Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A1-A11.
IN SEARCH OF HORIZONTAL GOVERNMENT: INSIGHTS ON
JOINING UP MUNICIPALITIES 1
Andrea Garlatti*
Udine State University, Department of Economics and Statistics
Luca Brusati
Udine State University, Department of Economics and Statistics
Paolo Fedele
Udine State University, Department of Economics and Statistics
1. Introduction.
Joined-up government and enhanced coordination are an evergreen in public management studies (Hood, 2005). As Pressman and Wildavsky (1984) put it: "no phrase expresses as frequent a complaint about the bureaucracy as does 'lack of co-ordination'. No suggestion for reform is more common than 'what we need is more co-ordination"' (p. 133).
A couple of decades of decentralization, often within "new public management"-inspired reforms have made coordination at the local level an even hotter topic on the agenda of many governments (Pollitt and Bouckaert 2011; Fedele and Ongaro 2008). Academic and practitioner literature, in fact, have often stressed the ways stronger inter-municipal coordination can be beneficial: attracting investors through increased services (Downs, 1994; Felbinger, 1984; Nunn & Rosentraub, 1997; Orfield, 1997); increasing professionalism and rationality of local bureaucrats (Barlow, 1991; Lind, 1997); reducing costs of services through economies of scale and reducing concomitantly taxes (Bunch & Strauss, 1992; Dolan, 1990); increasing planning capacity and reducing zero-sum competition for economic development (Fleischmann & Green, 1991; Rigos, 1995). Coordination among municipalities and joint service delivery have often been hot topics on the agenda of Italian decision-makers (Garlatti et al., 2013). The Italian local government system, in fact, is highly fragmented, inasmuch more than 60% of its 8,000 municipalities count less than 5,000 inhabitants. Small and very small municipalities often struggle to perform their functions, in both financial and organizational terms. Nevertheless, implementation of horizontal coordination
1	The paper is the result of the joint work of the three authors. Nevertheless in the final version, Luca Brusati
wrote the paragraph 1, Paolo Fedele wrote the paragraph 2, and Andrea Garlatti wrote the paragraph 3; the paragraph "Conclusions" was jointly written.
*Correspondence author
Received: 18 November; revised: 11 March; accepted: 30 March.	A1
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
has been problematic. Some attempts to merge municipalities have been stymied by political resistance at local level; attempts to coordinate service delivery fared better, but usually piecemeal approach and unstable arrangements. In the last few years, however, due to cuts in intergovernmental transfers and the impossibility to keep raising local charges, the issue is squarely back on the agenda.
This paper focuses on the policies implemented by Autonomous Region Friuli Venezia Giulia (FVG) to promote the joint delivery of municipal services. The analysis addressed in particular two research questions: 1) what were the "programme theories" entailed by the coordination devices used in FVG? The analysis covers the period 2006-2013. In 2006, in fact, a major reform in local government regulatory framework was established. 2) How did the specific context of FVG influence the "programme theories" when put to work? Results contribute to shed light on how the contextual factors (Pawson and Tilley, 2004) influence the implementation of joined up government initiatives. The analysis of FVG policies carried in this article, beside producing locally valid explanations, provides also the opportunity to attempt to extend findings' validity beyond the research site.
2. Analytical framework and methodological choices.
The study follows the realist evaluation logic (Pawson and Tilley 1997; 2004, Pawson et al., 2004). Therefore, it uses the CMO (context,mechanism, outcomes) framework, proposed by realist evaluation, to analyse the inter-municipal cooperation policies adopted in FVG. In realist analysis policies are hypothesis about social improvement shaped by a vision of change. Evaluation, by these lights, has the task of testing out the underlying programme theory, i.e. the hypothesis about how to achieve improvement (Pawson and Tilley,1997). Consequently, identifying the crucial programme mechanisms is the first step in a realist evaluation. This not a trivial task since the generation of policy ideas is often a complex process rather than the design of a single, fully rational policy architect (Lindblom, 1959). Once insulated the core policy mechanisms that a policy entails, realism utilises contextual thinking to address the issues of 'for whom' and 'in what circumstances' a programme will work. Through analysis of the "context', realist evaluation shed critical light to policy "miracle drugs" . For realism, it is axiomatic that certain contexts will be supportive to the programme theory and some will not. Context is not to be intended merely as localities, but as those conditions in which programmes are introduced. The interaction of the identified policy mechanisms with contextual features produce specific outcome patterns, i.e. the intended and unintended consequences of programmes, resulting from the activation of different mechanisms in different contexts.
A2
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
In the light of the CMO framework, in order to describe the coordination mechanisms, i.e. the policy's programme theories, we adopted the analytical framework suggested by Verhoest et al. (2010). Those authors classified a number of coordination devices that can be based alternatively on: structural solutions (e.g., a joint organizational unit or position) or management system (e.g., a joint planning system). Each device relies on a specific coordination logic (Ouchi 1980, Peter 1998, Thompson et al. 1991), namely hierarchy, market, or network. Context, for the purpose of the present work, will not be operationalized in details, but described in a narrative fashion so to have explaining factors "emerge" as relevant from field analysis (Glaser and Strauss 1967). As concerns outcome patterns, the paper will preliminary take into account the degree of implementation and the pace of actual implementation of the policy under analysis. Subsequently, the article provides some insights on how attempt to increase coordination affected personnel's arrangements and managerial practices in local governments.
In order to collect relevant information, first of all documental analysis has been carried out. Official and unofficial document have been reviewed to build up a longitudinal narrative of the policies adopted by FVG government in the period 2006-2013. In order to triangulate findings from documental analysis have been complemented by direct observation of some inter-municipal coordination events.
3. Case study
3.1 Context
Friuli Venezia Giulia is located in the North- East of Italy. Its population amounts to 1,239,234 distributed on a surface of 7,858.39 Km2. The intermediate tier (region, in the Italian public system) is represented by the Autonomous Region of FVG. FVG is one of the so-called "special statute" regions in the Italian intermediate government system: as such it enjoys a higher degree of law-making power than "ordinary statute" regions, due to socio-cultural, historical and geographic reasons. More specifically, FVG holds exclusive powers in defining the legal and financial framework where local governments operate. The intermediate level has been established merging two traditional territorial identities (Friuli and Venezia-Giulia). The local level is composed of 217 municipalities (as of 2014). A good portion of them, 91, is located in mountain areas, while 16 municipalities are located in partially mountain areas. Density is low especially in mountain areas: table 1 shows the demographic cluster per geographical type.
A3
Journal of Universal Excellence,
June 2015, year 4, number 2, pp. A12-A36.
Appendix
Table 1. FVG municipalities: demographic classes
	0-500	501-1.000	1.001-3.000	3.001-5.000	5.001-15.000	15.001-30.000	Over 30.000
Municipalities located in mountain	19	25	38	3	6	0	0
areas							
Municipalities located in partially mountain	0	0	4	1	8	1	2
areas							
Municipalities not located in mountain	0	4	43	16	39	6	2
areas							
Source: Piano di Valorizzazione Territoriale 2014, Regione Autonoma Friuli Venezia Giulia
Like other areas in Italy, campanilismo (Ongaro, 2011) is a very important aspect of (municipal) life symbolizing a sense of identity, pride and belonging to the place of birth, a feeling which is sometimes stronger than the sense of national or regional identity. On the other side, minorities are significantly present due both to the geographic position (FVG share border with Austria, Slovenia and Croatia) and to its peculiar historical events. However, like the rest of Italy, legalism - i.e the predominance and -to some extent- the institutionalization of administrative law in defining the mode of functioning of public administration- is largely dominant in the public sector (Capano, 2003, Ongaro, 2011;2010; Peters, 2008). Previous attempts at implementing inter-municipal cooperation might be included in the context (as defined by realistic evalution).As many other region in Italy, attempts at coordination were carried out, but policies aimed at municipal coordination were limited in scope and partly unsuccessful (Zanin, 2013). As of 2005, in fact ,only 3 experience of inter-municipal cooperation of some kind had been adopted. Finally and obviously, other elements can "emerge" as relevant from field analysis (Glaser and Strauss 1967).
3.2 Coordination mechanisms: the programme theory
Also on the basis of the criticisms by the Court of Auditors, in 2006 FVG passed a new overall legal framework for local government. This broader reform package also contained some provisions regarding inter-municipal coordination. According to the 2006 law, in fact, coordination can occur through different policy tools:
■ Convenzione, a contractual agreement between two municipalities to manage specific services during a specific time frame. Under such agreement, municipalities might (but are
A4
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
not required to) establish joint offices or units. Alternatively, joint functions can be delegated to one of the municipalities involved in the "convenzione".
■	Associazione intercomunale, a framework agreement between a number of municipalities lasting at least than six years and covering the joint delivery of a wide number of services. This agreement envisages, as a compulsory component, the establishment of joint offices or units. The framework agreement, approved by each local council involved, needs to be specified through service-specific agreements.
■	Unione di Comuni, a newly established second level tier that integrates neighbouring municipalities. Also in this case the time frame is at least six years. Since Unione represents a newly established tier, which is attributed legal personality and can deliberate its own statute, joint offices and unit are a compulsory component.
■	Consorzi tra Enti Locali, also a newly established organization which includes, beside municipalities, public authorities of different kind. Consorzi are meant to facilitate the joint delivery of public services.
■	Aster (Local Development Area). In the original version of the reform act, Associazioni intercomunali and unioni di Comuni were attributed the status of ASTER (local development Areas) in case they counted at least 30,000 inhabitants (or at least 15,000 inhabitants, but involved at least at 10 municipalities). Larger cities and mountain area municipalities were ex lege attributed the nature of ASTER. ASTERs were granted the opportunity of negotiating additional funds from FVG regional government to finance investment programmes. However, this policy option has been written off in 2009.
According to the reform act, Associazione intercomunale (framework agreement) and Unione di Comuni(newly established second level tier) can potentially develop into definitive mergers (fusioni) between municipalities. This further step requires carrying out local referenda. In the 2006 policy package, the adoption of the above mentioned form of inter-municipal cooperation is tied to extra-funding granted by the regional (intermediate) tier. More specifically, additional funding can occur in two form. First of all, municipalities can be entitled a one-off, non recurring extra grant just for establishing an Associazione intercomunale or Unione di Comuni. Furthermore, municipalities adopting coordination devices are granted an yearly extra-fund that lasts no more than six years. Since the third year, this yearly fund is progressively reduced to half of the first year amount. The funds are granted to the inter-municipal body as such, therefore member municipalities need to agree on how to allocate the extra-budget. These funds
A5
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
can be reversed in case FGV finds, via randomized control, that cooperation is not been implemented as planned by municipalities that are being granted extra money.
According to the 2006 law, in order to implement the policy package, FVG regional government issues a three-year "Local Enhancement Plan" (piano di valorizzazione territoriale) to be updated yearly. The plan's mandate, more specifically is to:
■	monitor and assess the number and type of coordination arrangements actually in place;
■	set the criteria for funding inter-municipal cooperation (both starting funds and yearly funds for six years).
■	Set the criteria for funding ASTERs (till this policy tool was in place);
■	set the criteria for funding mergers between municipalities (fusioni) that could potentially occur.
3.3 Outcomes
As concerns the first outcome pattern under analysis, i.e. degree and the pace of implementation, results are pretty impressive. The policy established in 2006, in fact, has involved most FVG municipalities. 202 municipalities out of 217 (originally 219 before two mergers occurred) took part, in fact, in some form of inter-municipal coordination. Table 2 show the progressive implementation over the period 2006-2013. Implementation was rather fast and widespread, but the preferred mode of inter-municipal coordination was the one with the lowest level of long-term commitment.
Table 2. Implemention of intermunicipal coordination (2006-2013)
Timing	Associazioni intercomunali	Unioni di comuni	Total	note
By May 2006	8	4	12	
By March 2007	35	5	40	
By February 2008	35	6	41	A merger occurred (active since 2009)
By February 2009	36	5	41	
By February 2010	37	4	41	
By February 2011	37	5	42	
By February 2012	37	5	42	
By February 2013	35	6	42	A merger occurred (active since 2014)
Source: Piano di Valorizzazione Territoriale 2014, Regione Autonoma Friuli Venezia Giulia (Local Enhancement Plan)
A6
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
Associazioni Intercomunali (framework agreement) in most cases grouped a number of municipalities between 4 and six. Only in few cases, framework agreements involved a larger number of local authorities.
Table 3. Establishment of Associazioni Intercomunali: numbers of municipalities involved
Municipalities involved	Associazioni Intercomunali
Up to 3	9
From 4 to 6	17
From 7 to 9	3
From 10 onwards	6
Source: Piano di Valorizzazione Territoriale 2014, Regione Autonoma Friuli Venezia Giulia (Local Enhancement Plan)
Unioni di Comuni, on the other hand, partly predictably, grouped generally a smaller numbers of municipalities (in all the cases never more than three municipalities).
Table 4. Establishment of Unioni di comuni: numbers of municipalities involved
Municipalities involved	Unioni di Comuni
Up to 3	6
From 4 to 6	0
From 7 to 9	0
From 10 onwards	0
Source: Piano di Valorizzazione Territoriale 2014, Regione Autonoma Friuli Venezia Giulia (Local Enhancement Plan)
4. Findings and conclusions.
It is time to try to answer the research questions formulated at the beginning of the article. First of all, what was the "programme theory" entailed by the coordination devices used in FVG? As said before, reconstructing and making sense of the intended mechanism for achieving change is not a trivial task. Referring to the above mentioned analytical framework by Verhoest et al. (2010), the policy package designed two main forms of integration. The first is clearly dynamic and contractual (Associazioni intercomunali, the framework agreement), although it includes some element of structural solutions inasmuch inter-municipal agreement envisages, as a compulsory component, the establishment of joint offices or units. The second is clearly structural (Unione di comuni). Unione, in fact, is represent a newly established tier, which is attributed legal personality and is based on joint offices and units. As concerns the coordination logic that the policy entails, it might be said that the 2006 law is made of two components. First of all, in the short-mid term the policy favoured coordination through market logic (meant as incentives) rather than hierarchy (authoritative decision) or network (voluntary decision based on cooperative behaviour). The main
A7
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
tool to promote inter-municipal coordination (via structure or contract) is, in fact, the additional funding provided by FVG in case associazioni or unioni are established. It is assumed that officials at the municipal level are rational and self-interested and therefore will not miss the opportunity to get extrafunds, adopting, therefore, coordination arrangements.
Anyway, besides this core "incentive based" policy, other mechanisms are supposed to work in the long term, according to the policy design. The incentives, in fact, are meant to be granted for no longer than six years. After this six years period, inter-municipal coordination arrangements already in place are expected to keep on walking on their own feet. The policy seems to entail that after the six years "incubation" period, municipalities will be willing to keep on jointly operating. This can happen through both a network logic and hierarchy logic. On one side, in fact, the policy assumes after a six years joint work period, officials will develop shared visions and beliefs so that cooperation will be sustained even without market-type incentive. On the other side, hierarchy and formal structure play a role especially as concerns unioni, which are government tiers established from scratch, but also as concerns associazioni since they are required to create joint offices. In other term, the policy seems to assumes that progressive institutionalization of formal structures should "lock in" municipalities and should represent an exit barrier that makes opt-out more costly and less likely.
Once clarified the programme theory, it is time to answer the second research question: how did the specific context of FVG influence the "programme theories" when put in practice? We obviously limit our analysis to the observed pattern of outcome, i.e. the policy's degree of implementation. Here, it might be argued that the short-mid term component of the policy (extra-funds to incentive structural and contractual form of cooperation) was successful. It might be noted that the lowest level of long-term commitment was largely preferred. When looking at this finding, this might be interpreted as the combined effect of strong local identities, which is a context specific factor, and the self-preservation attitude of officials that protect their turf. The latter is obviously not a local specific factor, but rather an evergreen in public management and policy. Although the incentives worked in the short run, they triggered primarily reversible arrangements: do these deals develop over time into tighter forms of integration? It is not possible at this stage of the process to assess the long term impact of the policy, which could be carried out only over time. Neither it is possibile, at the moment, to predict how localities will influence this outcome. Therefore, the big question, which underlies some relevant theoretical aspects as well, is to understand if strategic behaviors (such as gaming for the funds, Hood, 20006 and other strategic responses to institutional processes, Oliver, 1991) and resilience of previous arrangement (Peters, 2011) will prevail vis-à-vis the programme theory. Furthermore, beside the long term policy's achievements, a number of research
A8
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
questions, pretty relevant also for practice, might be addressed in the future: is the potential for synergies or political affinity a better predictor of inter-municipal coordination? What is the relative role of politicians and managers in deciding the patterns and implementing inter-municipal coordination? What is the pattern of change vs. resilience in processes and routines in the municipalities that cooperated or merged?
As stated in the introduction, the analysis of FVG policies, beside producing locally valid explanations, provides the opportunity to attempt to extend findings' validity beyond the research site. FVG, as mentioned earlier, represents a polar case since it enjoys a higher degree of lawmaking power also in defining the legal and financial framework where local governments operate. Therefore, it is a suitable site to discuss the classic argument that autonomization leads to better performance via differentiation: FVG should have been able to differentiate its policy leading to better performance (in this case enhanced coordination and efficiency). In the light of the findings, the argument needs at least refinement (Garlatti, 2014). FVG could differentiate the policy in the sense of opting for market-type incentive to cooperation, but not in the sense of favoring long-term commitment to coordination (for example financing only non-reversible form of coordination, like mergers). Therefore, the policy under analysis did not transform the pre-existing institutional landscape in a radical way. In the long term, this path-dependence could hinder the whole reform mechanism: municipalities could have been merely extra-financed for a number or years to go back to a fragmented landscape afterwards.
References
1.	Bunch, B. S., & Strauss, R. P. (1992). Municipal consolidation: An analysis of the financial benefits for fiscally distressed small municipalities. Urban Affairs Quarterly, 27, 615-629.
2.	Barlow, I. M. (1991). Metropolitan governance. London: Routlegde.
3.	Capano, G. 2003. 'Administrative Traditions and Policy Change: When Policy Paradigms Matter. The Case of Italian Administrative Reform During the 1990s', Public Administration, 81, 4, 781-801.
4.	Dolan, D. A. (1990). Local government fragmentation: Does it drive up the cost of government. Urban Affairs Quarterly, 26, 28-45.
5.	Downs, A. (1994). New visions for metropolitan America. Washington, DC: Brookings Institute
6.	Fedele P., Ongaro E. (2008) 'A common trend, different houses: a comparative analysis of devolution in Italy, Spain and the United Kingdom', Public Money & Management, vol.2, n.2; 85-92
A9
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
7.	Fleischmann, A., & Green, G. (1991). Organizing local agencies to promote economic development. American Review of Public Administration, 21, 55-67.
8.	Felbinger, C. L. (1984). Economic development or economic disaster. In R. D. Bingham & J. P. Blair (Eds.). Urban economic development (pp. 223-244). Beverly Hills: Sage.
9.	Garlatti, A., Lombrano, A. Zanin, L. (2013) Relazioni interaziendali nel governo locale, Forum Editrice, Udine
10.	Garlatti (2014) (a cura di) : Autonomia istituzionale e performance aziendale nei sistemi di pubblico impiego. Il caso del Friuli Venezia Giulia, Franco Angeli, Milano.
11.	Glaser, B.G.,Strauss, A.L. (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research, Chicago, Aldine Publishing Company
12.	Hood, C. (2006) Gaming in Targetworld: The Targets Approach to Managing British Public Services, Public Administration Review, Volume 66, Issue 4, pages 515-521
13.	Hood, C. (2005) The Idea of Joined-Up Government: A Historical Perspective. In Bogdanor, V. (2005) Joined up Government, Oxford University Press, Oxford
14.	Lind, M. (1997). A horde of Lilliputian governments. New Leader, 80, 6-7.
15.	Lindblom, C. (1959) The science of muddling through, Public Administration Review, Vol. 19, No. 2 (Spring, 1959), pp. 79-88
16.	Nunn, S., & Rosentraub, M. S. (1997). Dimensions of jurisdictional cooperation. Journal of the American Planning Association, 63, 205-219.
17.	Oliver, C. (1991): Strategic Responses to Institutional Pressures. Academy of Management Review, Vol.: 16, No.: 1, pp.: 145-179.
18.	Ongaro, E. (2011), The role of politics and institutions in the Italian administrative trajectory reform, Public Administration Vol. 89, No. 3, 2011 (738-755)
19.	Ongaro, E., 2010. The Napoleonic administrative tradition and public management reform in France, Greece, Italy, Portugal and Spain. In Painter M. and B. Guy Peters (2010) Tradition and Public Administration. Basingstoke: Palgrave Macmillan, 174-190
20.	Orfield, M. (1997). Metropolitics: A regional agenda for community and stability. Washington DC: Brookings Institute Press and Cambridge: Lincoln Institute for Land Policy.
21.	Ouchi, W., G. (1980)Markets, Bureaucracies, and Clans Administrative Science Quarterly Vol. 25, No. 1, pp. 129-141
22.	Pawson, R. and Tilley, N.,(1997) Realistic Evaluation, Sage,
23.	Peters, B.G., (2011) Institutional Theory in Political Science 3rd Edition, London: Continuum
A10
Journal of Universal Excellence,	Appendix
June 2015, year 4, number 2, pp. A12-A36.
24.	Peters, B.G., (2008). The Napoleonic Tradition, International Journal of Public Sector Management, 21 (2), 118-132.
25.	Peters, B.G., (1998) Managing horizontal government: the politics of coordination. Public Administration Vol. 76,pp 295-311
26.	Pollitt, C. and Bouckaert, G. (2011) Public management reform : a comparative analysis, Oxford, Oxford University Press.
27.	Painter, M. and Peters, B.G. (2010) The analysis of administrative tradition in Painter M. and B. Guy Peters (2010) Tradition and Public Administration. Basingstoke: Palgrave Macmillan, 3-16
28.	Pressman, J.L. and Wildavsky, A. (1984) Implementation. How Great Expectations in Washington Are Dashed in Oakland; Or, Why It's Amazing that Federal Programs Work at All, This Being a Saga of the Economic Development Administration as Told by Two Sympathetic Observers Who Seek to Build Morals on a Foundation, University of California Press
29.	Regione Autonoma Friuli Venezia Giulia (2014), Piano di Valorizzazione Territoriale(Local Enhancement Plan)
30.	Rigos, P. (1995). Metropolitan governance policy: Forty years of experimentation, research, and ideological conflict. Paper presented at the annual meeting, Southern Political Science Association, November, Atlanta.
31.	Thompson, G., J. Frances, R. Levacic and J. Mitchell. 1991. Markets, hierarchies and networks. London: Sage.
32.	Verhoest, K., Bouckaert, G., Peters, B. G. (2010)Janus-faced reorganization: specialization and coordination in four OECD countries in the period 1980—2005, International Review of Administrative Sciences September 2007 vol. 73 no. 3 325-348
33.	Zanin, L. (2013) Sviluppo e criticità delle Associazioni Intercomunali: analisi di un caso, in Garlatti, A., Lombrano, A. Zanin, L. (2013) Relazioni interaziendali nel governo locale, Forum Editrice, Udine
***
Andrea Garlatti is full professor of Public Management and Accounting at Udine State University, Department of Economics and Statistics. He is director of CIWE (Interdepartmental Center of Research on Welfare) of the same University.
Luca Brusati is associate professor of Management at Udine University, Department of Economics and Statistics. His research interests are public management reform, strategic management and interactive governance.
Paolo Fedele is assistant professor of Management at Udine University, Department of Economics and Statistics. His research interests are public management reform, strategic management and interactive governance
A11