On the diffusion and implementation of trust-based management in Scandinavia: cross-country survey evidence

Purpose – Thisstudyaimstoexaminethediffusionandimplementationoftrust-basedmanagement(TBM)in Scandinavian countries (Denmark, Norway and Sweden). TBM is a novel “ anti-New Public Management (NPM) ” innovation within the realm of New Public Governance (NPG), which asserts that leadership and control in public sector organizations should be practiced and designed based on the assumption that civil servants andemployeesin general are trustworthy. Theresearch questionsare as follows:How has TBMbeen diffused and implemented in Scandinavia? To what extent can the institutional logics framework increase understanding of similarities and differences between the Scandinavian countries (Denmark, Norway and Sweden)? Design/methodology/approach – The authors designed and submitted surveys to the municipal directors of the three Scandinavian municipal populations, thereby producing a unique cross-country dataset on TBM diffusion and implementation in Scandinavia (Denmark, Norway and Sweden). Findings – The authors ’ study shows that TBM has diffused widely among Scandinavian municipalities and has developed into a municipal-level concept across policy fields and sectors. While Denmark stands out as an earlier and more decisive TBM reformer, the results show that similarities in the diffusion and implementation of TBM in Scandinavian countries are more apparent than the differences. Originality/value – This study contributes to the public management literature and research on anti-NPM and NPG concepts by being the first wide-scale empirical study of TBM diffusion and implementation in the Scandinavian municipal sectors.


Introduction
The aim of this study is to examine the diffusion and implementation of trust-based management (TBM) in the Scandinavian countries.TBM is a management concept within the realm of new public governance (NPG) (Modell, 2021;Torfing and Bentzen, 2022).TBM asserts that leadership and control in public sector organizations should be practiced and designed based on the assumption that civil servants and employees in general are trustworthy.Excessive control, extensive performance measurement systems and transactional leadership styles should be replaced by decentralization, value-based management, involvement and looser forms of control (cf. SOU 2017(cf. SOU :56, 2017;;Bentzen, 2022).Hence, TBM involves sustaining and developing trust as an organizational resource to enable the distribution of autonomy and decision-making, which in turn allows for increased task complexity and innovation (Bentzen, 2022).Through its basic assumption of trustworthiness related to the institutional logic of the profession (Freidson, 2001), TBM marks a break with new public management (NPM), which is based on a market logic (Ouchi, 1980;Adler, 2001) and is claimed to hinder cooperation, coordination and collaboration in public sector organizations (Wiesel and Modell, 2014).
By examining TBM in Scandinavia we increase knowledge about the diffusion and implementation of a novel "anti-NPM" management innovation related to NPG.We are interested in the extent to which the innovation has reached the members of a social system (Rogers, 1995) or, more precisely in the case of TBM, to what extent dominant organizational actors have adopted TBM after encountering the global idea.Adoption refers to the formal or informal decision to commence implementation of a management concept (Siverbo, 2014).Our interest in implementation is focused on the purposes, measures and challenges associated with the new management concept (cf.Johnsen and Vakkuri, 2006;Siverbo, 2014).
Despite the longstanding criticism of NPM, few tangible alternatives to NPM-oriented public management systems have emerged.Many scholars argue that NPM, with its elements of corporatization, competition and performance management (Hood, 1995), is still the dominant administrative paradigm in most Western countries (Steccolini, 2019;Funck and Karlsson, 2020).Although it has been highly criticized, it has so far resisted replacement (Wiesel and Modell, 2014;Modell, 2021).TBM in Scandinavia is a rare example of an "anti-NPM" movement that has materialized into an apparent alternative, adapted to a local context.Through examining the diffusion and implementation of TBM, we add insights into the extent to which an NPG-oriented management concept has managed to challenge NPM.
In addition, inspired by an institutional logics framework (Friedland and Alford, 1991;Thornton et al., 2012), we examine and analyze similarities and differences in diffusion and implementation between the Scandinavian countries (Denmark, Norway and Sweden).Institutional logics can be defined as "a set of material practices and symbolic constructions that constitute organizing principles and which is available to organizations and individuals to elaborate" (Friedland and Alford, 1991, p. 248).Institutional logics exist at different levels (society, organizational fields, organizations, etc.), and as they specify what is rational and appropriate, they put normative pressure on organizations to comply with their formal and informal prescriptions.The institutional logics perspective allows for predicting both similarities and differences in the diffusion and implementation of TBM in Scandinavia.Diffusion and implementation may be similar because Scandinavian countries have the same society-level logics concerning trust and the notion of the welfare state (Johnsen and Vakkuri, 2006;Henriksson, 2012;Andreasson, 2017;Holmberg and Rothstein, 2020), which speaks in favor of the hypothesis that TBM has Scandinavia as the medium for diffusion and that there are no reasons to expect any major differences between the three different countries considering the strong fit between TBM and society-level logics.
However, the logics perspective is also open to TBM differences because of countryspecific institutional logics, possibly with historical roots.For instance, previous research shows that the Scandinavian countries are different in approaching reforms, mostly evidenced by how they implemented the predecessor movement of NPM (see, e.g.Christensen and Yesilkagit, 2006;Hansen, 2010;Green-Pedersen, 2002;Greve et al., 2016;Hansen, 2005;Knutsson et al., 2017;Kure and Malmmose, 2017;Torfing and Bentzen, 2022) and as will be further detailed below, certain differences appeared, notably concerning how far NPM was embodied in formal institutions in the three countries.Furthermore, there are structural differences between the three local government systems, and TBM reforms have somewhat different background contexts.In total this may have created different "legacies" and sedimented institutional arrangements (Cooper et al., 1996;Raynard et al., 2013) which affect the diffusion and implementation of management innovations in the Scandinavian countries.
To gain knowledge of the diffusion and implementation of TBM in Scandinavia and the conditions for (anti-NPM) NPG alternatives to obtain a foothold in similar societies with different approaches to the existing NPM-oriented administrative paradigm, we set out to answer the following research questions: RQ1.How has TBM been diffused and implemented in Scandinavia?

RQ2.
To what extent can the institutional logics framework increase understanding of similarities and differences between the Scandinavian countries (Denmark, Norway and Sweden)?
Our study shows that TBM has diffused widely among Scandinavian municipalities and has developed into a municipal-level concept across policy fields and sectors.While Denmark stands out as a somewhat earlier and more decisive TBM reformer, the overall finding is that the similarities in adoption and implementation are more apparent than the differences.The high degree of similarity is understood as a consequence of strong common institutional logics at the Scandinavian societal level.
In the next two sections, we present the TBM reform movements in Denmark, Norway and Sweden and use the institutional logics framework to discuss potential sources of similarities or differences in the diffusion and implementation of TBM.Next, the methods and results are presented.In the penultimate section, we discuss our findings and their implications for future research, as well as for public sector practitioners.In the final section, we present our conclusions.

Reform movements and literature review
The TBM reform in Denmark started in the wake of a municipal merger reform, creating units with a larger strategic capacity for administrative experiments and setting reform agendas (Torfing and Bentzen, 2022).Following a debate about the need for a more trust-oriented form of public organization, not least to curb the negative effects of NPM-oriented governance, a "trust reform" was first formally mentioned in 2011 in the foundation paper of the newly elected government (the Social Democratic Party, the Danish Social Liberty Party and the Socialistic Peoples Party).However, it would take almost two years before the trust reform was launched.The principles were negotiated between the government, representatives of municipalities and regions and unions, emphasizing that trust should be a point of departure in management and governance (Vallentin and Thygesen, 2017;Torfing and Bentzen, 2022).Since then, ideas of TBM have continuously played a central role in Danish national governance reforms launched by shifting governments.
In Norway, the idea of a trust reform existed as early as 2010 and was debated, especially in the health care and school sectors.Several local initiatives were adopted, but as a nationwide initiative, TBM was launched later than in Denmark and Sweden, in 2021.The development of TBM in Norway was linked to perceived problems with NPM and gradually became a general idea presumed to apply to all sectors.In the national election in 2017, the Labor Party had a "trust reform" as part of its political program.When they finally won the election in 2021 together with the Agrarian Party, the adoption of a nationwide trust reform was part of the two parties' joint political platform, the so-called Hurdalsplattformen.
The Swedish development of TBM was, as in Denmark and Norway, highly linked to alleged problems with NPM-oriented control, first communicated to Swedish citizens in a series of newspaper articles in 2013.The idea of TBM was initiated by the Swedish Social Democratic government in the 2014 election campaign.Thereafter, in 2016, the minister of internal affairs appointed a government committee consisting of experienced public officials as well as academics from the Swedish universities to develop and promote trust-based control (SOU 2017(SOU :56, 2017)).The committee work resulted in six official government reports and several research reports and was finalized in 2020.
Previous research on TBM indicates that the concept has started to diffuse, but so far, no clear picture of diffusion in the municipality sectors exists and possible country differences are unknown.In 2017, the Swedish Agency for Public Management studied the diffusion of TBM in central government agencies and noted that 56% of the agencies had initiated at least one project to develop TBM (Statskontoret, 2018).A study of the Norwegian public sector in 2020, based on convenience sampling, indicated that 38% of the Norwegian municipalities and 20% of central government agencies had adopted TBM (Johnsen et al., 2022).Furthermore, in a web-based survey addressing financial directors in 2021, an adoption rate of 59% in Swedish municipalities was observed (Siverbo, 2022).Although extant research indicates that TBM has diffused in Norway and Sweden, there is a lack of knowledge concerning adoption in Denmark and reliable and comparable data about the development in the Scandinavian countries.
It also seems that the TBM reform stands out in comparison with previous reforms in the ambition of being all-embracing (e.g.SOU 2017SOU :56, 2017;;Johnsen et al., 2022).TBM is adopted for various types of activities, vertical (hierarchical) relationships from the political level to the operational level and horizontal (cooperative) relationships, including external relationships with customers, citizens and other external parties.However, so far this more specified diffusion has not been the subject of scholarly attention.
Based on the current literature, providing a clear description of the practices that constitute TBM is challenging.TBM is usually not described in terms of tangible practices but as an overall approach (see, e.g.SOU 2019:43) or a philosophy (Bringselius, 2021;Johnsen et al., 2022).Accordingly, organizations adopting TBM have noted a lack of clarity about how to work with TBM post-adoption.There are comments about how "we are laying rails while the train is going" and "the boat is being built while we row" (Arnegaard and Halvorsen-Lowe, 2020).However, the study by Johnsen et al. (2022) indicates that the actual practice of TBM seldom involves changes in formal control, possibly due to difficulties in changing control structures and the anticipated ramifications that downplaying formalization may have for the reliability of public bureaucracies.For instance, a common view is that TBM functions well as a complement to, rather than a substitute for, formal control in the form of management-by-objectives (MBOs).A study of central government in Norway indicates that a trust approach survives under MBO regimes, even if some managers perceive MBO as rather control-oriented (Christensen et al., 2018).TBM is also seen as a method to realize the original intentions of MBO, as MBO rests on the notion that civil servants should be trusted to accomplish results (Bjurstrøm, 2021;Johnsen et al., 2022).Overall, previous studies indicate that TBM is not so much about reduced formalization as about new leadership styles, improved communication, increased professional discretion and better interaction between employers, employees and unions (Bentzen, 2016(Bentzen, , 2019a, b), b).However, in some cases, firstlevel managers became more present in the daily operational work to compensate for reduction of formal control of employees (Vallentin and Thygesen, 2017).Johnsen et al. (2022) call for more systematic research on how TBM plays out in practice.

IJPSM
The topic most often treated in the literature on TBM is the relationships within administrations, especially between first-level managers and employees.These studies build mostly on case studies of soft activities (e.g.elderly care, social care) in municipalities, but there are also examples from health care (Pettersen, 2011) and central agencies (Hasche et al., 2021).To some extent, this research illustrates potentials and benefits of TBM, but gives comparatively more insights into challenges and apprehensions (H akansson, 2022).Reported benefits from TBM, in line with intended purposes, are that employees experience more freedom in their work and are allowed to work as professionals and make their own judgments, but this occurs only when they have a good relationship with their manager (Vallentin and Thygesen, 2017).It is also observed that employees that are less controlled and possess a certain amount of discretion, have better opportunities to be caretaker oriented (Elmersj€ o and Sundin, 2021).
Reported challenges when implementing and using TBM relate primarily to the operational level and involve both first-level managers and employees (Bentzen, 2016(Bentzen, , 2019a(Bentzen, , b, 2021;;Vallentin and Thygesen, 2017;Torfing and Bentzen, 2020).Studies show that implementation of TBM has suffered from poor anchoring processes and consequentially unprepared managers and workers (see also Alexius and Sardiello, 2018;Hasche et al., 2021).Different opinions about the meaning of trust and how TBM is to be handled have evolved.First-level managers have received incomplete training in finding a balance between trust and control and, more specifically, in finding the right level of discretion for employees.If the room for discretion is too large, employees may feel worried and doubt that they have the necessary competence to make expected decisions.If the room for discretion is too limited, workers become disappointed and feel the change is meaningless, almost insulting (Bentzen, 2019a).Furthermore, as TBM increases the risk both for managers and employees, some of them are reluctant to accept TBM, especially if they fear cutbacks are coming, which means they will be left with greater responsibilities but insufficient resources (see also Astvik et al., 2020).
Our summary of the reform background and previous studies indicates that TBM adds complementary aspects to the current formal control structures rather than reducing them.In the adopters' view, TBM seems to be about changing the social "software" surrounding formal control systems, such as social controls, leadership and culture, so that it better promotes trustful relations.Related to the wish to increase the space for professional discretion of public service employees, ambitions to reduce overly micro-managed operations and processes and to facilitate coordination, seem also to be a general feature of TBM in Scandinavia.In addition, case-study research has provided knowledge on individual municipalities' purposes and challenges related to the implementation of TBM.

Analytical framework: trust and institutional logics Trust and control
Trust is commonly defined as readiness to accept vulnerability (Rousseau et al., 1998) and take a leap of faith in people (M€ ollering, 2006).In interpersonal relations, this readiness typically increases if trustees are considered to be able, benevolent and have integrity (Mayer et al., 1995).In the context of public sector organizations, trust potentially exists, or does not exist, in many different relationships within the organizations and between organizations and different external parties (Bouckaert, 2012).A common view is that trust in public sector employees increases with their level of professionalization (Freidson, 2001).
The core idea of TBM is to use existing trust in all types of relationships to increase autonomy and reduce costly, time-consuming and sometimes annoying formalizations.Having staff with competence, goodwill and integrity creates an environment where formal controls can be minimized.In addition, in relationships where trust is missing and weak -possibly a dysfunction of NPM-oriented and bureaucratic formal control practices -TBM is instead mobilized to induce trust, often through leadership interventions (Bentzen, 2019a, b, Bentzen, 2022).Thus, as a new management concept, TBM can be seen as a reaction to the market-based notion of coordination and control implied in NPM (Hood, 1995).TBM challenges the instrumental and moral notion that agents, and the public sector as such, perform best if incentivized and controlled.It embodies instead an idea that the public sector should rely more on trust and less on formal control.In doing so, the TBM movement aligns with the literature that sees trust and formal control as substituting mechanisms.A different but common view, in particular in the organizational control literature, is that formal control and trust complement each other (e.g.Dekker, 2004;Long, 2018;Bentzen et al., 2023).As indicated in the literature review and as will be further elaborated below, TBM in practice in Scandinavia serves more as a complement to formal control rather than a substitute.

Similarities and differences in Scandinavian institutional logics
The institutional logics perspective is increasingly becoming one of the most influential viewpoints within the realm of institutional theory (Modell, 2022).Friedland and Alford (1991) explained in their seminal work how institutional logics exist at different levels and guide the behavior and decision-making of individuals and organizations.These logics not only coexist but also compete and clash, sometimes forcing individuals to navigate institutional contradictions.In an extension of the institutional logics perspective, Thornton et al. (2012) emphasized and explained how institutional logics originate from different societal sectors and historical contexts.Of specific interest to our study is their insight into how organizations that share the same institutional logics at the overall societal level and consequently experience similar guidance and pressure, may still make different decisions about adoption and implementation of management concepts due to different historical developments at the individual country level.
The diffusion and implementation of TBM in Scandinavia may hence exhibit fundamental similarities because of shared institutional logics at the societal level (Friedland and Alford, 1991;Thornton et al., 2012), constituting a fertile ground for TBM-oriented practices and organizing principles.First, the three countries have similar ways of organizing society, managing institutions and conducting politics (Johnsen and Vakkuri, 2006).The Scandinavian model comprises high taxes, redistributive welfare services and extensive collaboration between unions, trade organizations and the state.The countries are quite similar in their responsibilities for providing services within education (preschool, primary and lower secondary schools), health and social care (primary health services, elderly care, home care, social welfare, etc.), utilities (water supply, sewerage, and waste) and culture (Rose and St ahlberg, 2005).Most of these activities are responsibilities for local government organizations (municipalities and regions) which, compared to many other countries, have extensive autonomy.They are directed by councils, constituting the highest municipal body, democratically elected every fourth year.
Second, no other region in the world reaches Scandinavia's level of social and institutional trust, that is, the high levels of trust between citizens and between citizens and societal institutions such as public sector organizations (Henriksson, 2012;Andreasson, 2017;Holmberg and Rothstein, 2020).This Scandinavian exceptionalism has developed over the years due to several interacting societal processes.Commonly emphasized factors are the role of the ambitious welfare state, the quality of government, the function of voluntary associations and the traditionally homogenous populations.The state is transparent, with relatively low levels of corruption and has, generally speaking, managed to create welfare and equality (Berg and Johansson, 2020).Against this background, it is not surprising that the first developments of TBM are found in Scandinavia (Johnsen et al., 2022;, cf. Bouckaert, 2012), although management tools based on similar ideas also exist elsewhere (see, e.g.Argento and Peda, 2015;Fledderus, 2015;Hood et al., 2021;Lahat and Sabah, 2021;Chiwawa, 2022).
However, institutional logics on the level of individual countries and other structural differences allow for differences in TBM diffusion and implementation.The municipal sectorsthe units of interest in this paperdiffer substantially in average size of the municipalities.The 98 municipalities in Denmark each have approximately 50,000 citizens, whereas the 290 Swedish municipalities have an average of 36,000 citizens while the 356 Norwegian municipalities average 15,000 citizens.The countries also differ in their approaches to NPM implementation.Even if all three countries have been characterized as "active reformers", they do not necessarily approach reforms in identical ways (Greve et al., 2016).According to Pollitt and Bouckaert (2017), descriptions of Scandinavian public management traditions at both central and local government levels need to be more specific rather than simply lumping the three countries together in a "Scandinavian" category.Denmark, Norway and Sweden exhibit differences in governmental traditions (Rhodes, 1999) and, more specifically, in adoption and implementation of NPM, as well as post-NPM strategies (Hansen, 2010).
In comparison to Norway and Sweden, Denmark stands out as a somewhat selective NPM reformer.Danish NPM reforms tend to be less formalized and less focused on evaluation (Greve et al., 2016;Hansen, 2005), possibly mirroring a more practical stance to the ideal of a clear-cut separation of politics and administration (Christensen and Yesilkagit, 2006).Denmark adopted NPM earlier, but the adoption was selective and limited to only some of the NPM elements (Green-Pedersen, 2002;Kure and Malmmose, 2017;Torfing and Bentzen, 2022).In addition, reform implementation in Denmark was comparatively slow (Kure and Malmmose, 2017;Torfing and Bentzen, 2022), tentatively explained by the consensus-driven policy processes in Denmark (OECD, 2000).
Norway distinguishes itself by being a more reluctant and pragmatic reformer, but also by its wealth, which relaxed pressure on Norway to become a high adopter of NPM as they lacked a strong motive for NPM reforms (Hood, 1995).Wealth in combination with a relatively well-functioning public sector reduced the need for radical reform (Knutsson et al., 2017).This may be a key explanation for the general view that Norway was a "reluctant reformer" (Christensen, 2003), lagging behind typical NPM countries (Christensen et al., 2008;Bezes et al., 2013).However, reluctance was also explained by skepticism toward NPM because it induced individualism in a collectively oriented political and administrative culture (Hansen, 2010).This may explain why Norway was first in Scandinavia to roll back NPM initiatives (Hansen, 2010).In addition, over the years a pragmatic approach to managerial tools appears to have evolved in Norway and in later years the municipal sector, especially, has shown a willingness to experiment with internal structures (Knutsson et al., 2017).
In comparison, Sweden appears to be a more thorough NPM reformer, with a higher degree of decentralization, a broader and deeper implementation of NPM reforms and a later debut of post-NPM reforms (Hansen, 2010;Knutsson et al., 2017).Sweden was among the group of countries in the 1980s considered to be the highest adopters of NPM (Hood, 1995).Denmark was an earlier adopter of some elements (see above), but like Norway one step behind Sweden in broadness and intensity of implementation (Hood, 1995;Christensen and Laegreid, 1999;Green-Pedersen, 2002;Knutsson et al., 2017).Possibly because of the larger investment in NPM reforms, Sweden has been later than Denmark and especially Norway in implementing post-NPM reforms.Denmark was more prone to implement such reforms as complements to NPM while Norway also took measures to roll back NPM reforms (Hansen, 2010).
In summary, we note that TBM adoption and implementation in Scandinavia takes place in somewhat different institutional contexts even if much is similar in the three countries (see Table 1).Of certain interest is that TBM in the three countries is introduced due to dissatisfaction with NPM, even though their actual approaches to NPM appear to differ.As

Trust-based management in Scandinavia
such, NPM has left them with somewhat different institutional "legacies" (Raynard et al., 2013) which can be expected to affect the diffusion of the NPG-oriented TBM alternative.
Such legacy effects have the potential to both incentivize a faster and broader adoption if there is a relation between the degree of NPM orientation and the urge for an alternative, or hinder adoption because NPM has become more strongly embodied in formal and informal institutions.In that sense, the potential effect of the partly different institutional settings is undetermined.To learn more about the driver behind the TBM diffusion and whether the TBM reform movement has managed to abstract TBM to a level that fits the Scandinavian high-trust culture or if its diffusion and implementation is constrained by the type and degree of NPM orientation, we explore the diffusion and implementation of TBM across the three Scandinavian countries.

Method
For practical reasons, the examination of the diffusion and implementation of TBM in Scandinavia was delimited to the municipal sectors of the three countries.Within this population we used a census (total sample) study, sending a survey questionnaire to all municipal directors.Municipal directors are the highest officials in municipalities, responsible for implementing municipal policy and for the overall management of their organizations.Their responsibility for and overview of, management issues make them appropriate respondents as TBM is within the overall realm of management.In late 2021, we sent out a web-based questionnaire to 744 directors (Denmark, 98; Norway, 356; Sweden, 290), and after three reminders, we received 385 replies, which constitute an overall response rate of 52%.The response rate in Denmark was 63%, in Norway 49% and in Sweden 52%.The relatively high response rate of the census study decreases the risk of respondent bias, always inherent in survey studies.Still, we expect that respondents with an interest in TBM were more likely to respond to the survey.Scores are therefore possibly somewhat skewed regarding the extent of TBM adoption.
The questionnaire was developed and designed in close cooperation by the authors of this paper, which means that at least one scholarly representative from each country participated.

IJPSM
The ambition was to map diffusion of the concept and to explore implementation issues in the form of purposes, measures and challenges related to TBM, while keeping the questionnaire short to ensure a good response rate.We developed new questions for all areas of interest, drawing on what has been evidenced as potentially interesting factors and aspects in previous case-based research on TBM but also on our own experiences from other research projects and in close cooperation with the municipal sector.
Questions related to diffusion were intended to capture adoption at the levels of organization, policy sector, type of relationship and the degree of formalization in adoption decisions.Questions about the purpose of TBM were selected based on developers' (e.g. the Swedish trust committee) stated aims with TBM and previous research on the subject.Questions about measures were guided by Bringselius' (2021) notion of TBM being realized through changes in organization, leadership, control and culture, as further evidenced by case studies reported above.Questions about implementation and realization challenges were inspired by previous case-study research (e.g.Bentzen, 2019aBentzen, , 2019b;;Vallentin and Thygesen, 2017).
The survey was first developed in the Norwegian language, which was possible to do cooperatively in the research group since the Scandinavian languages (Danish, Norwegian and Swedish) are quite similar.At this stage, the focus was on survey content and face validity, that is, that all questions were natural and understandable for respondents from all countries.In a second step, the survey was translated into Danish and Swedish.This was also done in close cooperation as the exact meaning of some words needed to be discussed.In a third step, the questionnaire was tested on three high-level managers, who, by and large, approved the formulations but suggested some minor changes.These suggestions were discussed in a final meeting in the research group and implemented in the final version of the survey.
Data were analyzed jointly for Scandinavia (unweighted average) and individually for Denmark, Norway and Sweden.In the findings section, data are mainly presented as frequencies of respondents agreeing that a certain statement was accurate, for instance "to a very high extent" or "to a rather high extent", where other alternatives were "to some extent", "to a small extent", "not at all" and "don't know".In presentations, "don't know" answers are treated as missing data.We assess the practical significance and relevance of country differences with Cohen's D, which is a measure of effect size.We highlight medium (Cohen's D > 0.5) and large (Cohen's D > 0.8) size effects.Since the sample is a census, although with non-responses, we do not assess statistical significance.

Diffusion and formalization
Based on the directors' answers it is reasonable to conclude that the concept of TBM has diffused widely in Scandinavia and that a clear majority of the municipalities have worked with the concept for some time (see Table 2).The share of municipalities that have abolished the concept or have lost momentum in implementation is low.Formal adoption is more common than informal adoption, and in one-fourth of the municipalities, both the political and administrative levels have made formal decisions to implement TBM.
As is also evident in Table 2, Denmark stands out as the Scandinavian country with the highest adoption rate and a conspicuously high degree of formal adoption, especially administrative formal adoption.

Targeted activities and relationships
The adoption rate in different policy areas indicates whether TBM is considered a generally applicable concept or is considered fit only for certain activities (see Table 3).The findings show that no policy area stands out, although there are minor differences.Furthermore, TBM

Trust-based management in Scandinavia
in Scandinavia is intended for a broad set of relationships.It is most frequently targeted for vertical relationships, especially at the operational level, but also for horizontal and nonhierarchical relations.In many municipalities, it is also applied in contact with external constituents, but not so much with the central government and voluntary sector.
As shown in Table 3, Danish municipalities are less prone to target unions for TBM.In addition, although the differences are not large, a higher share of Danish municipalities adopt TBM for a broader range of activities.

Purposes and measures
The municipal directors agreed that there are several reasons for adopting and implementing TBM.Broadly, the purpose is to increase employees' room for discretion, as well as their motivation (see Table 4).The lower part of the table shows that different measures to implement TBM have been taken, some of which are directly related to the implementation of TBM and some are more general, although closely related to the meaning of TBM.The most popular measures are development of trust-oriented organization cultures, leadership styles and belief systems (strategies/visions/basic values).Interestingly, despite declared ambitions to reduce overly extensive NPM control, few municipalities report reduced formal control or involve employees in control system design.
Denmark differs greatly from Norway and Sweden in that it does not link TBM to the purpose of improving strategic control.Denmark shows the largest interest in developing belief systems, particularly in comparison to Sweden and the lowest interest in developing a supporting organizational culture, especially compared to Norway.Norwegian municipalities diverge substantially from Danish and Swedish municipalities by being far more ready to admit that TBM adoption improves the image of the municipality.

Challenges
In general, the directors were not inclined to report implementation challenges (see Table 5).Only two challenges are reported to occur to a high extent in more than a third of the municipalities: the challenge of maintaining momentum when other issues call for attention and the challenge of remaining faithful to the idea of trust when things go wrong and politicians want to handle the situation with regulations.There are also challenges related to the concept of trust itself, what it is and how it should be operationalized, especially in Sweden and Norway.
Denmark differs from Norway and Sweden in that a lower share of directors report challenges overall, indicating that implementation at this point in time is perceived as easier in Denmark than in the other two countries.This is especially true when balancing control and trust.
In an additional analysis, we investigated whether challenges were more common among earlier or later adopters.However, this was not the case here.We also investigated whether

Trust-based management in Scandinavia
the eight municipalities where the implementation of TBM started to wane differed from the rest.Indeed, these municipalities experienced challenges on average twice as often as the others, but apart from that, they did not diverge substantially from other municipalities.

Concluding discussion
In this research, we set out to examine diffusion and implementation of TBM in Scandinavia and the extent to which the institutional logics framework can increase understanding of similarities and differences between the Scandinavian countries.
The first contribution of our research is new knowledge on the diffusion of a novel NPGoriented management innovation in Scandinavia.With new and unique data, we show that TBM has developed into a municipal-level concept not tied to any specific policy field or sector and has diffused widely among Scandinavian municipalities, which seems to be in line

IJPSM
with the intentions of the developers of the concept (SOU 2017:56;Johnsen et al., 2022).Our research is in line with prior studies that indicate wide diffusion of TBM in Sweden (Statskontoret, 2018;Siverbo, 2022) and Norway (Johnsen et al., 2022), but extends these studies by its inclusion and comparison of all three Scandinavian countries.This finding is interesting because management concepts connected to the NPG administrative paradigm have been difficult to spread widely, even though they have circulated globally since the early 2000s (Hood and Dixon, 2015;Steccolini, 2019;Modell, 2021).The fact that the NPG-oriented conceptaimed at restoring trust within public sector organizationswas introduced in Scandinavian high-trust societies most likely contributed to its popularity and diffusion.
Our second contribution is insights into TBM implementation, which complements previous mainly case study-based research.Our study supports the notion in previous studies that TBM implementation in reality is less about trust replacing control and more about coexistence of trust and formal control.Main TBM measures involve changing the social software (social controls, leadership, culture) surrounding formal control systems rather than making leeway for professional discretion and judgment by reducing formalization and bureaucracy (cf.Bentzen, 2016;Johnsen et al., 2022).Furthermore, our study confirms that implementation challenges noted in case studies do exist in the Scandinavian municipality population, such as confusion and disagreement of the meaning and realization of the concept and difficulties in persistence (Bentzen, 2016(Bentzen, , 2019a(Bentzen, , b, 2021;;Vallentin and Thygesen, 2017;Torfing and Bentzen, 2020).However, overall TBM seems to face relatively few implementation challenges, as indicated by the fact that no single challenge is significant for more than a third of the population.An important caveat to consider, though, is that the responding municipal directors might lack full insight into implementation issues and some municipalities may not have progressed far in the implementation process, thus not encountering obstacles yet.

Trust-based management in Scandinavia
The third contribution of this research is the understanding of whether TBM is a movement primarily mirroring a similarity in societal institutional logics regarding trust and welfare organization in Scandinavian countries, or whether country differences in structure, reform backgrounds, and NPM legacies create noteworthy differences in diffusion and implementation.Overall, our study indicates that the similarities between countries in TBM diffusion and implementation are more apparent than the differences.In general, the countries' municipalities adopt TBM for the same activities/sectors and the same relationships.Also, they predominantly implement it for the same reasons and with the same practices and face similar implementation challenges.The understanding of this is that the TBM movement has found an advantageous environment in the Scandinavian societal institutional logics (Friedland and Alford, 1991;Thornton et al., 2012).Common material practices and symbolic constructions related to societal trust and the notion of the welfare state have allowed TBM to reach a supracountry or supra-organizational field level (Friedland and Alford, 1991).While it is too early to say anything clear about the extent to which TBM will materialize into typical management practices in local governments, challenging the present NPM-oriented ones (cf.Modell, 2021), TBM has gained momentum as a Scandinavian reform movement that has managed to overcome the institutional differences and the different NPM legacies between countries.This indicates the importance of high levels of trust in the institutional environment for the diffusion of TBM, which may be important knowledge for other countries interested in joining the Scandinavian TBM movement.
However, although country-spanning similarities are best understood as consequences of supra-country societal institutional logics, some differences related to country-specific institutions and distinguishing features are worth emphasizing.Starting with Denmark, which stands out most in the comparison, the differences are likely connected to the time factor and the larger size and corresponding administrative competence of the Danish municipalities (cf.Johansson and Siverbo, 2009).Size and administrative capacity make a higher degree of formal adoption of TBM understandable, as well as a larger reliance on formal documents, when realizing TBM.Conversely, it explains the lower inclination to rely on softer controls, such as influencing organizational culture.Interestingly, the relatively greater inclination to formalize TBM in Denmark marks a break from the previous administrative tradition in Denmark of less formalization (Greve et al., 2016;Hansen, 2005).It is noteworthy that the larger Danish municipalities may have been more motivated to adopt TBM to counteract the risk of growth of formal control and bureaucracy in large units and at the same time are more prone to implement TBM with formal decisions and measures.Furthermore, the time factor makes both the wider diffusion and the confirming result of a longer tenure of TBM in Danish municipalities logical.It helps explain the less reported problems in Danish municipalities with finding a balance between control and trust, which may be an issue of experience.Another difference seems to be related to decisions made at the early stages of TBM development in Denmark.At the beginning, it was clearer in Denmark that the prime objective of the reform was allowing "public employees to focus more on their core tasks and less on documentation and paperwork" (Vallentin and Thygesen, 2017).This may explain the lower ambitions to use TBM to give top managers more time for strategic control.
The distinguishing features of Danish TBM adoption and implementation mean Denmark, known as an early but hesitant and selective NPM reformer, displays more decisiveness as an NPG reformer, at least at the municipal level.However, as TBM is only one expression of the NPG construct, it may be that the Danish municipalities have been selective in picking TBM and hesitant about other expressions of NPG.A further observation is that Denmark increasingly distinguishes itself as a general early adopter of reforms, as the country picked up both NPM and TBM before their Scandinavian neighbors and has managed to realize a municipal merger reform, which has only been discussed in Norway and Sweden in recent decades.

IJPSM
By contrast, the Norwegian municipal sector displays one striking distinguishing feature in its frank notion of adopting TBM for image management.Since resource dependence, in general, is larger in smaller units, one explanation may be the comparatively small size of Norwegian municipalities.This legitimacy ambition is interesting in relation to Norway's reputation as a reluctant and pragmatic reformer.While the adoption and implementation of TBM in the Norwegian municipal sector does not live up to the epithet of reluctance, image management means that it certainly lives up to expectations of pragmatism.
Sweden's reputation of being the most thorough NPM reformer could create expectations of the country also being a thorough TBM reformer, both because of proven change capabilities and perhaps also because of experiencing most downsides of NPM.Conversely, Sweden could be expected to be the least eager TBM adopter because of the large investments in NPM.Neither of these expectations appears to be valid, as Sweden does not stand out in any way in the comparisons.If anything, Sweden distinguishes itself by its high ambitions to investigate and develop the TBM concept in a public committee.Future research may examine if this procedure gives the public sector of Sweden a more solid ground for TBM implementation and justifies the epithet of being a thorough TBM reformer.
This study aimed to examine the diffusion and implementation of TBM in three Scandinavian countries.Based on a large-scale cross-national survey, we show that a majority of the municipalities in Denmark, Norway and Sweden have adopted the concept of TBM, with some variation in extent and content and, so far, moderate implementation challenges.Future research could examine more closely how differences in adoption and implementation impacts use and outcomes of TBM, preferably with a qualitative research strategy.
Due to a few methodological limitations our research should be interpreted with some caution.Although the municipal sectors constitute primary parts of the public sectors in Denmark, Norway and Sweden, this study does not provide insights regarding the diffusion and implementation of TBM in regional or state organizations.Also, the results must be interpreted with the organizational level of the respondents in mind.While municipal directors have a unique position for estimating the extent and form of TBM ideas, their top-down perspective is likely to differ from that of lower-level managers and employees, whose perspectives are not included in this study.Furthermore, although our process of translating survey questions was careful, it cannot be ruled out that some country differences were a consequence of the meaning of items having been slightly altered in the translation process.Finally, while we surveyed all municipalities in the Scandinavian countries, non-response bias might impose constraints on the true distribution of diffusion in the population of municipalities.

Table 2 .
*Cohen's D > 0.5 and **Cohen's D > 0.8 a For example trust-based control, trust-based leadership, trust reform or trust in the cooperation with citizens; b We processed answers to questions about if TBM adoption was a political and/or administrative decision or not formalized Source(s): Authors' own creation Adoption of TBM

Table 5 .
Challenges to TBM implementation