Beyond privatisation and cost savings: alternatives for local government reform

ABSTRACT Unsatisfactory results from privatisation have caused local governments to seek alternative reforms. Inter-municipal cooperation, mixed public/private delivery and contract reversals are three alternatives that have gained traction in the last decade. These alternatives help local governments manage markets for public service delivery as a dynamic process. They maximise government/market complementarities and address a wider array of public goals beyond cost efficiency concerns. The alternative reforms show how local governments balance citizen, labour and community interests to ensure efficiency, coordination and stability in public service delivery.


Introduction
The special issue published in 2007 (volume 33, issue 4) of this journal contained a set of articles which showed the debate on local government reform needed to move from a simple debate on privatisation to a broader discussion of alternative reforms. Reforms beyond privatisation were increasingly experienced worldwide, as shown by the expansion of municipal corporations in Portugal (Tavares and Camöes 2007), dynamic market management in different countries (Bel and Fageda 2007;Dijkgraaf and Gradus 2007), relational contracting (Brown, Potoski, and Van Slyke 2007) and reverse privatisation in the US (Hefetz and Warner 2007). In the decade that has elapsed since 2007, alternative reforms to privatisation have expanded, in practice as well as in scholarly analysis.
Inter-municipal cooperation, mixed public/private delivery and reverse privatisation have become much more common production choices over the last decades. Together they illustrate the importance of market management to secure benefits from alternative forms of service delivery. In this updated overview we present the latest empirical analyses of these reforms, exploring factors driving them and their outcomes, especially the evidence on costs. We find cost is but one concern local governments face. We highlight the critical role local governments play in managing markets, public values and community interests, especially labour.

Inter-municipal cooperation
While privatisation was based on the benefits of competition, inter-municipal cooperation reflects the importance of strategic collaboration as local governments form networks to achieve economies of scale and reduce negative externalities (McGuire and Agranoff 2011;. Inter-municipal cooperation in service production encompasses a variety of forms. To delimit different types of cooperative arrangements, Feiock and Scholz (2010) emphasise autonomy, viewed as the ease of entry and exit from a collaborative agreement. In Europe, these types of agreements can take the form of joint corporations, in which municipalities share ownership and production. The collaborative may jointly produce one or several services, contract to one of the members, or contract to an external partyeither for profit or for non-profit. While intermunicipal contracting is more limited in Europe, in the US it is more common than privatisation (Kim and Warner 2016;Hefetz, Warner, and Vigoda-Gadot 2012).
Cooperation has expanded in the last decade.  conduct a metaregression analysis on the existing multivariate empirical studies on factors explaining cooperation. While only 15 studies had been published on that topic before 2007, in the last decade (between 2007 and 2016) 34 studies were conducted (almost all of them published). While all 15 studies prior to 2007 were conducted for the US (but one for Mexico), the literature in the last decade has included analysis for Argentina, Brazil, France, Germany, Italy, Sweden, Portugal and Spain. While practically all pre-existing studies on factors explaining cooperation followed the methodology used to study factors explaining privatisation (see Fageda 2007, 2017), more recent research is developing unique methodological and empirical approaches to address cooperation in its own right. These include differences in factors driving cooperation (Kim 2018;, and use of alternative modelling approaches such as a hazard model in Germany (Bergholz 2018) or multi-level modelling to study duration of agreements in the US (Aldag and Warner Forthcoming).
European studies treat cooperation as a service delivery alternative to privatisation, but driven by similar objectives, primarily cost reduction. US studies give attention to a much broader array of motivating factors, beyond cost reduction, related to spatial location and organisational factors. Countries with fragmented local government face the challenge of regional coordination to address spillovers and externalities of service delivery, and intermunicipal cooperation is one response.
Empirical studies of inter-municipal cooperation find it is an important alternative to for-profit contracting (Brown, Potoski, and Van Slyke 2008;Hefetz, Warner, and Vigoda-Gadot 2012). While competition and efficiency are the primary factors driving privatisation, inter-municipal contracting is based on cooperation and is focused on longer term community concerns (Kim 2018). Inter-municipal cooperation also is used as a market management strategy to enhance competition in uncompetitive markets (Girth et al. 2012;Warner and Hefetz 2012), to achieve economies of scale while maintaining public control (Hefetz, Warner, and Vigoda-Gadot 2014b), to enhance market power of local governments in private contracting (Bel, Fageda, and Mur 2013) and to preserve service delivery in the face of fiscal stress, especially after the Great Recession (Kim and Warner 2016).
Scholarly analysis of the economic effects has expanded, particularly for European countries. In Bel and Warner's (2015) review, a total of eight empirical studies on cooperation and costs were found in the literature. Only one of them (Bel and Costas 2006) for Catalonia-Spain had been published before we wrote Bel, Hebdon, and Warner (2007). The other seven were conducted for Norway, Netherlands, Italy and other regions of Spain.
Most work in the US is based on case studies and finds cooperation results in cost savings less than half the time (Holzer and Fry 2011). A quantitative analysis of the economic effect of cooperation conducted for New York State by Bel, Qian, and Warner (2017) found similar results. Six additional studies on cooperation and costs have been identified for Germany, Czech Republic and, again, for the Netherlands, Norway and Spain. In short, in only one decade, fourteen more multivariate empirical studies on the effects of cooperation on costs have been conducted.
Empirical studies typically find that cooperation reduces costs, particularly when services with scale economies and small municipalities are involved. However, transaction costs can be higher than potential savings for services where scale economies are small, and municipalities were already operating close to their optimal production dimension. When cooperation is not primarily driven by cost savings, as in the US, then cost savings are less likely to be found.

Mixed delivery
Mixed delivery occurs when a government provides a service through both public and contracted delivery. This may involve dividing the jurisdiction into several districts or dividing the service into several parts and using public delivery in some and contracted delivery in others (Warner and Bel 2008;Warner and Hefetz 2008). According to Miranda and Lerner (1995), this 'redundancy' in delivery methods can be efficient because it makes benchmarking possible with the private sector, and also promotes bureaucratic competition in house. Thus, mixed delivery can promote competition by introducing pressures on public firms and by preventing monopolisation by private firms (Miranda and Lerner 1995;Girth et al. 2012;Bel, Brown and Warner 2014).
Mixed delivery is very common in the US and grew along with private contracting as a means to manage contracts (Warner and Hefetz 2008). In contrast, mixed delivery is less frequent in Europe (Warner and Bel 2008) where mixed public/private firms are more common. In this type of organisational form, the government and its private partners share ownership of the firm (Warner and Bel 2008;Cruz et al. 2014). The model is also referred to as 'partial privatisation' (Bel and Fageda 2010).
The rise in mixed delivery seeks to achieve the benefits of both market and public delivery (Warner and Hefetz 2008): private firms are interested in efficiency and profits; while also being interested in efficiency, the public sector is expected to provide fail-safe delivery and public accountability and involvement. There are few empirical studies on the effects of mixed delivery on service costs. Miranda and Lerner (1995) was the first study to address this issue. They found US cities with a higher percentage of services provided by mixed delivery incurred lower expenditures than those with a higher percentage of direct production. Recently, Bel and Rosell (2016) studied the effects of mixed delivery and costs in a single service (metropolitan bus) within one jurisdiction (Barcelona). They found private delivery is more costly than public delivery within this mixed delivery system, and that competition for contracts helps reduce costs.
Recent studies of mixed delivery have moved beyond a singular focus on costs and instead focus more broadly on the challenges of market management. Reform is no longer considered to be a simple choice to 'make or buy', but also the strategic management choice to both 'make and buy'. Hefetz, Warner, and Vigoda-Gadot (2014a) draw from the private sector management literature on 'concurrent sourcing', to build a theory for public sector mixed contracting. They argue that public managers face a broader range of contracting agents (both private for-profit and public inter-municipal) than private sector managers. In a study of 1474 US municipalities they found mixed delivery is more common with for-profit agents (38% of contracts) than among inter-municipal agents (13% of contracts). They argue that mixed delivery is a strategy to mitigate potential contracting risks, especially with for-profit agents where goal congruence and principal agent problems are more pronounced. Mixed delivery is used to promote market complementarities, and ensure attention to citizen interests, while seeking to reduce costs.
There are only a few studies that have analysed the determinants of mixed delivery and typically these look at mixed delivery along with for-profit and intermunicipal contracting: Brown, Potoski, and Van Slyke (2008); Warner and Hefetz (2008), Hefetz, Warner, and Vigoda-Gadot (2014a). Mixed delivery is generally found to be a strategy in service markets that face low competition (Girth et al. 2012). Mixed delivery also has been shown to reduce contract reversals )a point that will be explored in more detail in the next section.

Contract reversals
Contracting may involve movement towards market or back towards public delivery in a dynamic process over time. In the 2007 special issue, Hefetz and Warner (2007) noted that contracting reversals were part of a broader strategy of market management to ensure social choicea balance between market, public and citizen interests. Since that time, additional work in the US has found contract reversals to be part of a pragmatic managerial process (Hefetz and Warner 2016;Warner and Hefetz 2012). By contrast, in Europe and the global south, case study research has given new attention to this contract reversal process and a new name, 're-municipalisation', arguing that it is primarily political in motivation (McDonald 2016;Hall, Lobina, and Terhorst 2013). By contrast, the European papers using quantitative analysis (Chong, Saussier, and Silverman 2015;Gradus, Dijkgraaf, and Wassenaar 2014;Pérez-López, Prior, and Zafra-Gómez 2015) find reversals to be a strategic element of pragmatic market management, often driven by problems with higher costs and lower quality of service under privatisation. The most recent US data finds reversals more common in for-profit contracts than in inter-municipal contracts (Warner 2016). While case study research suggests remunicipalisation is an increasing trend, the US data show that contract reversals are not increasingthey are about the same level as new contracting outsuggesting a continuing market testing strategy as managers negotiate cross boundary interactions seeking complementarities as outlined by Hefetz (2016).

Managing union and community interests
While unionisation is often considered to be an obstacle to local government reform, most studies use private sector or national-level proxies and do not have local measures of public sector unionisation. Those few studies that have direct measures of unionisation find a more nuanced story. In a 2001 study of New York State, Warner and Hebdon (2001) found unions were only a negative factor when local governments implemented a single form of restructuringeither privatisation or inter-municipal cooperation. When more complex restructuring took place, for example, involving reverse privatisation and entrepreneurship, the union factor was not significant. One possible explanation for this finding was that job losses in single restructuring cases might be offset by job gains when a more complex mix of delivery alternatives is used. Jalette and Hebdon (2012) opened the union 'black box' and examined the actual strategies that unions took in the face of threatened job losses among local governments in Canada. They found that unions employed a range of strategies from strikes, legal actions, negotiations and suggesting alternatives, to acquiescence. Surprisingly, negotiations and suggesting alternatives were both more common and more effective than strikes. Also, the existence of the acquiescence strategy (support or no reaction), not often documented in the literature, also supported the idea of union practicality.
The most recent US studies have found unionisation has no effect on forprofit contracting but has a suppression effect on inter-municipal cooperation (Kim and Warner 2016)possibly because of the difficulty of crafting cooperative agreements across different public sector unions. Most studies use labour opposition to alternative service delivery, rather than actual measures of unionisation. These studies typically find no effect on for-profit contracting (Kim and Warner 2016;Hefetz, Warner, and Vigoda-Gadot 2014b), a positive effect on for-profit contracting (Lobao and Adua 2011) and a negative effect on contract reversals (Hefetz and Warner 2016), as professional managers understand how to manage labour interests alongside the other interests local governments must balancecitizen, budget and community concerns (Kim and Warner 2016). These results suggest the power of unionisation may be inside the organisation in forcing more professional management practices. Hefetz and Warner (2016) find unionised governments are more likely to monitor contracts, less likely to contract out in non-competitive markets and more likely to pursue cooperative labour management relations. Kim and Warner (2016) call this 'pragmatic municipalism'.

Conclusion
Market-based reforms rely on competition to achieve cost savings. But cost savings are hard to secure as most local government markets are non-competitive and transactions costs are high. Local governments have been leading in exploring new alternatives that address these problems. Mixed delivery and contract reversals recognise the important market management role of local government in creating competition between the public and private sector over time. Inter-municipal cooperation moves beyond competition to explore benefits of economies of scale and scope.
While twentieth century local government reforms focused primarily on the benefits of competition, twenty-first century reforms look more to the need for market management and the benefits of cooperationacross local governments, across public and private sectors and across labour, community and government interests. As fiscal pressures on local government mount, this pragmatic ability to engage in cross sector collaboration is the key reform for the future.

Revisited series
In the Revisited Series, Local Government Studies offers short updates of some of the journal's most cited articles of recent years. In these updates, the authors reflect on changes since their original contribution, while underlining the continuing relevance of the thinking behind the latter and indicating the direction in which this could be extended in the future.
Here, Germà Bel, Robert Hebdon and Mildred Warner revisit an article they published in 2007 in LGS Vol. 33 No. 4: 'Local government reform: privatisation and its alternatives'. In that article, they discussed the then existing empirical literature on the effects of privatisation and analysed how disappointment with cost savings and the challenges posed by contract management were leading local government policymakers to explore alternatives to privatisation, including municipal corporations, mixed delivery, contract reversals and relational contracting. The main conclusion of their analysis was that there was a need for a more comprehensive framework within which to study local government reform, giving attention to a wider array of alternatives. Here, they return to the issues raised in that paper, and discuss the main developments that have emerged as alternatives to local privatisation since 2007.

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributors
Germà Bel is Professor of Economics at University of Barcelona. His research focuses on public sector reform, privatisation regulation and competition, with special emphasis on local public services, infrastructure, environment and transport.
Robert Hebdon is Professor Emeritus, Organizational Behavior at McGill University's Faculty of Management, where his research focuses on dispute resolution, privatisation and workplace conflict.
Mildred Warner is Professor of City and Regional Planning at Cornell University in New York, USA. Her research focuses on local government service delivery, fiscal stress and policies related to economic development, the environment and social services.