Strasbourg, le 1^er décembre 2009                                                           LR-FS(2009)6

                                                                                                                         

COMITE EUROPEEN SUR LA DEMOCRATIE LOCALE ET REGIONALE

(CDLR)

COMITE D’EXPERTS SUR LES FINANCES AUX NIVEAUX LOCAL

ET REGIONAL ET LES SERVICES PUBLICS

(LR-FS)

ATELIER DE TRAVAIL

« ECONOMIES D’ECHELLE RESULTANT DE L’ACCROISSEMENT DE

LA TAILLE DES COLLECTIVITES LOCALES : QUELLES PREUVES ? »

Note du Secrétariat

établie par la Direction générale de la

démocratie et des affaires politiques

Direction des institutions démocratiques

This document is public. It will not be distributed at the meeting. Please bring this copy.

Ce document est public. Il ne sera pas distribué en réunion. Prière de vous munir de cet exemplaire.

Introduction

A sa dernière réunion (6-7 avril 2009), le Comité LR-FS a décidé de tenir, lors de sa deuxième réunion de 2009, un atelier sur deux thèmes apparentés :

1. les économies d’échelle résultant de l’augmentation de la taille des collectivités locales ;
2. le choix politique entre la promotion de la coopération entre les collectivités locales et celle de leur fusion (voir document LR-FS(2009)7).

En ce qui concerne le premier sujet, il a été décidé de se concentrer sur la collecte et l’examen des données issues des études réalisées sur la question. L’atelier aura pour but de présenter les possibilités réelles de parvenir à des économies d’échelle, sur la base des indications disponibles et d’exemples concrets.

Il était envisagé qu’un expert consultant élabore un document regroupant ces éléments, en collaboration avec les membres du Comité originaires des pays où les travaux considérés ont été menés. Malheureusement, en raison de réductions (temporaires) du personnel et de la préparation de la 16^e session de la Conférence ministérielle, le Secrétariat n’a pas été en mesure d’assurer la coordination de cette tâche. L’expert consultant, M. Jørgen Lotz (Danemark), a néanmoins élaboré, en vue de l’atelier, un document figurant en annexe, sur les économies d’échelle et les fusions de communes, s’appuyant plus particulièrement sur l’expérience danoise (anglais uniquement).

Lors de l’atelier, des éléments complémentaires seront apportés par M. Linze Schaap, de l’université de Tilburg (Pays-Bas), auteur de l’aperçu des questions d’actualité en matière de démocratie territoriale qui a servi de point de départ à la discussion informelle tenue par les ministres responsables des collectivités locales et régionales à Utrecht le 15 novembre, la veille de la 16^e session de la Conférence ministérielle. Ce document (MCL-19(2009)Inf2) sera envoyé séparément aux membres du LR-FS. M. Schaap mène actuellement une étude comparative sur la question des économies d’échelle.

De plus, tous les membres du LR-FS sont invités à présenter lors de l’atelier les résultats de recherches ou d’évaluations sur les fusions de communes dont ils auraient connaissance. Les membres disposant de tels documents sont priés de contacter le Secrétariat bien avant la réunion ([email protected]).

Action requise

Les membres sont invités à se préparer en vue d’une participation active à l’atelier. Les membres disposant de résultats de recherches ou d’évaluations sur les économies résultant de la fusion de collectivités locales sont invités à présenter ces informations dans le cadre de l’atelier. Ils sont également priés de contacter le Secrétariat bien avant la réunion.

Le Comité est invité, à l’issue de l’atelier, à examiner s’il convient de proposer au CDLR d’inclure dans le programme d’activités des travaux complémentaires sur cette question. Cette réflexion s’inscrira ainsi dans le cadre de l’examen plus général des nouvelles activités à entreprendre dans le prolongement de l’agenda d’Utrecht (voir point 4 de l’ordre du jour).

 

ECONOMIES OF SCALE AND MUNICIPAL AMALGAMATIONS

A survey with special attention to the Danish experience[1]

by Jørgen Lotz, Denmark. [email protected]

1.         The international findings

Byrnes (2002) presents three different methods to analyse economies of scale:

·                the “engineering estimates” involving the assembly of expert opinions regarding costs in a particular industry. This kind of evidence is subjective as managers will tend to defend their own plant size,

·                the “survivor technique” relying on actual trends in plant size rather than on opinions. This method is difficult to apply since the market is not efficient. Political decisions will tend to let also deeply inefficient plants survive,

·                estimating spending for the existing size distribution of plants, which may indicate an optimal size of a plant, this method “has formed the standard technique economists have employed for testing for scale economies in local government.”

Byrnes provides a survey of a number of such studies from UK and US from 1959-2001. He finds that:

·                29 per cent of the studies find evidence of U-shaped cost curves,

·                39 per cent find no statistical relationship,

·                8 per cent find evidence of economies of scale, and

·                24 find diseconomies of scale. 

He concludes that this reveals a “great deal of uncertainty about whether economies of scale exist in local government service provision.”

Byrnes also mentions several points of criticism: the studies use population as measure for output disregarding the demographic composition, they do not assign overhead costs to specific products, they do not differentiate between plant-level and firm level economies, and none of the studies mention the long run.  

To these observations could be added that economies of scale are difficult to demonstrate empirically because they are not always reaped as budget savings but are instead in the larger municipalities transformed into what is called “better service”, or they may already have been realised by the smaller ones entering into inter-municipal cooperation arrangements. Another possible source of bias is the problem of separating the effects of “size” from the effects of other drivers of municipal spending – demographic, social or geographic. And “scale economies” are not likely to result if small municipalities with low population density merge with other similar municipalities. These factors mean that economies of scale may exist even if they cannot be demonstrated based on present spending data, and this may explain the unclear results from the survey conducted by Byrnes.

2.         The Danish experience, the results

By a major structural reform of Danish local governments the number of municipalities was in 2007 reduced from 270 to 98, doubling the average size to 55.000 inhabitants. The objective was first of all to get better quality of municipal service, but also the argument of benefiting from economies of scale was used by the government.

The reform sparked much interest in the question of whether there are economies of scale and how strong they may be. A number of Danish tests, official as well as from the academic world, were published. Danish studies benefit from the advantage of central recording of expenditure data based on the same detailed, economic and functional nomenclature across all municipalities. 

The following table shows the resulting minimum optimal “firm-size” of some of these studies, some on the administrative spending of the municipalities, and some on the municipal spending on primary education.

Study (see list of reference)	“Optimal” municipal size total	Administration	Schools
Indenrigsministeriet (2000) and Møller et.al (2001)	18-25.000	18.000	50.000-75.000
Houlberg (2000)	30-35.000	30.000-50.000	100.000
Finansieringsudvalget (2004)	35-45.000	*	*

None of the studies based their conclusions on estimates of a continuous expenditure function[2], they preferred to use the method of grouping municipalities according to size – though different groupings used in each study – and test the significance of dummies applied to each group.

3.         The Danish experience: plant or firm specific studies

Most of the many Danish tests were for reasons of data availability of the so called “firm-specific effects” comparing municipal spending with the size of municipalities. Recently some Danish tests have been published on “plant-specific effects” comparing the unit costs and size of municipal institutions. These studies tend to suggest weaker gains from amalgamations than did the studies of firm-specific effects. For example, it was shown that big municipalities do not have proportionally more large schools than the small ones. Such studies are in particular interesting for spending on local administration and for spending on primary schools.

As to the studies of the economies of scale for spending on municipal administration the function “administration” is the relevant institution the whole municipality and firm- and plant-effects melts together.

Møller (2001) speculates that one factor having possibly contributed to an explanation for a development of economies of scale over time has been that more and more specialised and complicated functions have been delegated to the municipalities. Groes (2005) suggests that another explanation could be the new technology (IT) reducing the costs of administration relatively more in big units than in smaller ones.

The conclusions of Ministry of Interior and Houlberg have been challenged by Blom-Hansen (2005).  He argues that there is a bias in these studies because they are based on administrative costs as recorded in the municipal accounts. But practice differs, some municipalities centralise administration, others decentralise administration to the institutions. This means that the salary of an employee working on budgeting and accounting in an old age home is recorded as spending for old age care while the salary of an employee doing the same work but stationed in the city hall is recorded as an administrative cost. Blom-Hansen managed to compile data including also some of the decentralised employees and based on these adjusted data he found no economies of scale. Actually he found diseconomies of scale, the administrative costs per inhabitant increased with the size of the municipality.  Even though Blom-Hansens data may not yet be perfect they seem anyway better than those based on budget data.

In conclusion: the Commission on Administrative Structure (Strukturkommissionen 2004, p. 96) concluded that “economies of scale are in particular relevant for the administrative functions”.  However, Blom-Hansen suggests that these studies only confirm the intuitively appealing suggestion that large municipalities decentralise administration more than small municipalities.

For primary schools all studies found declining costs per inhabitants, the Ministry of Interior up to a size of 50.000-75.000, and Houlberg up to 100.000 inhabitants.  The Commission on Administrative Structure (Strukturkommissionen 2004) refers to several other studies including a DEA analysis suggesting better quality and lower unit costs in large schools. The commission noteworthy adds (p. 370) that “it is not examined whether bigger schools would have the same effect disregarding size of municipality”.

Blom-Hansen (2004b) argued against these studies that the connection from municipal size to economies of scale is not direct, it depends both on large schools being more efficient and on big municipalities having relatively large schools.

The analysis by Blom-Hansen finds economies of scale among primary schools based on data for individual schools (plants), the effects are found in particular for building related costs and costs of school administration. As to the costs of teaching, the economies of scale are found only up to a rather small school size of 100-150 pupils. He then turns to the correlation between school size and size of municipalities and finds, when he controls for population density, that there is no such correlation.

Blom-Hansen concludes that there are for primary schools economies of scale related to size of municipality. But to find savings by amalgamation of municipalities the new municipalities would need to close small schools and accept the increased costs of transportation.

It is noteworthy that a new Swedish report (Färre kommuner 2006) about the small municipalities in Sweden argues that a reform like the one in Denmark will not help in countries like Sweden and Finland. In these countries the major problem for the small municipalities is a hugely lower population density than in Denmark.

4.         Do amalgamations result in realisation of economies of scale?

And economies of scale may be only one of the forces released with amalgamations. Others - like a new desire to improve quality - may work in the opposite direction. It is plausible that in some cases the spending level agreed upon in amalgamated municipalities will tend to follow the municipality with the highest spending in any merger.

Moisio and Uusitalo (2006) have studied the economic effects over a period of ten years following 41 municipal amalgamations that took place in Finland between 1970 and 1981.  They found that spending per inhabitant after ten years had increased more in the amalgamated municipalities than in their comparison groups. Only the spending on general administration decreased, but by far less than the increase in spending for other functions.  Moisio and Uusitalo also refer to earlier Swedish studies of expenditure growth in amalgamated municipalities supporting evidence of higher and not lower expenditure growth in merged municipalities compared to others. 

The Finnish study may be seen as a warning that economies of scale are not the only factor having an effect in cases of municipal amalgamation. Two points are in particular worth mentioning: The difference in results could also stem from the fact that the Finnish study is based on data from earlier years and that the existence of economies of scale may be a relatively recent phenomenon.

In DenmarkHoulberg (2008) has attempted to compare the same way as the Finnish study cost developments over time in amalgamated with non-amalgamated municipalities. His analysis included only changes one year after the reform – 2007 – and he warns that it is still too early to make statements on the effects of the amalgamation reform. His tentative findings are that the development in spending in amalgamated municipalities has not been different from that in non-amalgamated municipalities.    

Literature

Blom-Hansen, Jens and Larsen, Brian (2005): “Jo større, jo dyrere”. Nordisk Administrativt Tidsskrift 1/2005, 86. årgang.

Blom-Hansen, Jens (2004a): ”Stordriftsfordele i den kommunale serviceproduktion? Børnepasning som eksempel.” Nordisk Administrativt Tidsskrift 4/2004. 85. årgang. Pp. 277-293.

Blom-Hansen, Jens (2004b): ”Stordriftsfordele I den kommunale serviceproduktion? Skoleområdet som eksempel”. In Blom-Hansen, Jens, Nørgaard, Asbjørn Sonne and Pallesen, Thomas (eds) Politisk ukorrekt. Aarhus Universitetsforlag.

Byrnes, Joel and Dollery, Brian (2002): “Do Economies of Scale Exist in Australian Local Government? A Review of the Research Evidence” Urban Policy and Research, Vol. 20, No. 4, 391-414.

Finansdepartementet Stockholm (2006) ”Färre kommuner?” Arbetsmateriel 2006-02-01. 

Groes, Nils (2006): “Der skal knofedt til” in: 6 bud på kommunale udfordringer. Kommunernes Landsforening 2006. 

Houlberg, Kurt (2000). “Kommunale stordriftsfordele – hvor finder vi dem, og hvor store er de?” København. AKF Forlaget.

Houlberg, Kurt (2008). ”Har kommunerne realiseret stordriftsfordelene i forbindelse med Strukturreformen?” Administrativ Debat nr. 2, August 2008.

Indenrigsministeriet (2000). ”Størrelseseffekter i den kommunale sektor”. København. Indenrigsministeriet.

Moisi, Antti and Uusitalo, Roope (2006):”The impact of municipality mergers on local public expenditure in Finland” (unpublished)  [email protected]

Møller, Inge Lene and Niels Jørgen Mau Pedersen (2001) “Economies of scale in local governments – theoretical and empirical investigations of Danish municipalities” Paper for the 57^th Congress of International Institute of Public Finance in Linz, Austria, August 2001 ([email protected])

Appendix

Municipal co-operation[3]

Most of the municipal services are home-produced, the municipalities are suppliers as well as producers. In many cases there is no alternative, either because of security of supply, or because no alternative market exists, or the service is an expression of public authority which cannot without special legislation be delegated to others.  But there are many exemptions from this general rule.

In many cases independent producers make contracts with the municipality on the supply of services, this is quite common in the field of social services. Such supply is in the law regarded as public supply because of the strict conditions in the contracts on supervision and control by the municipality.

Municipal co-operation in the supply of other services like public utilities and highly specialised services was foreseen already in the 1970-reform of the structure of local government. Over the years, as the municipalities have been asked to supply more and more functions, and the legislation has become more complex, the municipalities have established co-operation on a wider selection of functions. There is now a rising interest among municipalities for more formalised co-operation and the average municipality now participates in about 30 such arrangements.

The Commission survey of cooperative arrangements The Commission report includes the result of a survey about municipal co-operation, defining co-operation as projects approved by the council. The survey reported 1835 co-operation arrangements, 36 per cent of these involved no more than two counties or municipalities, 77 per cent no more than five units. Municipalities larger than 100.000 inhabitants participate in more co-operation arrangements than the average. The small municipalities prefer co-operation on the supply of social and health related services. Among the larger municipalities co-operation among the technical services dominates relatively.       Another difference is that the larger municipalities typically are the suppliers of the coordinated service, the smaller municipalities typically cooperate as users and let others supply the service.

In evaluating municipal co-operation arrangements the Commission on Structural Reform found that such co-operation can enable smaller municipalities to obtain economies of scale and access to more specialised expertise. But the commission also concluded that cooperative arrangements tend to reduce the possibility to differentiate the service to the needs of the individual clients, and that cooperative arrangements tend to obscure responsibility and reduce the possibility for the citizens to have influence.

The conclusion of the commission was on this background that when a change in the distribution of functions is considered it cannot be advised to disregard the importance of the size of the municipalities.

It is not advisable to increase the number of functions delegated to the municipalities if the new municipal functions become too difficult to handle for the smallest municipalities so that it needs to be supplemented with recommendations for more cooperative arrangements.

The political agreement in Parliament on a structural reform states that the work of the Commission formed the basis for the political process that resulted in the agreement. The political decision was to base the reform on amalgamations aiming at sufficient size for the municipalities to deliver “high quality welfare to the Danish people”.