Abolishing the Act on System of Choice in Swedish eldercare: on arguments and replacements in the municipalities

Purpose – Since 2009 Swedish municipalities may apply the Act on System of Choice (LOV) in, among other things, eldercare. About half of the 290 Swedish municipalities have chosen this within home-care services for older citizens, thus creating conditions for a welfare mix where private and public providers compete. Some of these municipalities later made decisions to abolish LOV. This article aims to analyse the arguments put forwardbymunicipalpoliticianstoabolishLOVanddiscussesifthecaseofabandoningLOVrepresentsacaseofre-municipalization. Design/methodology/approach – Qualitative method was used to analyse decision protocols and media materials from 20 Swedish municipalities that had abolished LOV in home-care services. Findings – The article shows that politics and ideology seem to have only a limited significance in abolishing LOV. The most important arguments found in the empirical materials were instead pragmatic and related to the transaction costs: in smaller municipalities about the weak position of private providers and in larger municipalities about reported cases of welfare crime and extensive needs to control and review. In smaller municipalities, LOV was replaced by public monopoly and in larger municipalities by other types of procurements. Originality/value – Withitsfocusoneldercareinparty-dominatedmunicipalities,thearticleaddsknowledge to the literature on drivers of re-municipalization but also discusses possible delimitations of the concept of re-municipalization.


Introduction
Elements of new public management (NPM) and marketization have pervasively changed the Swedish welfare state and thus also the eldercare (Vabø and Szebehely, 2012;Feltenius, 2017;Moberg, 2017Moberg, , 2023;;Blix and Jordahl, 2021;Blomqvist and Winblad, 2022).In Sweden, eldercare, tax-funded and a municipal area of responsibility, mainly consists of nursing homes and home-care services [1].Since 2009 the Act on System of Choice in the Public Sector (LOV) has given Swedish municipalities the right to decide about introducing so-called "choice systems" in, among other things, eldercare.In 2023 more than half of the Swedish municipalities have a system of choice in operation in home-care services, where private and public providers compete in most cases.
Simultaneously international research pays attention to the fact that marketization and privatization in the public sector are not irreversible nor a lineal development (e.g., Bel et al., 2018;Voorn et al., 2021;Berlin et al., 2023).The case of LOV in Sweden is no exception.Only two years after the introduction of the new act, in 2011, one municipality decided to abolish user choice according to LOV in home-care services (henceforth referred to as abolishing LOV).To date, 23 municipalities have decided to abolish LOV.
In the literature, abolishing LOV has been discussed as a possible case of "re-municipalization" (Montin, 2016;Jordahl and Persson, 2021).The concept suggests that the provision of services returns from the private sector back to the municipal sectorthat is, a return to the status as before the privatization (B€ onker et al., 2016;Voorn, 2021).But then why does this return happen?International research makes a categorization in two different drivers behind re-municipalization: political drivers including ideological statements against privatization and private providers, and pragmatic drivers including transaction costs, management and quality (e.g., Clifton et al., 2021;Voorn et al., 2021;Albalate et al., 2022;Warner, 2023).According to the literature, pragmatic drivers appear to dominate regarding technical services and infrastructure.Some scholars suggest that ideology could be more evident as a driver concerning social services (Petersen et al., 2015;Gradus and Budding, 2020).
Hence, it could be assumed that political drivers would be prominent for abolishing LOV in eldercare in Sweden, a country also with party-dominated municipalities since municipal politics is organized through the political parties.With few exceptions, the parties in the Swedish municipalities are the same as at national level (Lidstr€ om, 2012).If political drivers are frequent, homecare would probably represent a clear-cut case of re-municipalization describing a transfer from the private sector back to the municipal sector.However, if it turns out that homecare is not that different from other sectors such as technical services, and that pragmatic drivers are salient, another important question arisesnamely, what happens afterwards?It might not necessarily imply a total "backsourcing" of the provision (see Warner, 2008).Rather, other regulatory systems for sustaining private elements could be introduced, thus with the possible preservation of a "welfare mix" of both private and public providers (Ascoli and Ranci, 2002).
We argue that it is urgent not only to explore drivers behind the abolition of a regulatory system for privatization such as LOV.It is also important to investigate more in detail what replaces it (cf.Voorn et al., 2021).Instead of returning to the previous public monopoly, it might be the case that marketization is maintained by other regulatory tools.If so, the abolition of LOV as a possible case of re-municipalization might be questionedat least based on the current definition of re-municipalization, with a focus exclusively on ownership and provision.
In this article, we examine the abolition of LOV in home-care services in Swedish municipalities.We address two questions: First, what are the arguments for the abolition of LOV as stated in the municipalitiespolitical, pragmatic or both?Second, does the abolition of LOV, after also having considered what replaces it, represent a case of re-municipalization?
Next, we introduce the Swedish case.Thereafter the concept of "re-municipalization" is introduced.We then discuss our research method and empirical materials, after which we present our findings.In the final and concluding section, we discuss the results and our contribution to the literature.

The Swedish case
Sweden is divided into 290 municipalities with extensive areas of responsibility within welfare (Lidstr€ om, 2012).Local self-government is strong, also in statutory and mandatory social services such as eldercare.The principle of local self-government gives Swedish municipalities the right to decide about the marketization of, for example, eldercare and to choose between two different models for procurement (Feltenius and Wide, 2023).

IJPSM 37,2
The first model derives from the Public Procurement Act (LOU), originating in the 1990s and which, with its current design, is based on European Union Directive 2004/18/EC (Bretzer et al., 2016).In LOU, the contract is awarded to the provider with the most attractive bid concerning the quality as well as costs of servicesor one of the two.In Swedish eldercare, LOU has mainly been applied in the procurement of nursing homes (Feltenius, 2017;Segaard and Saglie, 2017;Lindmark et al., 2023).
The other model is based on the Act on System of Choice in the Public Sector (LOV), a national law from 2009.It is voluntary for the municipalities to apply LOV within a social service area, e.g., eldercare.In LOV there is freedom of establishment.All private providers who meet the requirements stated in the municipality's tender document are allowed to sign a contract with the municipality and then become eligible in a system of choice.The citizens who have been granted home-care services choose which provider will carry out their efforts (Feltenius and Wide, 2019).The municipalities cannot influence the number of providers or the market share of the various providers, for example, they cannot determine what proportion of users will have private homecare services.Instead, the distribution among public and private providers depends on the citizens' choices as well as the interest of private companies to apply for approval.The idea behind LOV was, among other things, that citizens' preferences and not political priorities should govern the organization of welfare services and accordingly strengthen the power of the individual citizen (Government Bill, 2008).In LOV, the approved providers receive a fixed sum from the municipality, based on, for instance, the number of granted hours of service provided each month.The citizen's fee is also fixed and paid to the municipality but covers only a minor portion of the total costabout 6% (Socialstyrelsen, 2014).The remaining cost is financed by the municipalities through local income taxes as well as central government grants.
In 2010, the first 45 municipalities implemented LOV into operation.In 2023, 159 municipalities had LOV in operation in at least one social service area.Figure 1 shows the development over time.Statistics by type of service are not available over time for LOV in operation, which means that all service areas are included in the figure.However, almost all municipalities that have adopted LOV have done so within home-care services meaning that the figure gives a good overview of the development.There are only one or two municipalities per year that instead only have LOV within, for example, adult daily activities or companion service (SKR, 2018, 2019, SKR 2023a).In 2011, one municipality decided to abolish LOV in homecare; in 2014, two more municipalities followed.To date, 23 municipalities have made decisions to abandon LOV in home-care services.
Previous research has shown that until 2017 the abolishing municipalities had, on average, smaller populations, lower population density, lower income tax base, older inhabitants, and higher electoral support for left parties than municipalities with LOV in operation (Jordahl and Persson, 2021).In 2023, the small, rural municipalities have been joined by large, urban municipalities (SKR 2023b).
The concept of re-municipalization means that the ownership or provision of, for example, services return from the private sector back to the municipal sectora return to the status as before the privatization.This is evident from the definition provided by Bart Voorn: "remunicipalization refers to the return of previously privatized (or contracted-out) services to municipal authorities" (Voorn, 2021 p. 442).However, there has been some confusion in the literature regarding the concept.Judith Clifton and others have addressed this ambiguity and highlight two important aspects of the concept.Firstly, the development should occur exclusively at the municipal level and not, for example, as a transfer of responsibilities from the national to the municipal level.Secondly, there is a trajectory of transfer from the municipality to the private sector, and then back to the municipality again (see Clifton et al., 2021 pp. 295-297).
This development is also described as a "reverse privatization" since public services are moved away from privatization.It implies that the direction of privatization is not inevitable, linear or irreversible (Clifton et al., 2021).Already a decade ago, re-municipalization was frequent in European countries inamong other thingswater, electricity, public transport, waste management and housing (Hall, 2012).Research on local governments in the United States has shown similar patterns (e.g., Hefetz and Warner, 2004;Warner, 2008;Warner and Hefetz, 2012;Damanpour et al., 2020).
Re-municipalization can take several different forms.It includes the creation of municipally owned corporations, the repurchase of shares sold to private companies and the re-insourcing of outsourced services, for example after the expiration of concession contracts (Wollman, 2018;Voorn et al., 2021).Previous research on re-municipalization has mainly dealt with the sectors of energy, waste and water, as well as public transportation.Less research has dealt with social services (Hung and Lu, 2022).One reason might be that in many countries, unlike in Sweden, the municipalities' role in social services is limited.Yet, until recently, research on re-municipalization in a Swedish context has been sparse (but see Jansson et al., 2021;Jordahl and Persson, 2021;Berlin et al., 2022;Carlstr€ om et al., 2023).

Political drivers
Political drivers are based on ideological statements against private providers or private ownership in certain activities and services.For example, it may be an ambition to regain democratic control over critical infrastructure, such as water and energy, or not to let in forprofit providers in tax-funded public services (e.g., Hall, 2012;Clifton et al., 2021;Warner and Aldag, 2021).According to previous research, ideology does not have a substantial effect on re-municipalization (Hanna and McDonald, 2021;Berlin et al., 2023;Warner, 2023).However, it must be noted that party politicization at the local level is low in countries investigated such as the United States (Warner and Aldag, 2021;Warner, 2023) while also social services have been largely neglected in previous research.Instead, the literature has been devoted mainly to technical services and infrastructure such as water and waste services (see Petersen et al., 2015;Gradus and Budding, 2020;Hanna and McDonald, 2021).
In Sweden, privatization in social services has long been an ideological issue and a divider in politics.A common argument by left-wing parties is that tax-financed activities should be carried out by the public sector and not by for-profit companies.Liberal and conservative parties are in general more positive about privatization (Stolt and Winblad, 2009;Guo and Willner, 2017;Hardell et al., 2020), accordingly influencing privatization at the local level (Elinder and Jordahl, 2013a;Lindh and Johansson Sev€ a, 2018).
Another argument is that privatizations challenge the universalistic and collective core of the Swedish welfare modelthat is, accessible, inclusive and equal social services of high quality (e.g., Vabø and Szebehely, 2012).A system of choice might instead contribute to increased inequality since citizens have different conditions to make informed choices and the services of the providers may vary (Moberg et al., 2016;Moberg, 2017Moberg, , 2023)).

Pragmatic drivers
In the literature, pragmatic reasons behind re-municipalization mainly concern management issues and aspirations to reduce transaction costs (e.g., Clifton et al., 2021;Voorn, 2021;Warner and Aldag, 2021;Albalate et al., 2022;Berlin et al., 2023;Warner, 2023).More specifically, pragmatic reasons include efforts to possible efficiency gains, cost savings and quality improvements as well as problems with contract management, monitoring and control (e.g., Voorn, 2021 for a review).Additionally, the literature mentions the absence of private providers and a low level of competition (Jansson et al., 2021).
The transaction costs for privatization in social services should not be underestimated.There are costs of monitoring and evaluating the private providers as well as the contract management itself.If these costs are too large, it may be more economically beneficial to perform the services "in-house", i.e., only by the public provider (Elinder and Jordahl, 2013b).
In Swedish municipalities with LOV in home-care services, the transaction costs may be unreasonable if there are just a few private providersor none (Jordahl and Persson, 2021; Abolishing the Act on System of Choice Jansson et al., 2021;Feltenius and Wide, 2023).Even without any private providers, the costs of administering the choice system remain, for example, to keep the tender document and routines up to date.The same applies if there are just a couple of citizens who choose private providers.
The costs for monitoring and control to ensure quality and avoid sector failures and misconduct may as well be extensive (e.g., Hall, 2012;Jansson et al., 2021).Previous investigations of eldercare have discussed misconduct among private providers as well as major "failures" in eldercare, for example, fraud and elder neglect and abuse (Klingberg Hjort and Martinsson, 2021;Berlin et al., 2022).

Data and methodology
Identifying the municipalities We used data from the Swedish Association of Local Authorities and Regions (SKR) to find municipalities that have decided to abolish LOV (SKR, 2020(SKR, , 2023b)).Thereafter, to ensure that the decision concerned home-care services and not another area, we contacted each municipality and asked for the date of the decision as well as the relevant decision protocols [2].We identified 23 municipalities that have abolished LOV in homecare.In three of these municipalities, we failed to obtain decision protocols and other empirical materials whereupon they have been excluded [3].Accordingly, we investigate 20 municipalities [4].

Empirical materials
We wanted to capture the arguments behind the decision to abolish LOV-what were the stated reasons?Most of the municipalities made their decisions several years ago.The closer in time the source is to what it provides information about, the more credible the information is since people have a strong tendency to either forget or rationalize afterwards.Therefore, instead of conducting interviews long afterwards, we chose to analyse materials contemporary to the political decision.
First, we gathered the municipalities' decision protocols as described above.These contained a decision justification which could be brief or extensive.Some also had supplements, such as investigations.Second, we gathered media materials by using Retriever's digital news archive, which provides articles from print and digital editorial media as well as from radio and television.For each municipality, we conducted searches on the municipality's name in combination with "home-care services" (hemtj€ anst).Thereafter we selected the material that dealt with the abolition of LOV [5].Both news articles and debate articles are included.
Information about what has replaced LOV in home-care servicespublic monopoly or another procurement modelwas gathered from the municipalities' websites complemented with media material.

Analytical frameworkabolishing LOV
Based on previous research on re-municipalization, we constructed an analytical framework with political and pragmatic arguments and a draft of indicators of reasons for abolishing LOV.After reviewing the empirical materials, we made some adjustments to the indicators to make them clearer and include all arguments appearing in the materials.Thus, our analytical framework was designed through interplay between theory and materials (Table 1).The political reasons include ideological arguments against profits in public welfare and arguments to safeguard universalism.The pragmatic reasons are more diverse and include the weak position of private providers, lack of stability regarding the number of providers, private providers' elder abuse, private providers' fraud, private providers' inadequate routines and, finally, administrative burden to maintain the system of choice.

IJPSM 37,2
Analytical frameworkreplacing LOV There are two plausible outcomes in the analysis of what LOV had been replaced by (Table 2).The first outcome is homecare only by the public provider, i.e., public monopoly.The other outcome is procurement according to the LOU, i.e., continued privatization.In this case, either all home-care services in the municipality can be procuredor a specified share of it.In the latter case, the municipality keeps a part "in-house."

Findings
Arguments to abolish LOV in home-care services In the empirical materials, we searched for indicators of political and pragmatic reasons to abolish LOV in homecare.A summary of the investigation is presented in Table 3 which shows the arguments found in the material from each municipality as well as the municipalities' population.Since the material is different in length and degree of detail between the municipalities, some individual arguments may be missing in the presentation.Overall, however, we assess that the analysis gives a fair picture of municipalities that have abolished LOV.

Political reasons
One finding is that political arguments for abolishing LOV are few in the materials.Only one municipality explicitly mentions problems with profits in welfare, namely the small northern municipality of Robertsfors.The social-democratic chairman states that LOV is not only   being abolished for pragmatic reasons but also ideological reasons: "We don't want people to take money out of welfare and give it to private companies.We want people to reinvest back into schools and the elderly" (V€ asterbottens Folkblad, 2015).
In a few other municipalities, it is stated that in a system with LOV politicians have less opportunity to influence and control eldercare regarding working conditions for the staff or continuity for the users: "There should be as few staff as possible involved with a user.It should be uniform.But we cannot control that if we outsource the work" (Nya Kristinehamns-Posten, 2011).It is also stated in the municipality of Vindeln that it is better to have a municipal provider for all care activities, to ensure the quality of the activities (V€ asterbottens Folkblad, 2012).We choose to interpret these statements as arguments to safeguard universalism in eldercare, i.e., all citizens should have the same quality of care, regardless of provider.
This means that, in the materials, not much indicates that LOV is abolished for ideological reasons.On the other hand, we find ideological arguments from centre-right politicians who oppose the abolition of LOV.For example, a politician in the municipality of B astad says: "It's a fundamental ideological issue for us the Moderates, the right to choose, and to keep LOV costs nothing" (Helsingborgs Dagblad, 2016).In a reservation to the decision, the Liberals in the municipality of S€ odert€ alje state: "The Act on System of Choice puts the individual senior in the centre.We stand up for LOV" (S€ odert€ alje kommun, 2014).
Yet, in many municipalities, centre-right politicians support the abolition of LOV.In some municipalities they are even the driving force, as in Tj€ orn, an island municipality on the west coast: "It's never fun for a Moderate to say that we have decided to abolish LOV" (G€ oteborgs-Posten, 2018).Another example is the municipality of Str€ angn€ as: "We cannot automatically continue with a system that doesn't work, just because you stand behind a political view that promotes freedom of choice" (Dagens Samh€ alle, 2021).

Pragmatic reasons
Pragmatic reasons were frequent in the empirical materials.When we compiled our findings, we noted interesting differences between small and large municipalities (see population size in Table 3): different problems were described, and different arguments were put forward.Especially the larger municipalities Link€ oping, Sollentuna, and S€ odert€ alje deviate.However, common in all municipalities is that the costs of maintaining the system are seen as burdensome concerning administration and supervision.In addition, the municipalities see it as problematic that they cannot influence the number of private providers in LOV due to the principle of free establishment.
A common remark that we find in the material of smaller municipalities is that there have been private providers in the past, but too few citizens have chosen them.Therefore, these providers have closed their operations in the municipalities.In other small municipalities, one or a few private providers remain but with just a few users, and in a few more municipalities there has never been a private provider.
Accordingly, a general statement in the material is that citizens with granted home-care services have not been interested in choosing private providers, but also that private companies are not interested in applying.A typical comment is: "It [LOV] doesn't work in small municipalities" (Skaraborgsbygden, 2017).In some smaller municipalities, the last private provider has just filed for bankruptcy or has not extended its contract for quality reasons.Thus, it was seen as "the right opportunity" to abolish LOV.Another few municipalities have only exceptionally had private providers and chose to abolish LOV rather than spend a lot of administrative time updating the tender document in accordance to current procedures and organization.The materials show that the politicians consider that an extensive administration is required to maintain LOV, mainly to continuously update and revise the tender document and routines.In municipalities with approved private providers, the administration is also required to inform the citizens about selectable providers and how the choice should be made.
One example of a municipality with few users with a private provider is Finsp ang.The year prior to the decision to abolish LOV, twelve home-care recipients had chosen the sole private provider.This provider completed a total of 38 h out of the municipality's over 10,000 annual home-care hours (Norrk€ opings Tidningar, 2018).Yet an administrative resource equivalent to 0.25 annual workers was required to administer the system of choice (Finsp angs kommun, 2018).In the municipality of Osby, four citizens had chosen the only approved private provider and the abolition of LOV was estimated to save 250,000 Swedish krona (SEK) annually (Sveriges Radio, 2018).In a decision protocol of the municipality of Ving aker, the same thing was described: "It has been the case throughout the years that the municipality of Ving aker has had more private providers approved than there have been users who have chosen private providers" (Ving akers kommun, 2018).A social-democratic politician in the municipality of Vindeln says: "No one [of the older citizens] wants private home-care services, and instead we can transfer the resources that LOV requires to other activities" (V€ asterbottens-Kuriren, 2015).
In the municipality of B astad, the abolition of LOV is justified by the fact that the previous private providers had very few users and there were currently no private providers left in the municipality.The local party Bj€ arepartiet argues that by abolishing LOV, the burden on the administration is reduced: The introduction of LOV has taken a lot of administrative effort for a pressured administration with contracts of around 40 pages.The contracts must be followed up from a quality point of view and involve supervision that is often not done (Helsingborgs Dagblad, 2015).
In the municipality of Robertsfors, it is stated that the only private provider showed shortcomings during contract follow-ups, concerning, for example, time registration, routines and deviation reports.This along with the costs of maintaining LOV led the municipality to abolish LOV (Sveriges Radio, 2015).
The situation is different in the larger and often more centrally located municipalities.For example, in the municipality of S€ odert€ alje, a large municipality in the capital area with many different providers, arguments are put forward based on experiences of cancelled contracts, cheating and even police raids, which according to the Social Democrats demanded a large and expanding administrative apparatus for supervision: "With sixteen or more companies, based on the experiences we now have, at least ten more [administrative] positions are needed" (LT, 2014).A problem described is that in LOV, the municipality cannot restrict the number of private providers.The municipality of Sollentuna, also in the capital area, had similar problems with 15-20 different home-care providers.After suspending many contractors due to cheating and fraud, the municipality decided to abolish LOV (Mitt-i-Stockholm, 2022).Link€ oping, a large town, also describes a situation with providers' withdrawn permissions, terminated contracts and fraud: "Cheating is widespread, and the system requires large costs for the municipality" (Corren, 2020).
Several municipalities state problems with a lack of stability.For example, contracts may be terminated due to misconduct, or providers resign or go bankrupt.If so, the municipality must be prepared to assume responsibility for providing the granted care, but there are also problems with continuity for the users.For example, in the municipality of V€ armd€ o, it is stated that the continual withdrawal or bankruptcy of private providers has created problems for both the municipality and the citizens: ". . . it has meant a burden on the whole organization and as a result, many homecare recipients have had to change providers" (V€ armd€ o Kommun, 2020).

IJPSM 37,2
After LOVthen what?When a municipality decides to abolish LOV, there is already a strategy of action for how to provide home-care services in the future.Usually, this strategy is the result of an investigation conducted by the municipal administration.There are two plausible replacements: (1) Delivering homecare only by the public provider, i.e., a public monopoly.
(2) Procuring homecare according to the LOUeither all homecare in the municipality or a specified share of it, i.e., continued privatization.
In our investigation, the first outcome is the most common as presented in Table 4.
As shown above, a common argument to abolish LOV is that few citizens have chosen private providers or that there are not even any private providers, most likely because these municipalities are not lucrative enough for private companies.In all these cases, LOV has been replaced by a public monopoly meaning that when LOV is abolished, the municipality will reintroduce the system that was applied before LOV.Consequently, in the future, all home-care services will be carried out only by the public provider.
However, in another group of municipalities, the situation is different: There have been too many private providers and several reported cases of cheating and fraud.This group includes the large municipalities of Sollentuna, S€ odert€ alje and Link€ oping.In these cases, LOV is instead replaced by another procurement model -LOU.In LOU, a public procurement made by the municipality's administration determines which companies get the contracts.The evaluation of the companies is based on quality.In these cases, the economic compensation is fixed and regulated by the municipality, regardless of the provider.
In for example Sollentuna, the municipality has signed a contract with six home-care providers for three years.The municipally owned corporation Solom had to submit their tender in the procurement just like the private companies (Sollentuna kommun, 2022).In S€ odert€ alje, originally 40% of the total number of home-care hours were procured and shared between five private providers with fixed shares.The homecare is procured every four years and in the most recent round (2022) it was not possible to fill all the procured shares.As a result, the municipality now have a 67% share (LT, 2022)  Abolishing the Act on System of Choice S€ odert€ alje still have a choice since a provider is selectable until it reaches its capacity ceiling by the procurement.Link€ oping have chosen another model -LOU without user choice.More precisely, the municipality is divided into six geographical areas with one provider responsible for all users in each area: The public provider is responsible in three areas and private providers in the other three (Link€ oping kommun, 2023).V€ armd€ o, an archipelago municipality close to the capital, has introduced a similar model with four geographical areas (V€ armd€ o kommun, 2022).The municipality of Str€ angn€ as will implement LOU with geographical areas in 2025.This decision was explained as follows: "By assigning providers of homecare based on geographic area, better conditions are provided for planning, coordination, scheduling, and reduced administration" (Str€ angn€ as kommun, 2023).In these three municipalities, marketization remains with both private and public providers, but the citizen is assigned a provider by residential address and not by user choice.
In sum, we detect the following pattern: When LOV is abolished in smaller municipalities because of a weak position of private providers; it is replaced by public monopoly, i.e., an obvious case of re-municipalization.However, in larger municipalities, where the reason rather is misconduct or a high turnover by private providers (which also are too many) as well as high administrative costs, LOV is replaced by LOU.The latter implies a continued privatization with a mix of different providers, both private and public.An important difference is that the municipality has significantly more control in LOU since it is the municipality that decides upon the number of providers and their distribution, not the market.

Concluding discussion
The findings of the present study demonstrate that political arguments seem to have a very limited significance in abolishing LOV in Swedish home-care services, despite the partydominated municipalities.The most frequent reasons found in the materials were instead pragmatic.This result supports previous research on re-municipalization showing that ideology can indeed be an important driver in privatization but largely lacks significance in re-municipalization (e.g., Jansson et al., 2021;see Hung and Lu, 2022).It means that drivers of re-municipalization in this case of social services do not deviate from technical services as have been suggested by other scholars (e.g., Petersen et al., 2015;Gradus and Budding, 2020;Jordahl and Persson, 2021) it is a matter of pragmatism anyway.
Our empirical investigation included 20 Swedish municipalities and we discovered different patterns.In smaller municipalities, there were few private providers of homecare, with few users.The most frequent arguments for abolishing LOV were the weak position of private providers and the administrative burden.In smaller municipalities, LOV was replaced by a public monopoly.In the larger municipalities, there was a high turnover or even an experienced uncontrollable quantity of private providers and in some cases also reports of fraud and inadequate routines.The most frequent argument of larger municipalities was also the administrative burden, but in this case for control and review.In larger municipalities, LOV was replaced by another type of procurement -LOU.
The latter implies that a decision to abolish LOV does not necessarily imply a return to a public monopoly.When LOV is replaced by LOU, the model with different providers continues through another legal framework.It might have implications for the discussion on the concept of re-municipalization.Re-municipalization refers to a transfer from the municipality as a provider to the private sector, and then back to the municipality again (e.g., Voorn, 2021;Clifton et al., 2021).This is not a feasible description of the development in all the Swedish municipalities.Thus, decisions to abandon LOV cannot be said to represent a clearcut case of re-municipalization.

IJPSM 37,2
Our findings raise the question of whether the concept of re-municipalization is too narrow.In LOV, the municipality's control is limited.LOV, a choice system with freedom of establishment, hands over the power to the market to decide the number of providers and their share since it is decided by the citizens' choices and the private companies' interests to apply.Consequently, a transition from LOV to LOU implies a development from the market's control of the distribution between providers to the municipality's control of the distribution.In LOU, the municipality decides the exact number of providers and often also the market share of each provider.
The transition to LOU is not covered by the current definition of re-municipalization.Hence, the concept fails to quite capture important aspects of how marketization at the local level may change and develop over time and in different ways in different local contexts.A suggestion is broadening the concept to include more dimensions.In its current form, remunicipalization is restricted to the provision of public services.It could be developed to also cover the control of public services.Such a broadening of the concept with both provision and control dimensions would be valuable in exploring and seeking to understand the next chapter of marketization at the local level in different countries.
We emphasize two limitations.Firstly, due to feasibility, our investigation was limited to the abolition of LOV in home-care services, with neither nursing homes nor the abolition of LOU included.Thus, generalization to Swedish eldercare overall should be cautiously done.Secondly, our empirical materials consist of secondary sources in the form of texts.While there are several advantages to such contemporary sources, the materials vary in length and detail between municipalities which may result in individual arguments for specific municipalities being missing.Interviews or surveys would have allowed us to ask more precise questions systematically but may have been affected by the time gap.We encourage future studies to further investigate the adoption and abolition of both LOV and LOU in all areas of Swedish eldercare using different methodological approaches.2. In most municipalities, a political decision to abolish user choice according to LOV is first taken by the care committee, then by the executive committee and finally by the popularly elected municipal council.
4. All 23 municipalities are listed in the supplemental file, "Municipalities that have abolished LOV in home-care services." 5. The quotations in the empirical parts of this article are translated from Swedish to English by the authors.
Figure 1.Development of the act on system of choice (LOV) in Swedish municipalities Own work based on the empirical analysis; statistics on population from Statistics Sweden's StatisticalDatabase (2022)

Notes 1 .
"Home-care services" involve individually adapted interventions in the older citizen's own home regarding personal care and services.
. Citizens in Sollentuna and