and Phuong Tra Tran (2012), ”L’efficacité des contrats de partenariat en France : une première évaluation quantitative”, Revue d’économie industrielle (140), pp. 81-110.

Spiller, Pablo T. (2009), “An institutional theory of public contracts: Regulatory implications”, in Ménard, Claude and Michel Ghertman (eds.), Regulation, Deregulation, Reregulation: Institutional Perspectives, Edward Elgar.

Vosges Matin (2016), “Les succès d’AquaNova America”, 20 November,

Williamson, Oliver E. (1985), The Economic Institutions of Capitalism, The Free Press, New York, NY.

Williamson, Oliver E. (1976), “Franchise bidding for natural monopolies: In general and with respect to CATV”, Bell Journal of Economics 7(1), pp. 73–104.


← 1. See:

← 2. See, for example, the report from two French Senators - Les contrats de partenariats: Des bombes à retardement? - Rapport d’information n° 733 (2013-2014) by Mr. Jean-Pierre Sueur and Mr. Hugues Portelli, 16 July 2014, available at

← 3. See, for example, Le Monde and AFP (2014).

← 4. The Court of Auditors is a quasi-judicial body of the French government charged with conducting financial and legislative audits of most public institutions and some private institutions.

← 5. The number of French regions was reduced from 22 to 13 regions in France since January 2016. This may justify the creation of regional observatories that would benefit from scale economies and would be easier to create and manage with fewer regions.

← 6. Concession contracts are also touched by the translation in national law of the concession European Directive 2014/23/UE, but more lightly (see Saussier and Tirole, 2015).

End of the section – Back to iLibrary publication page