By Germà Bel (Universitat de Barcelona) and Marianna Sebő (Universitat de Barcelona)
Growing skepticism expressed by local governments towards private-sector participation in public service provision has led many local authorities to experiment with new forms of public service delivery. In recent decades, one of the alternatives most frequently adopted has been inter-municipal cooperation (IMC). IMC is seen as a tool that can lower costs by exploiting economies of scale, while maintaining greater government control over production, something that is not readily achievable with privatization. Further benefits of IMCs include the enhanced cross-jurisdictional coordination, service quality and inter-municipal reciprocity. Concerns over stability, equity and universality may also stimulate cooperation.
Thus, cost savings is one of the key rationales for intermunicipal cooperation, if not the most important one, in practice. Yet, empirical evidence does not always confirm cost savings as an effect of cooperation. What’s more, data-based studies conducted in (in chronological order) in Spain, Norway, the Netherlands, Italy, France, Czech Republic, Sweden, Germany and USA report divergent results. Thus, the main motivation for our research was to account for the divergence in the outcomes reported for IMC agreements, and to explore the factors that best explain this variation.
Subsequently, we had to choose a methodology that can explain reported differences, through synthesizing previous findings. To do so, we chose the strategy of the meta-regression methodology. This approach has been widely used, for example, in psychotherapy and in studies of expectations and different types of elasticity, as well as to analyze the effects of privatization on the costs or productivity of local public service delivery, among other topics. Following this line, before specifying our model, we had to construct our database which included all studies on the effects of IMC. We examined academic literature, unpublished studies and “gray literature”. In all cases we used the following key-words on the websites: ‘inter-municipal cooperation’, ‘interlocal cooperation’, ‘interlocal contracting’, ‘joint contracting’ and ‘shared delivery’.
Next, we had to make sure the homogeneity of the definition of variables included. To do so, we needed to exclude studies that didn’t include costs (but rather efficiency for example) or where IMC was not defined homogenously. After all these refinements, we ended up with a database comprising 18 published and unpublished studies, with a total of 111 estimations.
While the model itself contained environmental and technical variables, we payed special attention to variables that reflect hypotheses from theories on the relationship between consts and IMC. From the theoretical point of view, we examined the effects of IMC on costs via the following channels:
- Since the IMC can help to approach the optimal boundary levels with the lowest average costs for the provision of a given service, economies of scale can be enhanced. Most empirical papers report that as the size of the municipalities grows, IMC tends to be less cost-advantageous. IMC is especially beneficial for smaller municipalities since they are less likely than larger municipalities to use competitive bidding.
- If delivering a service requires specialized investments, and if performance measurement is difficult, that service will incur high transaction costs. However, sharing the service-related transaction costs between the participants can contribute to the cost-saving effect of IMC. Alternatively, it can be argued that participants will incur information and coordination costs, negotiation and division costs, enforcement and monitoring costs and bargaining costs which the IMC might not be able to improve.
- IMC governance can take the form of informal agreements, formal contracts between the parties, joint-bodies for governance, or the delegation of power and resources to supra-municipal bodies. When deciding how to organize the IMC, delegation seem to have advantages in guaranteeing the participants’ joint interests and managing cooperation in different services. Because of this, economies of scale and scope in monitoring and coordination can be exploited. This course of action is frequent in Spain – comarcas and mancomunidades, France – communautés and, more recently, in Italy.
We found that studies conducted in municipalities with small population sizes tend to find IMC more cost advantageous, which is consistent with the hypothesis in relation to scale economies. Moreover, studies in which the governance of cooperation is delegated to supra-municipal governments also tend to report IMC more cost advantageous. However, we did not find any significant overall relationship between service-related transaction costs and the cost advantages of IMC.
Next, to understand better the effects of transaction costs we extended our study with the aim to disentangle the various facets of its components. We classified the service-related transaction costs, based on the nature of a given service, focusing on its asset specificity; and, on the other, ease of measurement. We took survey-based indicators first from Brown and Potoski (2005) but we didn’t get significant results. Next, using measures reported by Hefetz and Warner (2012), a somewhat different picture emerged. Both asset specificity and contract management difficulty are significant and were associated with IMC cost advantages. Still, these results should be treated as preliminary finding and more empirical research on transaction costs and IMC is encouraged.
We have not been able to consider here questions of service quality, given that the empirical evidence is extremely scarce. Yet, we believe the main implication that can be drawn by policy makers from our results is that ‘one size does not fit all’: IMC can be cost-advantageous for some services, but not for others. The possibility of exploiting scale economies, particularly in the case of small municipalities, seems to be robustly associated with cost savings. Moreover, the way the governance of the cooperation is arranged matters, highlighting the need to carefully consider the coordination and supervision costs involved.
Germà Bel is Professor of Economics and Public Policy in the Department of Econometrics, Statistics and Applied Economics at Universitat de Barcelona (UB). Has been visiting professor at Princeton u. and Cornell U., and visiting researcher at Harvard U. Cornell U., EUI at Firenze, and U. de Paris-Sorbonne. He is the director of the Observatory of Public Policies Analysis and Evaluation at UB. His research focuses on public sector reform, and pays special attention to local public services, and of infrastructure and transportation.
Marianna Sebő is a PhD candidate in Economics at Universitat de Barcelona (UB), and fellow researcher and the of Governments and Market research unit at UB. Her research focuses on interaction between Government and Markets, and on Games, Behavior and Incentives. Her academic interests include public policy analysis, economics of regulation and industrial organization. She is currently working on topics concerning public service delivery in Barcelona and her metropolitan area.