Regional heterogeneity in Italy: transport devolution and corruption

Contents lists available atScienceDirect

Land Use Policy

journal homepage:www.elsevier.com/locate/landusepol

Regional heterogeneity in Italy: Transport, devolution and corruption

Fabio Carlucci

, Andrea Cirà

, Giovanni Immordino

, Giuseppe Ioppolo

, Tan Yigitcanlar

b_{Department of Economics, University of Messina, Piazza Pugliatti, 1, 98122 Messina, Italy} c_{Department of Economics and Statistics, University of Naples, Via Cintia, 21, 80126 Naples, Italy}

d_{School of Civil Engineering and Built Environment, Queensland University of Technology, 2 George Street, Brisbane, QLD 4001, Australia}

A R T I C L E I N F O

A B S T R A C T

In the transport related literature little attention has been devoted to the institutional aspects involved in the strategic interaction between the central government and its local bodies. This might seem surprising given the relevance of the problem. This paper analyses the impact of the devolution of administrative powers on the transport sector. Specifically, we review some recent developments in the policies adopted in the transport sector by several advanced countries. Then, we analyze the relationship between mobility and devolution from a pure efficiency perspective. We describe the Italian devolution experience and present the latest data on corruption and per capita income to testify the heterogeneity among Italian regions. The main insight of our Tieboutian's framework is that decentralization is more sensible whenever there exists a lot of heterogeneity among regions in terms of per capita income and in the corruptibility of public officials, as this is the case for Italy. However, the Italian decentralization process might have been slow-downed due to fairness considerations.

1. Introduction and background

A strand of the public economics literature has emphasized the role of the devolution of power in providing more market-oriented services and an increase in productive efficiency, in the transport sector (Oates, 1972, 1999; Canning et al., 2010). The public goods literature lays out a theory that determines the best solution for the assignment of functions to different levels of government in specific sectors of the public finance. In a seminal paper, Tiebout (1956)argued that local public finance leads to a more efficient supply of local services, to an improved allocative efficiency, and reflects social preferences more adequately. Indeed, the national provision of local public goods is not efficient and this may be true for the public transport sector.

The assumptions of Tiebout's model, defined “extreme” by its author, are very restrictive. While under certain circumstances, compe-tition among local governments might lead to an efficient allocation of local public goods, a possibility which requires empirical verification (Bewley, 1981; Liu and Zhang, 2013), the few analysis on the benefits of political decentralization in terms of efficiency do not provide a

clear-cut answer (Balaguer-Coll et al., 2010). Economists commonly perceive the economic relevance of decentralization.1 _{For example,}

regarding development planning,de Valk (1990)observes that decen-tralization appears less a political goal and more an instrument to increase effectiveness and efficiency. Nevertheless, the evaluation of economic effects of this political process can be difficult.

Seabright (1996) examines devolution process not only from a political point of view but also from a strictly economic perspective. He focuses on the trade-off between policy coordination and accountabil-ity, defined as “the probability that the welfare of a given region can determine the re-election of the government”.2_{In a centralized political}

system, there is the highest possible coordination, while the level of accountability is low; conversely, decentralization of powers to local governments diminishes coordination but leads to a higher standard of accountability. Finally, regional devolution stands at an intermediate position in terms of both coordination and accountability. The scope of the devolution determines the size of the electorate in charge of appointing or rejecting a government: the higher is the level of centralization, the lower the control of local communities on

politi-http://dx.doi.org/10.1016/j.landusepol.2017.04.020

Received 9 February 2017; Received in revised form 27 March 2017; Accepted 14 April 2017

⁎_{Corresponding author.}

E-mail addresses:fcarlucci@unisa.it(F. Carlucci),acira@unime.it(A. Cirà),giovanni.immordino@unina.it(G. Immordino),giuseppe.ioppolo@unime.it(G. Ioppolo), tan.yigitcanlar@qut.edu.au(T. Yigitcanlar).

1_{We refer to decentralization as the transfer of powers and responsibilities from central government to regional and local governments. Thus, in this paper, we use the words}

“decentralization” and “devolution” interchangeably (seeLeone and Carroll, 2010).

2_{It is worth noting that the interests of political representatives do not coincide automatically with those of citizens (Barro, 1970).}

Available online 02 May 2017

0264-8377/ © 2017 Elsevier Ltd. All rights reserved.

cians. This might mean that the central government is not motivated to implement policies that would increase the welfare of marginal localities that is of citizens that do not have the power to affect the government's re-election. The relationship between accountability and local welfare is an important connection between the political and the economic implications of devolution.

From an economic point of view,Seabright (1996)approaches the question of devolution underlining the important role played by the externalities of local policies on other localities. Where these external-ities are negative, that is in case of competitive local policies, decentralization increases population welfare; whereas if there are positive spillovers between localities, that is in case of complementary local policies, centralization increases population welfare. It is also important to underline how the government perceives the geographical context at local scale, the degree of criticality of the areas for intervention, ex-ante, from an environmental, social, economic and infrastructure viewpoint, the established functional mix and, finally, the relationships between objectives of conservation and those of finding new forms of sustainable consumption as a whole (Ioppolo et al., 2013).

Alesina and Spoalore (1997) focus on a trade-off between the benefits of centralization, in terms of economies of scale, and the “political costs” of heterogeneity of population that is difficult to satisfy for a centralized government. Moreover, the economic efficiency of devolution requires the full coincidence of the costs and benefits of the decisions taken by the political institutions (Haynes et al., 2005). This means that the social benefits from all decisions and policies must fall on the same citizens that bear the relative costs. Indeed, in the transport and mobility sector widespread negative externalities (Perveen et al., 2017) may lead to considerable difficulties in implementing political decisions due to the opposition by local communities that fear sacrifice on their part (so-called not in my back yard– NIMBY-syndrome). In this case the wide concept of sustainability passes through the capacity of local governance to involve the central government in a‘coopetitive’ action (Ioppolo et al., 2012).

As previously pointed out, improved allocative efficiency is the main objective of decentralization measures. From a productive efficiency viewpoint, however, in the transport field, frequently, the provision of transport services could lead to economies of scale because of the largefixed costs. In the local transport sector it is possible to achieve scale economies for an urban, a provincial or, as an upper limit, a regional catchment area. Then, devolution does not have a negative effect on production efficiency. Thus the right scale of administrative decentralization should depend on the technical production frontier.

The remaining sections of this paper analyze the impact of the devolution of administrative powers on the transport sector. The main insight of our Tieboutian's framework is that decentralization is more sensible whenever there exists a lot of heterogeneity among regions in terms of per capita income and in the corruptibility of public officials, as this is the case for Italy. Moreover, the Italian decentralization process might have been slow-downed due to fairness considerations.

2. Centralization and devolution of transport policy

In order to analyze the impact of decentralization on the transport industry, we start by examining recent developments in the transport policy of advanced countries.

The industrial era was characterized by the rigidity of the manu-facturing process and by modal oriented transport systems. In the post-Fordist era, the greater flexibility of production processes required a logistically integrated intermodal system (Stough and Rietveld, 1997). These changes induced an increasing complexity of transport system and a proliferation of stakeholders and (private and public) decision makers with respect to the industrialization era.

The expansion in the set of decision makers for the transportation public policy has caused several problems of coordination and

coopera-tion among institucoopera-tions involved in the transport field. Those are especially due to political differences between local and central governments and to different economic interests between areas of the same country.

For example, from the point of view of land-use and infrastructure planning, the expansion in the set of policy makers in the transport sector has extended the cost and time required to obtain implementable infrastructure decisions. Ranging from the search for community consensus to decision implementation, the several stages might be subject to a series of slowdowns due to excessive fragmentation of the institutional environment and to the greater difficulties in identifying costs and benefits of investments relating to ever-larger social areas.

Moreover, the objectives of the institutions in charge of planning, financing and operating transport activities have changed over the years. Examining the evolution of transport policy in the EU and in the US, it appears that institutions for planning,financing and operating transport systems were initially motivated by mobility and productive efficiency. Later, “external goals” became important political objec-tives. Safety, economic development and technological innovation are examples of external goals that, for social and cultural reasons, are now between the main objectives of the political process in the mobility sector. The growth in the number of goals expanded also the range of actors involved in transport policy, and this in turn enhances conflicts between different governmental institutions. Finally, during the 1960s environmental issues arose as an important goal in transport policy, and today the quality of the environment is a key factor in mobility decision making (Stough and Rietveld, 1997).3

During the past several decades, in order to simplify the government structure a large number of developing and industrial countries have implemented decentralization reforms.

In the US, for example, in 1962 The Federal Aid Highway Act established the requirement that planning be done at the regional level and no project could be imposed on a region. In 1991 the ISTEA (Intermodal Surface Transportation Efficiency Act) gave direct control on same investment choices to the regional planning organizations and balanced the power between DOTs (Departments of Transportation) and MPOs (Metropolitan Planning Organizations).4The US experience is interesting since it allows appreciating the consequences of the devolution in terms of the interaction and cooperation between central government and local authorities. However, the reform has stimulated a constructive collaboration between DOTs and MPOs, with significant benefits in addressing a wide array of urban and suburban transport issues. This result has been accomplished thanks both to the localized knowledge of regional planners and to the advantages of centralized state government in terms of scale economies, internalization of externalities, political legitimacy5_{and administrative uniformity that,}

in the transport sector, increases safety level (Taylor and Scheitzer, 2005).

Another important example is the UK national transport policy, which changed drastically during the 1990s, turning into the regulation of demand, with the aim of making better use of infrastructures. In line with a general trend toward devolution, with the 2004 Traffic Management Act (TMA) and the 2008 Local Transport Bill, the UK has provided greater power for the 150 existing Local Traffic Authorities (LTA).

The devolution of powers was achieved by delegating the admin-istration of the roads of each jurisdiction to the LTAs, authorities potentially more competent thanks to their localized knowledge

3_{Supporters of the devolution of powers claim that it is a device for the conservation of}

natural resources, thanks to the increase of social participation in policy making (Ahmed and Mbwambo, 2004).

4_{MPOs are agencies with responsibility and authority on metropolitan transport}

planning and operation.

5_{Central governments make easier to take an unpopular decision for a local}

(Brannigan and Paulley, 2008). Nevertheless, in the early years of the reform, LTAs chose to use few of the new powers available to them. Moreover, as underlined by Canning et al. (2010), some of the LTAs managers perceived that devolution might cause a conflict between the national goal of reducing road traffic and the local goal of economic development, that is “the tensions between reducing car dependence and development growth principles” (Hull, 2005).6

In Australia, a federal nation with a unique system of three-tier governance, traditionally control over explicit forms of urban govern-ance, notably planning, transport and urban services, has been vested in the states (Gleeson, 2001). In recent years, the growing accessibility, mobility and sustainability challenges of Australian cities have pushed local governments to play a more active role in the preparation and delivery of transport policies (Yang et al., 2015). For instance, the Roads and Transport Strategy (2006–2016) of Australian Local Govern-ment Association emphasize the necessity of devolution and partner-ship with State and Federal governments for funding.

The previous analysis has highlighted that while the knowledge of local planners (regional, sub-regional, metropolitan and urban) has helped to implement environmentally and citizens oriented transport policies (Ollivier-Trigalo, 2009), the devolution has also created problems of institutional coordination.

3. Mobility and devolution: a pure efficiency perspective From a social point of view, mobility services are a very important asset in developed areas while are less important in poor areas (Preston, 2009). In a country where there is a considerable income gap, conflicts may arise due to the different weights assigned by the population to mobility. For instance, if central governments choose eco-compatible transport systems funding this policy with an increase of the taxes, it may then happen that this choice leads to social discontent and to unsustainability of the transports policy.

On one hand, from the previous debate on the merits and demerits of government decentralization we have concluded that the rationale in favor of decentralization mainly rely on the idea first expressed by Tiebout (1956) that the inhabitants of the different regions have different tastes, for instance different transport preferences. Therefore, the same level of mobility services will not satisfy all of them.

On the other hand, decentralized provision will make possible to give different mobility standards to different regions. However, as underlined byPrud’homme (1994) decentralization might be accom-panied by more corruption, if corruption is more widespread at the local than at the national level (Fisman and Gatti, 2002).

In what follows, we will take a somehow different stand to argue that in Italy a higher degree of decentralization in transport sector is particularly important. The main argument is that devolution is more sensible when there is a lot of heterogeneity between regions both in terms of per capita income and in terms of corruption.

We adapt thefindings of a study byImmordino and Pagano (2010) who argue that stricter laws require more incisive and costlier enforce-ment. Since enforcement activity depends both on available tax revenue and the honesty of officials, the optimal legal standard of a benevolent government is increasing in per-capita income and decreasing in officials’ corruption. In this context, the rationale for the introduction of legal standards arises from a market failure. Specifically, a real externality occurs when an agent's decision affects the payoff of others directly rather than via the information conveyed by the market mechanism. An example of an externality may arise in a public-good setting, where firms providing transport services affect consumers’

utility via the level of pollution. Each consumer, being small, does not want to bear the cost to penalize pollutingfirms. Hence, as a result, the environment's quality will be sub-optimally low.

In such a setting social welfare can be determined as:

W=u s( ) −c s( ) +y− ,e (1)

where u(s)− c(s) is the surplus (consumers’ utility minus firms’ costs) associated with legal standard s, and y− e is the endowment net of enforcement costs. The legal standard constrains the technological choice of transportfirms limiting the environmental harm caused by their production process.

Sincefirms have the incentive to deviate from the legal standard, the standard must be assisted by appropriate enforcement, by ensuring that the following incentive constraint holds:

c s( ) ≤c q( ) +p e l( ) , ∀q≠ .s (2) where c(s) is thefirms’ costs, c(q) is the cost of an alternative choice and p(e)l is an expected penalty paid by afirm when deviate from the law. Under this constraint,firms prefer to abide by the law rather than deviating to an alternative choice q and paying the expected penalty p (e)l, which in turn depends on the actual penalty l and the enforcement activity e. Indeed, bureaucrats enforce the standard by inspecting transport firms, detecting non-compliance with probability p(e). In setting these variables, the government must take into account also the feasibility constraint:

e≤y−c s( ), (3)

where the resources devoted to enforcement activity, e, cannot exceed the economy's per-capita endowment y net of the resources used in the production of the infrastructure c(s).

However, if the law is enforced by corruptible officials: given a certain level of enforcement e chosen by the government, the prob-ability of detection is not determined anymore by a simple technical relationship p(e). Indeed, self-interested enforcers will try to take advantage of their position to extract bribes blunting the deterrence of enforcement. The bribe that an official can extract from a firm is denoted byαl where 0 < α < 1. The magnitude of α depends on how corrupt, unreliable and costly is the judicial system and on the fraction of corrupt officials, since this fraction will determine the probability with which afirm will face extortion.

Since now thefirm pays αl irrespective of compliance, it is more difficult to incentivize it for given enforcement e and penalty l. Under the new incentive constraint thefirm pays the expected bribe αp(e)l if it complies and p(e)l if it does not comply:

c s( ) ≤c q( ) + (1 −α p e l) ( ) , ∀q≠ .s (4) Under standard assumptions on the convexity of the cost function c (·) and the concavity of the detection technology p(·), the comparison between (2) and (4) immediately shows that with corruption the enforcement required to police a given standard s is larger than without it. The reason being that extortion by bureaucrats makes it harder to incentivizefirms. Indeed, the enforcement required for a given standard s is increasing in the degree of corruption,α.

Then, the optimal standard with corrupt enforcers is found max-imizing the social welfare(1)subject to the incentive constraint(4)and the feasibility constraint(3)and is characterized as follows: The optimal standard with corrupt enforcers is decreasing in the degree of corruptionα and weakly increasing in per-capita income y.7

Thisfinding corresponds to a Tieboutian framework in which local tastes and preferences are carried out through local policy. However, many concerns such as externalities and fairness have often been leveled against Tieboutian analysis.

6_{Analyzing the German experience,Schöller-Schwedes (2010)underlines the}

impor-tance of transport system integration from a technical, political and social point of view. A lack of cooperation between local governments in planning and operating transport activities is one of the causes of the low integration of transport system in a“too

4. Italian devolution: transport and infrastructures issues

In the area of corruption research, continuous efforts have been conducted to develop a systematic way to measure the phenomenon. Measures of corruption currently in use are perception-based (Kapoor and Ravi, 2012) or are based on objective indicators drawing on data other than the surveys.

In order to map the risk of corruption at subnational level, existing empirical studies generally adopt objective measures, as the number of prosecuted corruption-related cases (ANAC, 2013).

In this section, we present the latest data for the Italian regions on the number of reported crimes (per 100,000 inhabitants) related to corruption and the data on per capita income (an average for the period of 2004–2013) (Table 1). Those data testify the huge heterogeneity between Italian regions in both those dimensions. For instance, the northern region of Lombardy is about twice as rich as the southern regions of Apulia, Calabria, Campania and Sicily. Even in terms of corruption it is apparent the usual North-South Italian divide. Indeed, there are not regions from the North and only 2 from the Centre above the average number of reported crimes of 1,2, while there are 6 out of 7 southern and insular regions with a corruption higher than the average. In what follows, we briefly describe the Italian devolution experi-ence and the main related issues in the fields of transport and infrastructure.

Italy began the decentralization of administrative responsibility with thefiscal federalism of the 2001 Constitutional reform. Law 42/ 2009 and its subsequent law-decrees further implemented the decen-tralizing process.

The purpose of the reform was to achieve a federal state founded on principle of subsidiarity and on the redistribution of legislative powers between the State and the Regions (Ferrari, 2002).

However, as shown below (in Section 5), there is also a strict correlation between social fairness and devolution. Therefore, the Italian federalist model deals also with the key principles of fairness and equity (Frosini, 2009). For this reason, in the current Italian political debate on merits and demerits of the devolution, there are arguments for equity promoting consequences. In order to obtain agreement of both the opposition and the majority in Parliament, the reform was limited to the administrative functions and legislative

powers, while the asymmetry betweenfiscal power and public funding remains. In other words, local governments spending (Municipalities, Provinces, Metropolitan Cities and Regions) is not linked to the power of setting taxes, despite Article 119 of the Italian Constitution sets out that“Municipalities, Provinces, Metropolitan Cities and Regions shall havefinancial autonomy with respect to revenues and expenditures”.

Moreover, during the 2010–2013 period, in order to reduce public deficit, Italian governments implemented a fiscal program that in-creased the tendency to overlap central and local government's autonomy (Bordignon, 2013). Though the framework law of 2009 imposes a more complete federal fiscal structure by 2017, the im-plementation of the process has been very slow (European Committee of the Regions, 2014). In an attempt of relaunching federalist reforms, the Italian Parliament approved Law 56/2014 with the aim of establish-ing metropolitan cities and redefinestablish-ing competences of the provinces. The constitutional reform proposed by Renzi government with the aim of converting the Senate into a federal parliament representing local and regional administrations was rejected by the constitutional refer-endum of December 4th, 2016.

Despite the measures adopted, state and local-level policies fre-quently overlap also in the transport and infrastructures sector. For example, the Regions implement environmental policy, whereas the central government has the competency over the realization of the major infrastructures projects, in a framework defined at the EU level. Moreover, the planning of local public transport is decentralized to local governments, while the central government still intervenes to make up budgetary deficits. This causes a mismatch between the competencies of different government levels and a welfare and produc-tion efficiency loss.

As underlined above, another crucial issue is the North-South economic divide. In the transport field, South Italy suffers from an infrastructure and logistics gap compared to Northern Italy regions (Carlucci et al., 2017). Having this in mind the law onfiscal federalism (Law no. 42/09) includes the redistribution of transport infrastructure investments from one region to another through national funds to be assigned to Municipalities, Provinces, Metropolitan Cities and Regions. Fiscal federalism reform will also introduce a rewards type system and impose a tax burden depending on financial performance of local governments. This mechanism is in contrast with the goal of horizontal equity, because it leads to a further increase of the North-South divide, as public bodies in southern Italy have worsefinancial performances than in the North.

However, as underlined by Prud’homme (1994), the provision of public services is a complex activity composed by several functions, each exhibiting different characteristics: the planning, the choice of investment, the construction and maintenance of facilities, the market regulation. From an efficiency point of view, each function should be delegated to a specific national or local government. For instance, the technical, organizational and institutional complexity of transport planning requires that it should remain a central government function. For other functions such as the pricing, knowledge of local realities is crucial. Finally, there are functions like regulation that require the simultaneous involvement of different levels of government. In these circumstances both central government independence and local gov-ernments accountability, which reduces lobbying, are needed. However, if several levels of government must simultaneously be involved, as in Italy, there is a risk of vertical competition and institutional conflict, which could implies (production and allocative) efficiency losses. More experiences carry out the need to build a new model of governance (from national level to local one), sustainability oriented, in order to overcome problems related to the different point of view (Ioppolo et al., 2016).

Indeed, issues of vertical government competition, overlap of competences and allocative inefficiency might arise in systems char-acterized by heterogeneity and where the link between spending, and taxation is not strong enough. It is the Italian case, in which tax and

Table 1

Per capita income (euros) and corruption. Reported crimesa_{per 100,000 inhabitants}

(average 2004–2013).

Data sources: SAeT, 2010 (National Anticorruption Agency), ANAC (National Antic-orruption Authority) andISTAT, 2016.

Region Per capita income Corruption Area

Aosta Valley 33,581 1.1 North-West

Lombardy 32,309 1.2 North-West

Trentino-Alto Adige 32,215 0.9 North-East Emilia-Romagna 31,194 0.6 North-East

Lazio 29,387 1.8 Centre

Veneto 29,226 0.5 North-East

Friuli Venezia Giulia 28,414 0.5 North-East

Piedmont 27,759 0.4 North-West Tuscany 27,379 0.8 Centre Liguria 26,530 0.9 North-West Marche 25,880 1.1 Centre Umbria 23,708 0.6 Centre Abruzzo 21,158 1.6 Centre Molise 19,672 1.6 South Sardinia 19,114 1.8 Islands Basilicata 17,426 2.9 South Apulia 16,883 1.5 South Sicily 16,527 1.0 Islands Campania 16,260 1.9 South Calabria 16,120 1.9 South

a_{The crimes considered are those regulated by the Italian Penal Code artt. 318, 319,}

expenditure are still quite centralized, subnational governments get central government transfers and there is heterogeneity between regions. This leads to underprovision in some transportation areas and overprovision in other areas, because as highlighted byDuc (2008) tastes and preferences are not homogeneous across locations.

Moreover, many political powers overlap since, in absence of a strong link between taxation and spending, land use and large transportation infrastructures are matter of concurrent competencies of the State and the Regions.8 _{This concurrent legislative power, as}

defined by the reform of the Title V of the Italian Constitution, has a negative impact in terms of policy efficiency and sustainability because the complex relationship between national and local governments. Since the current system of decentralization causes inefficiency and little accountability, further reforms of the Italian public finance are needed.

5. Mobility and devolution: adding fairness considerations

As explained byFehr and Schmidt (2006), in their contribution to the Handbook of the Economics of Giving, Altruism and Reciprocity, people often care for the well-being of others and this may have important economic consequences: “This evidence indicates that a substantial percentage of the people are strongly motivated by other-regarding preferences and that concerns for the well-being of others, for fairness and for reciprocity, cannot be ignored in social interactions.”

The goal of this section is to add fairness considerations to the previous model, to study how equity affects our main result, and to suggest how the Italian decentralization process might have been slow-downed due to fairness considerations. Let us assume that a country is made up of two regions: (N)orth and (S)outh. Then social welfare can be written as:

Wi=u s s( ,i j) −c s( ) +i y− ( ;e s αi i) (5) where i is either N or S and j is the other region. Moreover, the function e(·) is obtained from(4)when the incentive constraint is binding. The main difference with the objective function in (1) is that now the welfare of region i depends also on the standard of region j through consumers utility u(·). While almost all models of fairness, altruism and reciprocity start out by making some fairly specific assumptions about the players’ utility functions. Alternatively, we start from a general preference relation and after showing our results we will propose, as an example, a specific functional form.

Let us consider three possible cases: (i)first, the regions are poor and the feasibility constraint(3)is binding for both of them; (ii) second, one region (S) is poor and the other is rich (N) so that only the feasibility constraint of S is binding, and; (iii) third, both regions are rich and the feasibility constraint is not binding for neither of them.

In (i) the standards are implicitly determined by the binding feasibility constraints and fairness consideration do not play any role:

yi= ( ;e s αi i) +c s( )i for =i N S, . (6) In the case (ii) the standard of the poor region, say the South is again determined by the binding feasibility constraint and is indepen-dent from the other region's standard. Conversely, the standard for the North is found maximizing the social welfare (5) subject to the incentive constraint(4)(the feasibility constraint is slack by assump-tion) and is characterized by the followingfirst order condition:

u s′( ,N sS) = ′(c sN) + ′( ;e sN αN). (7) Totally differentiating(7)gives:

ds ds u u c s e s α = − = − ″ (⋅) ″ (⋅) − ″( ) − ″( ; ) N S W s s W s N S N N N N N ∂ ∂ ∂ ∂ ∂ , , N N S N N 2 2 2 ₍₈₎

under standard assumptions on the convexity of the cost function c(·) and the concavity of detection technology p(·) and utility function u(·), it is possible to show that the sign of the previous expression depends only on the numerator. In words, the standard chosen in the North region depends positively on that in the South region if the sign of the cross derivative of u(·) is positive.

In the case (iii) both regions are rich and choose the standard keeping into account the choice of the other region. Thus the standards are the outcome of the Nash equilibrium of a game played by the two regions. Formally, the outcome is found solving the system offirst order conditions, where one is Eq.(7)and the other is easily obtained by(7) inverting subscripts N and S. Finally, totally differentiating both first order conditions yields the slopes of the regions reaction curves (one is identical to(8)while the other is easily obtained from(7)by inverting subscripts N and S), which are both positive if the signs of the cross derivatives of the utility functions are positives. If this is the case the standard in one region is an increasing function of the standard in the other region. In other words, standards are strategic complements.9

We conclude the section by showing how the previous conditions are satisfied for a specific functional form proposed by Cox et al. (2007). ⎪ ⎪ ⎧ ⎨ ⎩ u s s s θs α α s s α ( , ) = ( + )/ if ≠ 0 if = 0. i j iα jα i jα ₍₉₎

In the previous expressionα reflects the curvature of the indiffer-ence curves andθ, that is interpreted as the “emotional state” of player i, captures whether preferences are altruistic or spiteful. Straightforward calculations show that the sign of the cross derivative of the utility function is positive forθ sufficiently large. In other words, if the consumers are sufficiently altruistic the standards in the two regions (except for case (i) are, differently from the pure efficiency baseline model, positively related to each other and do not depend uniquely on the region income and degree of corruption.

6. Conclusions

Public sector plays a key role in mobility services. Moreover, in Italy, corruption is a relevant issue in transport infrastructurefinancing and service provision.

As pointed out by theItalian Ministry of Justice (2012)in a report, corruption has effects on the Italian economy, where the total direct costs of corruption amount to 60 billion Euros per year (approximately 4% of GDP), and more specifically in major public works (mainly transport infrastructure) the costs connected to the corruption are estimated around 40% of the contract cost.

This paper examines the prospects and constraints of centralization and devolution for the transport policy. It offers an overview of several examples of devolution in the transport sector and focuses on the Italian devolution in the mobilityfield.

The main insight generated by the analysis is that, under a pure efficiency perspective, decentralization is more sensible whenever there exists a lot of heterogeneity between regions in terms of per capita income and in the corruptibility of public officials. This seems to be the case for Italy.

The idea behind this result is that corruption induces a lower legal standard because it has a“deterrence-blunting effect”: when enforcers are corrupt, policing a given standard requires more enforcement

8_{Other areas of concurrent competences are foreign trade, job safety, culture and the}

environment.

9_{In game theory the decisions of two or more players are called strategic complements}

if they mutually reinforce one another, and they are called strategic substitutes if they mutually offset one another. SeeBulow et al. (1985).

resources than otherwise, so that it is efficient for the regulator to choose a less ambitious standard than he would with loyal enforcers. Moreover, if income is low the society will content itself with lower mobility standards. This result sheds some light on our research question—since different degrees of corruption and per-capita income imply different optimal mobility standards, a common policy is sensible whenever the regions are not too heterogeneous in terms of those characteristics. In practice, to choose the candidate regions to an optimal common policy “area” we should weight both per-capita income and the degree of corruption to get a synthetic measure of homogeneity between the regions. The assembly, for practical pur-poses, of this weighted index is left for future research.

Turning to the Italian decentralization process in the transport and infrastructures sector, we have argued that despite the measures adopted, state and local-level policies frequently overlap causing inefficiency and little accountability. We believe that the Italian decentralization process might have been slow-downed due to fairness considerations.

Finally, our prospective research will particularly concentrate on further testing the findings generated in this paper by undertaking cross-country comparative studies—countries with relatively higher and lower corruption levels to that Italy.

Acknowledgements

The authors would like to thank the Land Use Policy Editor and referees for valuable comments on the earlier version of the manuscript. Fabio Carlucci is Scientific Director of the Laboratory of Agricultural Economics, Transport and Tourism (LabEATT);Giovanni Immordino is member of the Center for Studies in Economics and Finance (CSEF); Andrea Cirà, Giuseppe Ioppolo and Tan Yigitcanlar are members of the Research & Mobility Group(R & M).

References

Ahmed, A.E., Mbwambo, J.S., 2004. Does decentralization have a positive impact on the use of natural resources? Term Paper for the Interdisciplinary Course, International Doctoral Studies Program at ZEF (Zentrum für Entwicklungsforschung), Bonn. Alesina, A., Spoalore, E., 1997. On the number and size of nations. Q. J. Econ. 112 (4),

1027–1056.

ANAC (Autorità Nazionale Anticorruzione). 2013. Corruzione sommersa e corruzione emersa in Italia: modalità di misurazione e prime evidenze empiriche.http://www. anticorruzione.it/portal/rest/jcr/repository/collaboration/Digital%20Assets/ anacdocs/Attivita/Pubblicazioni/RapportiStudi/Metodologie-di-misurazione.pdf. (accessed 07.11.16).

Balaguer-Coll, M.T., Prior, D., Tortosa Ausina, E., 2010. Decentralization and efficiency of local government. Ann. Reg. Sci. 45, 571–601.

Barro, R., 1970. The control of politicians: an economic model. Public Choice 14, 19–42. Bewley, T., 1981. A critique of Tiebout's theory of local public expenditures.

Econometrica 49 (3), 713–740.

Bordignon, M., 2013. Economic Crisis and Recentralization of Government: The Italian Experience. IEB's Report on Fiscal Federalism. pp. 62–66.

Brannigan, C., Paulley, N., 2008. Funding for local authority transport and land-use schemes in the UK. Transp. Policy 15, 379–386.

Bulow, J.I., Geanakoplos, J.D., Klemperer, P.D., 1985. Multimarket oligopoly: strategic substitutes and complements. J. Polit. Econ. 93, 488–511.

Canning, P.E., Hellawell, E.E., Hughes, S.J., Gatersleben, B.C.M., Fairhead, C.J., 2010. ‘Devolution’ of transport powers to local government: impacts of the 2004 Traffic Management Act in England. Transp. Policy 17, 64–71.

Carlucci, F., Cirà, A., Forte, E., Siviero, L., 2017. Infrastructure and logistics divide: regional comparisons between North Eastern & Southern Italy. Technol. Econ. Dev. Econ. 23 (2), 243–269.

Cox, J.C., Friedman, D., Gjerstad, S., 2007. A tractable model of reciprocity and fairness. Games Econ. Behav. 59, 17–45.

De Valk, P., 1990. State decentralization and participation. In: de Valk, P., Wekwete, K.

(Eds.), Decentralizing for Participatory Planning: Comparing the Experiences of Zimbabwe and Other Anglophone Countries in Eastern and Southern Africa. Avebury Press, Aldershot, pp. 62–84.

Duc, V., 2008. The Economics of Measuring Fiscal Decentralisation Part 2: New Fiscal Decentralisation Indices. Economics Discussion/Working Papers 08-14, The University of Western Australia. Department of Economicspp. 53–93. European Union. Committee of the Regions, 2014. Division of Powers Between the

European Union, Member States, Candidate and Some Potential Candidate Countries, and Local and Regional Authorities: Fiscal Decentralisation or Federalism. Fehr, E., Schmidt, K.M., 2006. The economics of fairness, reciprocity and altruism–

experimental evidence and new theories. In: Kolm, S., Ythier, J.M. (Eds.), Handbook of the Economics of Giving, Altruism and Reciprocity. Elsevier, Amsterdam, pp. 615–691.

Ferrari, E., 2002. Planning, building and environmental law after the recent Italian devolution. Eur. Public Law 8 (3), 357–365.

Fisman, R., Gatti, R., 2002. Decentralization and corruption: evidence across countries. J. Public Econ. 83, 325–345.

Frosini, T.E., 2009. Introduction to Italianfiscal federalism. federalismi.it 19: 1–6.http:// www.federalismi.it/ApplOpenFilePDF.

cfm?artid=14466&dpath=document&dfile=09112009183249.

pdf&content=Introduction+to+italian+fiscal+federalism+-+stato+-+dottrina +-+. (accessed 07.11.16).

Gleeson, B., 2001. Devolution and state planning systems in Australia. Int. Plann. Stud. 6 (2), 133–152.

Haynes, K.E., Gifford, J.L., Pelletiere, D., 2005. Sustainable transportation institutions and regional evolution: global and local perspectives. J. Transp. Geogr. 13, 207–221. Hull, A., 2005. Integrated transport planning in the UK: from concept to reality. J. Transp.

Geogr. 13, 318–328.

Immordino, G., Pagano, M., 2010. Legal standards, enforcement and corruption. J. Eur. Econ. Assoc. 8 (5), 1104–1132.

Ioppolo, G., Saija, G., Salomone, R., 2012. Developing a Territory Balanced Scorecard approach to manage projects for local development: two case studies. Land Use Policy 29, 629–640.

Ioppolo, G., Saija, G., Salomone, R., 2013. From coastal management to environmental management: the sustainable eco-tourism program for the mid-western coast of Sardinia (Italy). Land Use Policy 31, 460–471.

Ioppolo, S., Cucurachi, S., Salomone, R., Saija, G., Shi, L., 2016. Sustainable local development and environmental governance: a strategic planning experience. Sustainability 8 (2), 180.

ISTAT, 2016. Delitti denunciati dalle forze di polizia all’autorità giudiziaria. (accessed 07.11.16).http://dati.istat.it/Index.aspx?DataSetCode=dccv_delittips.

Italian Ministry of Justice, 2012. Rapporto sulla corruzione. (accessed 07.11.16).https:// www.giustizia.it/giustizia/it/mg_6_1_1.wp?contentId=NOL783861.

Kapoor, M., Ravi, S., 2012. Determinants of corruption: government effectiveness vs. cultural norms. B.E. J. Econ. Anal. & Policy 12 (1), 1–24.

Leone, R., Carroll, B.W., 2010. Decentralisation and devolution in Canadian social housing policy. Environ. Plann. C: Gov. Policy 28 (3), 389–404.

Liu, A., Zhang, J., 2013. Fiscal decentralization and environmental infrastructure in China. B.E. J. Econ. Anal. Policy 13 (2), 733–759.

Oates, W.E., 1972. Fiscal Federalism. Harcourt Brace Iovanovich Inc., New York. Oates, W.E., 1999. An essay onfiscal federalism. J. Econ. Lit. 37 (3), 1120–1149. Ollivier-Trigalo, M., 2009. Politiques de transport: Où en sont les Régions? Actions,

doctrines et institutionnalisation. Revue D’Économie Régionale et Urbaine 3, 471–490.

Perveen, S.T., Yigitcanlar, M., Kamruzzaman, J., Hayes, J., 2017. Evaluating transport externalities of urban growth: a critical review of scenario-based planning methods. Int. J. Environ. Sci. Technol. 14 (3), 663–678.

Preston, J., 2009. Epilogue: transport policy and social exclusion– some reflections. Transp. Policy 16, 140–142.

Prud’homme, R., 1994. On the Dangers of Decentralization. Policy Research Working Paper 1252. World Bank, Washington.

Seabright, P., 1996. Accountability and decentralisation in government: an incomplete contracts model. Eur. Econ. Rev. 40, 61–89.

SAeT Servizio Anticorruzione e Trasparenza, 2010. Relazione al Parlamento, Dipartimento della Funzione Pubblica. Autorità Nazionale Anticorruzione, Rome. Schöller-Schwedes, O., 2010. The failure of integrated transport policy in Germany: a

historical perspective”. J. Transp. Geogr. 18, 85–96.

Stough, R.R., Rietveld, P., 1997. Institutional issues in transport systems. J. Transp. Geogr. 3, 207–214.

Taylor, B.D., Scheitzer, L., 2005. Assessing the experience of mandated collaborative inter-jurisdictional transport planning in the United States. Transp. Policy 12, 500–511.

Tiebout, C.M., 1956. A pure theory of local expenditures. J. Polit. Econ. 64, 416–424. Yang, J., Yuan, M., Yigitcanlar, T., Newman, P., Schultmann, F., 2015. Managing

knowledge to promote sustainability in Australian transport infrastructure projects. Sustainability 7 (7), 8132–8150.